diff options
Diffstat (limited to 'drivers/net/ethernet')
196 files changed, 25415 insertions, 4353 deletions
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig index eadcb053807e..9a8308553520 100644 --- a/drivers/net/ethernet/Kconfig +++ b/drivers/net/ethernet/Kconfig @@ -34,6 +34,7 @@ source "drivers/net/ethernet/adi/Kconfig" source "drivers/net/ethernet/broadcom/Kconfig" source "drivers/net/ethernet/brocade/Kconfig" source "drivers/net/ethernet/calxeda/Kconfig" +source "drivers/net/ethernet/cavium/Kconfig" source "drivers/net/ethernet/chelsio/Kconfig" source "drivers/net/ethernet/cirrus/Kconfig" source "drivers/net/ethernet/cisco/Kconfig" diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile index 1367afcd0a8b..4395d99115a0 100644 --- a/drivers/net/ethernet/Makefile +++ b/drivers/net/ethernet/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_NET_BFIN) += adi/ obj-$(CONFIG_NET_VENDOR_BROADCOM) += broadcom/ obj-$(CONFIG_NET_VENDOR_BROCADE) += brocade/ obj-$(CONFIG_NET_CALXEDA_XGMAC) += calxeda/ +obj-$(CONFIG_NET_VENDOR_CAVIUM) += cavium/ obj-$(CONFIG_NET_VENDOR_CHELSIO) += chelsio/ obj-$(CONFIG_NET_VENDOR_CIRRUS) += cirrus/ obj-$(CONFIG_NET_VENDOR_CISCO) += cisco/ diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig index 426916036151..acd53173fcc0 100644 --- a/drivers/net/ethernet/amd/Kconfig +++ b/drivers/net/ethernet/amd/Kconfig @@ -179,10 +179,8 @@ config SUNLANCE config AMD_XGBE tristate "AMD 10GbE Ethernet driver" - depends on (OF_NET || ACPI) && HAS_IOMEM && HAS_DMA + depends on ((OF_NET && OF_ADDRESS) || ACPI) && HAS_IOMEM && HAS_DMA depends on ARM64 || COMPILE_TEST - select PHYLIB - select AMD_XGBE_PHY select BITREVERSE select CRC32 select PTP_1588_CLOCK diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h index 34c28aac767f..b6fa89102526 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h @@ -857,6 +857,48 @@ */ #define PCS_MMD_SELECT 0xff +/* SerDes integration register offsets */ +#define SIR0_KR_RT_1 0x002c +#define SIR0_STATUS 0x0040 +#define SIR1_SPEED 0x0000 + +/* SerDes integration register entry bit positions and sizes */ +#define SIR0_KR_RT_1_RESET_INDEX 11 +#define SIR0_KR_RT_1_RESET_WIDTH 1 +#define SIR0_STATUS_RX_READY_INDEX 0 +#define SIR0_STATUS_RX_READY_WIDTH 1 +#define SIR0_STATUS_TX_READY_INDEX 8 +#define SIR0_STATUS_TX_READY_WIDTH 1 +#define SIR1_SPEED_CDR_RATE_INDEX 12 +#define SIR1_SPEED_CDR_RATE_WIDTH 4 +#define SIR1_SPEED_DATARATE_INDEX 4 +#define SIR1_SPEED_DATARATE_WIDTH 2 +#define SIR1_SPEED_PLLSEL_INDEX 3 +#define SIR1_SPEED_PLLSEL_WIDTH 1 +#define SIR1_SPEED_RATECHANGE_INDEX 6 +#define SIR1_SPEED_RATECHANGE_WIDTH 1 +#define SIR1_SPEED_TXAMP_INDEX 8 +#define SIR1_SPEED_TXAMP_WIDTH 4 +#define SIR1_SPEED_WORDMODE_INDEX 0 +#define SIR1_SPEED_WORDMODE_WIDTH 3 + +/* SerDes RxTx register offsets */ +#define RXTX_REG6 0x0018 +#define RXTX_REG20 0x0050 +#define RXTX_REG22 0x0058 +#define RXTX_REG114 0x01c8 +#define RXTX_REG129 0x0204 + +/* SerDes RxTx register entry bit positions and sizes */ +#define RXTX_REG6_RESETB_RXD_INDEX 8 +#define RXTX_REG6_RESETB_RXD_WIDTH 1 +#define RXTX_REG20_BLWC_ENA_INDEX 2 +#define RXTX_REG20_BLWC_ENA_WIDTH 1 +#define RXTX_REG114_PQ_REG_INDEX 9 +#define RXTX_REG114_PQ_REG_WIDTH 7 +#define RXTX_REG129_RXDFE_CONFIG_INDEX 14 +#define RXTX_REG129_RXDFE_CONFIG_WIDTH 2 + /* Descriptor/Packet entry bit positions and sizes */ #define RX_PACKET_ERRORS_CRC_INDEX 2 #define RX_PACKET_ERRORS_CRC_WIDTH 1 @@ -973,10 +1015,47 @@ #define TX_NORMAL_DESC2_VLAN_INSERT 0x2 /* MDIO undefined or vendor specific registers */ +#ifndef MDIO_PMA_10GBR_PMD_CTRL +#define MDIO_PMA_10GBR_PMD_CTRL 0x0096 +#endif + +#ifndef MDIO_PMA_10GBR_FECCTRL +#define MDIO_PMA_10GBR_FECCTRL 0x00ab +#endif + +#ifndef MDIO_AN_XNP +#define MDIO_AN_XNP 0x0016 +#endif + +#ifndef MDIO_AN_LPX +#define MDIO_AN_LPX 0x0019 +#endif + #ifndef MDIO_AN_COMP_STAT #define MDIO_AN_COMP_STAT 0x0030 #endif +#ifndef MDIO_AN_INTMASK +#define MDIO_AN_INTMASK 0x8001 +#endif + +#ifndef MDIO_AN_INT +#define MDIO_AN_INT 0x8002 +#endif + +#ifndef MDIO_CTRL1_SPEED1G +#define MDIO_CTRL1_SPEED1G (MDIO_CTRL1_SPEED10G & ~BMCR_SPEED100) +#endif + +/* MDIO mask values */ +#define XGBE_XNP_MCF_NULL_MESSAGE 0x001 +#define XGBE_XNP_ACK_PROCESSED BIT(12) +#define XGBE_XNP_MP_FORMATTED BIT(13) +#define XGBE_XNP_NP_EXCHANGE BIT(15) + +#define XGBE_KR_TRAINING_START BIT(0) +#define XGBE_KR_TRAINING_ENABLE BIT(1) + /* Bit setting and getting macros * The get macro will extract the current bit field value from within * the variable @@ -1119,6 +1198,82 @@ do { \ ioread32((_pdata)->xpcs_regs + (_off)) /* Macros for building, reading or writing register values or bits + * within the register values of SerDes integration registers. + */ +#define XSIR_GET_BITS(_var, _prefix, _field) \ + GET_BITS((_var), \ + _prefix##_##_field##_INDEX, \ + _prefix##_##_field##_WIDTH) + +#define XSIR_SET_BITS(_var, _prefix, _field, _val) \ + SET_BITS((_var), \ + _prefix##_##_field##_INDEX, \ + _prefix##_##_field##_WIDTH, (_val)) + +#define XSIR0_IOREAD(_pdata, _reg) \ + ioread16((_pdata)->sir0_regs + _reg) + +#define XSIR0_IOREAD_BITS(_pdata, _reg, _field) \ + GET_BITS(XSIR0_IOREAD((_pdata), _reg), \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH) + +#define XSIR0_IOWRITE(_pdata, _reg, _val) \ + iowrite16((_val), (_pdata)->sir0_regs + _reg) + +#define XSIR0_IOWRITE_BITS(_pdata, _reg, _field, _val) \ +do { \ + u16 reg_val = XSIR0_IOREAD((_pdata), _reg); \ + SET_BITS(reg_val, \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH, (_val)); \ + XSIR0_IOWRITE((_pdata), _reg, reg_val); \ +} while (0) + +#define XSIR1_IOREAD(_pdata, _reg) \ + ioread16((_pdata)->sir1_regs + _reg) + +#define XSIR1_IOREAD_BITS(_pdata, _reg, _field) \ + GET_BITS(XSIR1_IOREAD((_pdata), _reg), \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH) + +#define XSIR1_IOWRITE(_pdata, _reg, _val) \ + iowrite16((_val), (_pdata)->sir1_regs + _reg) + +#define XSIR1_IOWRITE_BITS(_pdata, _reg, _field, _val) \ +do { \ + u16 reg_val = XSIR1_IOREAD((_pdata), _reg); \ + SET_BITS(reg_val, \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH, (_val)); \ + XSIR1_IOWRITE((_pdata), _reg, reg_val); \ +} while (0) + +/* Macros for building, reading or writing register values or bits + * within the register values of SerDes RxTx registers. + */ +#define XRXTX_IOREAD(_pdata, _reg) \ + ioread16((_pdata)->rxtx_regs + _reg) + +#define XRXTX_IOREAD_BITS(_pdata, _reg, _field) \ + GET_BITS(XRXTX_IOREAD((_pdata), _reg), \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH) + +#define XRXTX_IOWRITE(_pdata, _reg, _val) \ + iowrite16((_val), (_pdata)->rxtx_regs + _reg) + +#define XRXTX_IOWRITE_BITS(_pdata, _reg, _field, _val) \ +do { \ + u16 reg_val = XRXTX_IOREAD((_pdata), _reg); \ + SET_BITS(reg_val, \ + _reg##_##_field##_INDEX, \ + _reg##_##_field##_WIDTH, (_val)); \ + XRXTX_IOWRITE((_pdata), _reg, reg_val); \ +} while (0) + +/* Macros for building, reading or writing register values or bits * using MDIO. Different from above because of the use of standardized * Linux include values. No shifting is performed with the bit * operations, everything works on mask values. diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c index 8a50b01c2686..a6b9899e285f 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c @@ -150,9 +150,12 @@ static int xgbe_dcb_ieee_setets(struct net_device *netdev, tc_ets = 0; tc_ets_weight = 0; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - DBGPR(" TC%u: tx_bw=%hhu, rx_bw=%hhu, tsa=%hhu\n", i, - ets->tc_tx_bw[i], ets->tc_rx_bw[i], ets->tc_tsa[i]); - DBGPR(" PRIO%u: TC=%hhu\n", i, ets->prio_tc[i]); + netif_dbg(pdata, drv, netdev, + "TC%u: tx_bw=%hhu, rx_bw=%hhu, tsa=%hhu\n", i, + ets->tc_tx_bw[i], ets->tc_rx_bw[i], + ets->tc_tsa[i]); + netif_dbg(pdata, drv, netdev, "PRIO%u: TC=%hhu\n", i, + ets->prio_tc[i]); if ((ets->tc_tx_bw[i] || ets->tc_tsa[i]) && (i >= pdata->hw_feat.tc_cnt)) @@ -214,8 +217,9 @@ static int xgbe_dcb_ieee_setpfc(struct net_device *netdev, { struct xgbe_prv_data *pdata = netdev_priv(netdev); - DBGPR(" cap=%hhu, en=%hhx, mbc=%hhu, delay=%hhu\n", - pfc->pfc_cap, pfc->pfc_en, pfc->mbc, pfc->delay); + netif_dbg(pdata, drv, netdev, + "cap=%hhu, en=%#hhx, mbc=%hhu, delay=%hhu\n", + pfc->pfc_cap, pfc->pfc_en, pfc->mbc, pfc->delay); if (!pdata->pfc) { pdata->pfc = devm_kzalloc(pdata->dev, sizeof(*pdata->pfc), @@ -238,9 +242,10 @@ static u8 xgbe_dcb_getdcbx(struct net_device *netdev) static u8 xgbe_dcb_setdcbx(struct net_device *netdev, u8 dcbx) { + struct xgbe_prv_data *pdata = netdev_priv(netdev); u8 support = xgbe_dcb_getdcbx(netdev); - DBGPR(" DCBX=%#hhx\n", dcbx); + netif_dbg(pdata, drv, netdev, "DCBX=%#hhx\n", dcbx); if (dcbx & ~support) return 1; diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c index d81fc6bd4759..dd03ad865caf 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c @@ -208,8 +208,9 @@ static int xgbe_init_ring(struct xgbe_prv_data *pdata, if (!ring->rdata) return -ENOMEM; - DBGPR(" rdesc=0x%p, rdesc_dma=0x%llx, rdata=0x%p\n", - ring->rdesc, ring->rdesc_dma, ring->rdata); + netif_dbg(pdata, drv, pdata->netdev, + "rdesc=%p, rdesc_dma=%pad, rdata=%p\n", + ring->rdesc, &ring->rdesc_dma, ring->rdata); DBGPR("<--xgbe_init_ring\n"); @@ -226,7 +227,9 @@ static int xgbe_alloc_ring_resources(struct xgbe_prv_data *pdata) channel = pdata->channel; for (i = 0; i < pdata->channel_count; i++, channel++) { - DBGPR(" %s - tx_ring:\n", channel->name); + netif_dbg(pdata, drv, pdata->netdev, "%s - Tx ring:\n", + channel->name); + ret = xgbe_init_ring(pdata, channel->tx_ring, pdata->tx_desc_count); if (ret) { @@ -235,12 +238,14 @@ static int xgbe_alloc_ring_resources(struct xgbe_prv_data *pdata) goto err_ring; } - DBGPR(" %s - rx_ring:\n", channel->name); + netif_dbg(pdata, drv, pdata->netdev, "%s - Rx ring:\n", + channel->name); + ret = xgbe_init_ring(pdata, channel->rx_ring, pdata->rx_desc_count); if (ret) { netdev_alert(pdata->netdev, - "error initializing Tx ring\n"); + "error initializing Rx ring\n"); goto err_ring; } } @@ -476,8 +481,6 @@ static void xgbe_unmap_rdata(struct xgbe_prv_data *pdata, if (rdata->state_saved) { rdata->state_saved = 0; - rdata->state.incomplete = 0; - rdata->state.context_next = 0; rdata->state.skb = NULL; rdata->state.len = 0; rdata->state.error = 0; @@ -518,8 +521,6 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb) rdata = XGBE_GET_DESC_DATA(ring, cur_index); if (tso) { - DBGPR(" TSO packet\n"); - /* Map the TSO header */ skb_dma = dma_map_single(pdata->dev, skb->data, packet->header_len, DMA_TO_DEVICE); @@ -529,6 +530,9 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb) } rdata->skb_dma = skb_dma; rdata->skb_dma_len = packet->header_len; + netif_dbg(pdata, tx_queued, pdata->netdev, + "skb header: index=%u, dma=%pad, len=%u\n", + cur_index, &skb_dma, packet->header_len); offset = packet->header_len; @@ -550,8 +554,9 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb) } rdata->skb_dma = skb_dma; rdata->skb_dma_len = len; - DBGPR(" skb data: index=%u, dma=0x%llx, len=%u\n", - cur_index, skb_dma, len); + netif_dbg(pdata, tx_queued, pdata->netdev, + "skb data: index=%u, dma=%pad, len=%u\n", + cur_index, &skb_dma, len); datalen -= len; offset += len; @@ -563,7 +568,8 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb) } for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { - DBGPR(" mapping frag %u\n", i); + netif_dbg(pdata, tx_queued, pdata->netdev, + "mapping frag %u\n", i); frag = &skb_shinfo(skb)->frags[i]; offset = 0; @@ -582,8 +588,9 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb) rdata->skb_dma = skb_dma; rdata->skb_dma_len = len; rdata->mapped_as_page = 1; - DBGPR(" skb data: index=%u, dma=0x%llx, len=%u\n", - cur_index, skb_dma, len); + netif_dbg(pdata, tx_queued, pdata->netdev, + "skb frag: index=%u, dma=%pad, len=%u\n", + cur_index, &skb_dma, len); datalen -= len; offset += len; diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c index 21d9497518fd..506e832c9e9a 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -710,7 +710,8 @@ static int xgbe_set_promiscuous_mode(struct xgbe_prv_data *pdata, if (XGMAC_IOREAD_BITS(pdata, MAC_PFR, PR) == val) return 0; - DBGPR(" %s promiscuous mode\n", enable ? "entering" : "leaving"); + netif_dbg(pdata, drv, pdata->netdev, "%s promiscuous mode\n", + enable ? "entering" : "leaving"); XGMAC_IOWRITE_BITS(pdata, MAC_PFR, PR, val); return 0; @@ -724,7 +725,8 @@ static int xgbe_set_all_multicast_mode(struct xgbe_prv_data *pdata, if (XGMAC_IOREAD_BITS(pdata, MAC_PFR, PM) == val) return 0; - DBGPR(" %s allmulti mode\n", enable ? "entering" : "leaving"); + netif_dbg(pdata, drv, pdata->netdev, "%s allmulti mode\n", + enable ? "entering" : "leaving"); XGMAC_IOWRITE_BITS(pdata, MAC_PFR, PM, val); return 0; @@ -749,8 +751,9 @@ static void xgbe_set_mac_reg(struct xgbe_prv_data *pdata, mac_addr[0] = ha->addr[4]; mac_addr[1] = ha->addr[5]; - DBGPR(" adding mac address %pM at 0x%04x\n", ha->addr, - *mac_reg); + netif_dbg(pdata, drv, pdata->netdev, + "adding mac address %pM at %#x\n", + ha->addr, *mac_reg); XGMAC_SET_BITS(mac_addr_hi, MAC_MACA1HR, AE, 1); } @@ -907,23 +910,6 @@ static void xgbe_write_mmd_regs(struct xgbe_prv_data *pdata, int prtad, else mmd_address = (pdata->mdio_mmd << 16) | (mmd_reg & 0xffff); - /* If the PCS is changing modes, match the MAC speed to it */ - if (((mmd_address >> 16) == MDIO_MMD_PCS) && - ((mmd_address & 0xffff) == MDIO_CTRL2)) { - struct phy_device *phydev = pdata->phydev; - - if (mmd_data & MDIO_PCS_CTRL2_TYPE) { - /* KX mode */ - if (phydev->supported & SUPPORTED_1000baseKX_Full) - xgbe_set_gmii_speed(pdata); - else - xgbe_set_gmii_2500_speed(pdata); - } else { - /* KR mode */ - xgbe_set_xgmii_speed(pdata); - } - } - /* The PCS registers are accessed using mmio. The underlying APB3 * management interface uses indirect addressing to access the MMD * register sets. This requires accessing of the PCS register in two @@ -1322,7 +1308,8 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata) for (i = 0; i < pdata->hw_feat.tc_cnt; i++) { switch (ets->tc_tsa[i]) { case IEEE_8021QAZ_TSA_STRICT: - DBGPR(" TC%u using SP\n", i); + netif_dbg(pdata, drv, pdata->netdev, + "TC%u using SP\n", i); XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_TC_ETSCR, TSA, MTL_TSA_SP); break; @@ -1330,7 +1317,8 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata) weight = total_weight * ets->tc_tx_bw[i] / 100; weight = clamp(weight, min_weight, total_weight); - DBGPR(" TC%u using DWRR (weight %u)\n", i, weight); + netif_dbg(pdata, drv, pdata->netdev, + "TC%u using DWRR (weight %u)\n", i, weight); XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_TC_ETSCR, TSA, MTL_TSA_ETS); XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_TC_QWR, QW, @@ -1359,7 +1347,8 @@ static void xgbe_config_dcb_pfc(struct xgbe_prv_data *pdata) } mask &= 0xff; - DBGPR(" TC%u PFC mask=%#x\n", tc, mask); + netif_dbg(pdata, drv, pdata->netdev, "TC%u PFC mask=%#x\n", + tc, mask); reg = MTL_TCPM0R + (MTL_TCPM_INC * (tc / MTL_TCPM_TC_PER_REG)); reg_val = XGMAC_IOREAD(pdata, reg); @@ -1457,8 +1446,9 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel) /* Create a context descriptor if this is a TSO packet */ if (tso_context || vlan_context) { if (tso_context) { - DBGPR(" TSO context descriptor, mss=%u\n", - packet->mss); + netif_dbg(pdata, tx_queued, pdata->netdev, + "TSO context descriptor, mss=%u\n", + packet->mss); /* Set the MSS size */ XGMAC_SET_BITS_LE(rdesc->desc2, TX_CONTEXT_DESC2, @@ -1476,8 +1466,9 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel) } if (vlan_context) { - DBGPR(" VLAN context descriptor, ctag=%u\n", - packet->vlan_ctag); + netif_dbg(pdata, tx_queued, pdata->netdev, + "VLAN context descriptor, ctag=%u\n", + packet->vlan_ctag); /* Mark it as a CONTEXT descriptor */ XGMAC_SET_BITS_LE(rdesc->desc3, TX_CONTEXT_DESC3, @@ -1533,6 +1524,8 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel) packet->tcp_payload_len); XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, TCPHDRLEN, packet->tcp_header_len / 4); + + pdata->ext_stats.tx_tso_packets++; } else { /* Enable CRC and Pad Insertion */ XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, CPC, 0); @@ -1594,9 +1587,9 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel) rdesc = rdata->rdesc; XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN, 1); -#ifdef XGMAC_ENABLE_TX_DESC_DUMP - xgbe_dump_tx_desc(ring, start_index, packet->rdesc_count, 1); -#endif + if (netif_msg_tx_queued(pdata)) + xgbe_dump_tx_desc(pdata, ring, start_index, + packet->rdesc_count, 1); /* Make sure ownership is written to the descriptor */ dma_wmb(); @@ -1618,11 +1611,12 @@ static void xgbe_dev_xmit(struct xgbe_channel *channel) static int xgbe_dev_read(struct xgbe_channel *channel) { + struct xgbe_prv_data *pdata = channel->pdata; struct xgbe_ring *ring = channel->rx_ring; struct xgbe_ring_data *rdata; struct xgbe_ring_desc *rdesc; struct xgbe_packet_data *packet = &ring->packet_data; - struct net_device *netdev = channel->pdata->netdev; + struct net_device *netdev = pdata->netdev; unsigned int err, etlt, l34t; DBGPR("-->xgbe_dev_read: cur = %d\n", ring->cur); @@ -1637,9 +1631,8 @@ static int xgbe_dev_read(struct xgbe_channel *channel) /* Make sure descriptor fields are read after reading the OWN bit */ dma_rmb(); -#ifdef XGMAC_ENABLE_RX_DESC_DUMP - xgbe_dump_rx_desc(ring, rdesc, ring->cur); -#endif + if (netif_msg_rx_status(pdata)) + xgbe_dump_rx_desc(pdata, ring, ring->cur); if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, CTXT)) { /* Timestamp Context Descriptor */ @@ -1661,9 +1654,12 @@ static int xgbe_dev_read(struct xgbe_channel *channel) CONTEXT_NEXT, 1); /* Get the header length */ - if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, FD)) + if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, FD)) { rdata->rx.hdr_len = XGMAC_GET_BITS_LE(rdesc->desc2, RX_NORMAL_DESC2, HL); + if (rdata->rx.hdr_len) + pdata->ext_stats.rx_split_header_packets++; + } /* Get the RSS hash */ if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, RSV)) { @@ -1700,14 +1696,14 @@ static int xgbe_dev_read(struct xgbe_channel *channel) INCOMPLETE, 0); /* Set checksum done indicator as appropriate */ - if (channel->pdata->netdev->features & NETIF_F_RXCSUM) + if (netdev->features & NETIF_F_RXCSUM) XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES, CSUM_DONE, 1); /* Check for errors (only valid in last descriptor) */ err = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ES); etlt = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ETLT); - DBGPR(" err=%u, etlt=%#x\n", err, etlt); + netif_dbg(pdata, rx_status, netdev, "err=%u, etlt=%#x\n", err, etlt); if (!err || !etlt) { /* No error if err is 0 or etlt is 0 */ @@ -1718,7 +1714,8 @@ static int xgbe_dev_read(struct xgbe_channel *channel) packet->vlan_ctag = XGMAC_GET_BITS_LE(rdesc->desc0, RX_NORMAL_DESC0, OVT); - DBGPR(" vlan-ctag=0x%04x\n", packet->vlan_ctag); + netif_dbg(pdata, rx_status, netdev, "vlan-ctag=%#06x\n", + packet->vlan_ctag); } } else { if ((etlt == 0x05) || (etlt == 0x06)) @@ -2026,9 +2023,9 @@ static void xgbe_config_tx_fifo_size(struct xgbe_prv_data *pdata) for (i = 0; i < pdata->tx_q_count; i++) XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_TQOMR, TQS, fifo_size); - netdev_notice(pdata->netdev, - "%d Tx hardware queues, %d byte fifo per queue\n", - pdata->tx_q_count, ((fifo_size + 1) * 256)); + netif_info(pdata, drv, pdata->netdev, + "%d Tx hardware queues, %d byte fifo per queue\n", + pdata->tx_q_count, ((fifo_size + 1) * 256)); } static void xgbe_config_rx_fifo_size(struct xgbe_prv_data *pdata) @@ -2042,9 +2039,9 @@ static void xgbe_config_rx_fifo_size(struct xgbe_prv_data *pdata) for (i = 0; i < pdata->rx_q_count; i++) XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, RQS, fifo_size); - netdev_notice(pdata->netdev, - "%d Rx hardware queues, %d byte fifo per queue\n", - pdata->rx_q_count, ((fifo_size + 1) * 256)); + netif_info(pdata, drv, pdata->netdev, + "%d Rx hardware queues, %d byte fifo per queue\n", + pdata->rx_q_count, ((fifo_size + 1) * 256)); } static void xgbe_config_queue_mapping(struct xgbe_prv_data *pdata) @@ -2063,14 +2060,16 @@ static void xgbe_config_queue_mapping(struct xgbe_prv_data *pdata) for (i = 0, queue = 0; i < pdata->hw_feat.tc_cnt; i++) { for (j = 0; j < qptc; j++) { - DBGPR(" TXq%u mapped to TC%u\n", queue, i); + netif_dbg(pdata, drv, pdata->netdev, + "TXq%u mapped to TC%u\n", queue, i); XGMAC_MTL_IOWRITE_BITS(pdata, queue, MTL_Q_TQOMR, Q2TCMAP, i); pdata->q2tc_map[queue++] = i; } if (i < qptc_extra) { - DBGPR(" TXq%u mapped to TC%u\n", queue, i); + netif_dbg(pdata, drv, pdata->netdev, + "TXq%u mapped to TC%u\n", queue, i); XGMAC_MTL_IOWRITE_BITS(pdata, queue, MTL_Q_TQOMR, Q2TCMAP, i); pdata->q2tc_map[queue++] = i; @@ -2088,13 +2087,15 @@ static void xgbe_config_queue_mapping(struct xgbe_prv_data *pdata) for (i = 0, prio = 0; i < prio_queues;) { mask = 0; for (j = 0; j < ppq; j++) { - DBGPR(" PRIO%u mapped to RXq%u\n", prio, i); + netif_dbg(pdata, drv, pdata->netdev, + "PRIO%u mapped to RXq%u\n", prio, i); mask |= (1 << prio); pdata->prio2q_map[prio++] = i; } if (i < ppq_extra) { - DBGPR(" PRIO%u mapped to RXq%u\n", prio, i); + netif_dbg(pdata, drv, pdata->netdev, + "PRIO%u mapped to RXq%u\n", prio, i); mask |= (1 << prio); pdata->prio2q_map[prio++] = i; } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index db84ddcfec84..c75204909702 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -183,9 +183,10 @@ static int xgbe_alloc_channels(struct xgbe_prv_data *pdata) channel->rx_ring = rx_ring++; } - DBGPR(" %s: queue=%u, dma_regs=%p, dma_irq=%d, tx=%p, rx=%p\n", - channel->name, channel->queue_index, channel->dma_regs, - channel->dma_irq, channel->tx_ring, channel->rx_ring); + netif_dbg(pdata, drv, pdata->netdev, + "%s: dma_regs=%p, dma_irq=%d, tx=%p, rx=%p\n", + channel->name, channel->dma_regs, channel->dma_irq, + channel->tx_ring, channel->rx_ring); } pdata->channel = channel_mem; @@ -235,7 +236,8 @@ static int xgbe_maybe_stop_tx_queue(struct xgbe_channel *channel, struct xgbe_prv_data *pdata = channel->pdata; if (count > xgbe_tx_avail_desc(ring)) { - DBGPR(" Tx queue stopped, not enough descriptors available\n"); + netif_info(pdata, drv, pdata->netdev, + "Tx queue stopped, not enough descriptors available\n"); netif_stop_subqueue(pdata->netdev, channel->queue_index); ring->tx.queue_stopped = 1; @@ -330,7 +332,7 @@ static irqreturn_t xgbe_isr(int irq, void *data) if (!dma_isr) goto isr_done; - DBGPR(" DMA_ISR = %08x\n", dma_isr); + netif_dbg(pdata, intr, pdata->netdev, "DMA_ISR=%#010x\n", dma_isr); for (i = 0; i < pdata->channel_count; i++) { if (!(dma_isr & (1 << i))) @@ -339,7 +341,8 @@ static irqreturn_t xgbe_isr(int irq, void *data) channel = pdata->channel + i; dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR); - DBGPR(" DMA_CH%u_ISR = %08x\n", i, dma_ch_isr); + netif_dbg(pdata, intr, pdata->netdev, "DMA_CH%u_ISR=%#010x\n", + i, dma_ch_isr); /* The TI or RI interrupt bits may still be set even if using * per channel DMA interrupts. Check to be sure those are not @@ -386,8 +389,6 @@ static irqreturn_t xgbe_isr(int irq, void *data) } } - DBGPR(" DMA_ISR = %08x\n", XGMAC_IOREAD(pdata, DMA_ISR)); - isr_done: return IRQ_HANDLED; } @@ -436,43 +437,61 @@ static void xgbe_tx_timer(unsigned long data) DBGPR("<--xgbe_tx_timer\n"); } -static void xgbe_init_tx_timers(struct xgbe_prv_data *pdata) +static void xgbe_service(struct work_struct *work) +{ + struct xgbe_prv_data *pdata = container_of(work, + struct xgbe_prv_data, + service_work); + + pdata->phy_if.phy_status(pdata); +} + +static void xgbe_service_timer(unsigned long data) +{ + struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + + schedule_work(&pdata->service_work); + + mod_timer(&pdata->service_timer, jiffies + HZ); +} + +static void xgbe_init_timers(struct xgbe_prv_data *pdata) { struct xgbe_channel *channel; unsigned int i; - DBGPR("-->xgbe_init_tx_timers\n"); + setup_timer(&pdata->service_timer, xgbe_service_timer, + (unsigned long)pdata); channel = pdata->channel; for (i = 0; i < pdata->channel_count; i++, channel++) { if (!channel->tx_ring) break; - DBGPR(" %s adding tx timer\n", channel->name); setup_timer(&channel->tx_timer, xgbe_tx_timer, (unsigned long)channel); } +} - DBGPR("<--xgbe_init_tx_timers\n"); +static void xgbe_start_timers(struct xgbe_prv_data *pdata) +{ + mod_timer(&pdata->service_timer, jiffies + HZ); } -static void xgbe_stop_tx_timers(struct xgbe_prv_data *pdata) +static void xgbe_stop_timers(struct xgbe_prv_data *pdata) { struct xgbe_channel *channel; unsigned int i; - DBGPR("-->xgbe_stop_tx_timers\n"); + del_timer_sync(&pdata->service_timer); channel = pdata->channel; for (i = 0; i < pdata->channel_count; i++, channel++) { if (!channel->tx_ring) break; - DBGPR(" %s deleting tx timer\n", channel->name); del_timer_sync(&channel->tx_timer); } - - DBGPR("<--xgbe_stop_tx_timers\n"); } void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata) @@ -512,6 +531,7 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata) RXFIFOSIZE); hw_feat->tx_fifo_size = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, TXFIFOSIZE); + hw_feat->adv_ts_hi = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, ADVTHWORD); hw_feat->dma_width = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, ADDR64); hw_feat->dcb = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, DCBEN); hw_feat->sph = XGMAC_GET_BITS(mac_hfr1, MAC_HWF1R, SPHEN); @@ -759,112 +779,12 @@ static void xgbe_free_rx_data(struct xgbe_prv_data *pdata) DBGPR("<--xgbe_free_rx_data\n"); } -static void xgbe_adjust_link(struct net_device *netdev) -{ - struct xgbe_prv_data *pdata = netdev_priv(netdev); - struct xgbe_hw_if *hw_if = &pdata->hw_if; - struct phy_device *phydev = pdata->phydev; - int new_state = 0; - - if (!phydev) - return; - - if (phydev->link) { - /* Flow control support */ - if (pdata->pause_autoneg) { - if (phydev->pause || phydev->asym_pause) { - pdata->tx_pause = 1; - pdata->rx_pause = 1; - } else { - pdata->tx_pause = 0; - pdata->rx_pause = 0; - } - } - - if (pdata->tx_pause != pdata->phy_tx_pause) { - hw_if->config_tx_flow_control(pdata); - pdata->phy_tx_pause = pdata->tx_pause; - } - - if (pdata->rx_pause != pdata->phy_rx_pause) { - hw_if->config_rx_flow_control(pdata); - pdata->phy_rx_pause = pdata->rx_pause; - } - - /* Speed support */ - if (phydev->speed != pdata->phy_speed) { - new_state = 1; - - switch (phydev->speed) { - case SPEED_10000: - hw_if->set_xgmii_speed(pdata); - break; - - case SPEED_2500: - hw_if->set_gmii_2500_speed(pdata); - break; - - case SPEED_1000: - hw_if->set_gmii_speed(pdata); - break; - } - pdata->phy_speed = phydev->speed; - } - - if (phydev->link != pdata->phy_link) { - new_state = 1; - pdata->phy_link = 1; - } - } else if (pdata->phy_link) { - new_state = 1; - pdata->phy_link = 0; - pdata->phy_speed = SPEED_UNKNOWN; - } - - if (new_state) - phy_print_status(phydev); -} - static int xgbe_phy_init(struct xgbe_prv_data *pdata) { - struct net_device *netdev = pdata->netdev; - struct phy_device *phydev = pdata->phydev; - int ret; - pdata->phy_link = -1; pdata->phy_speed = SPEED_UNKNOWN; - pdata->phy_tx_pause = pdata->tx_pause; - pdata->phy_rx_pause = pdata->rx_pause; - ret = phy_connect_direct(netdev, phydev, &xgbe_adjust_link, - pdata->phy_mode); - if (ret) { - netdev_err(netdev, "phy_connect_direct failed\n"); - return ret; - } - - if (!phydev->drv || (phydev->drv->phy_id == 0)) { - netdev_err(netdev, "phy_id not valid\n"); - ret = -ENODEV; - goto err_phy_connect; - } - DBGPR(" phy_connect_direct succeeded for PHY %s, link=%d\n", - dev_name(&phydev->dev), phydev->link); - - return 0; - -err_phy_connect: - phy_disconnect(phydev); - - return ret; -} - -static void xgbe_phy_exit(struct xgbe_prv_data *pdata) -{ - if (!pdata->phydev) - return; - - phy_disconnect(pdata->phydev); + return pdata->phy_if.phy_reset(pdata); } int xgbe_powerdown(struct net_device *netdev, unsigned int caller) @@ -889,13 +809,14 @@ int xgbe_powerdown(struct net_device *netdev, unsigned int caller) netif_tx_stop_all_queues(netdev); + xgbe_stop_timers(pdata); + flush_workqueue(pdata->dev_workqueue); + hw_if->powerdown_tx(pdata); hw_if->powerdown_rx(pdata); xgbe_napi_disable(pdata, 0); - phy_stop(pdata->phydev); - pdata->power_down = 1; spin_unlock_irqrestore(&pdata->lock, flags); @@ -924,8 +845,6 @@ int xgbe_powerup(struct net_device *netdev, unsigned int caller) pdata->power_down = 0; - phy_start(pdata->phydev); - xgbe_napi_enable(pdata, 0); hw_if->powerup_tx(pdata); @@ -936,6 +855,8 @@ int xgbe_powerup(struct net_device *netdev, unsigned int caller) netif_tx_start_all_queues(netdev); + xgbe_start_timers(pdata); + spin_unlock_irqrestore(&pdata->lock, flags); DBGPR("<--xgbe_powerup\n"); @@ -946,6 +867,7 @@ int xgbe_powerup(struct net_device *netdev, unsigned int caller) static int xgbe_start(struct xgbe_prv_data *pdata) { struct xgbe_hw_if *hw_if = &pdata->hw_if; + struct xgbe_phy_if *phy_if = &pdata->phy_if; struct net_device *netdev = pdata->netdev; int ret; @@ -953,7 +875,9 @@ static int xgbe_start(struct xgbe_prv_data *pdata) hw_if->init(pdata); - phy_start(pdata->phydev); + ret = phy_if->phy_start(pdata); + if (ret) + goto err_phy; xgbe_napi_enable(pdata, 1); @@ -964,10 +888,11 @@ static int xgbe_start(struct xgbe_prv_data *pdata) hw_if->enable_tx(pdata); hw_if->enable_rx(pdata); - xgbe_init_tx_timers(pdata); - netif_tx_start_all_queues(netdev); + xgbe_start_timers(pdata); + schedule_work(&pdata->service_work); + DBGPR("<--xgbe_start\n"); return 0; @@ -975,8 +900,9 @@ static int xgbe_start(struct xgbe_prv_data *pdata) err_napi: xgbe_napi_disable(pdata, 1); - phy_stop(pdata->phydev); + phy_if->phy_stop(pdata); +err_phy: hw_if->exit(pdata); return ret; @@ -985,6 +911,7 @@ err_napi: static void xgbe_stop(struct xgbe_prv_data *pdata) { struct xgbe_hw_if *hw_if = &pdata->hw_if; + struct xgbe_phy_if *phy_if = &pdata->phy_if; struct xgbe_channel *channel; struct net_device *netdev = pdata->netdev; struct netdev_queue *txq; @@ -994,7 +921,8 @@ static void xgbe_stop(struct xgbe_prv_data *pdata) netif_tx_stop_all_queues(netdev); - xgbe_stop_tx_timers(pdata); + xgbe_stop_timers(pdata); + flush_workqueue(pdata->dev_workqueue); hw_if->disable_tx(pdata); hw_if->disable_rx(pdata); @@ -1003,7 +931,7 @@ static void xgbe_stop(struct xgbe_prv_data *pdata) xgbe_napi_disable(pdata, 1); - phy_stop(pdata->phydev); + phy_if->phy_stop(pdata); hw_if->exit(pdata); @@ -1374,7 +1302,7 @@ static int xgbe_open(struct net_device *netdev) ret = clk_prepare_enable(pdata->sysclk); if (ret) { netdev_alert(netdev, "dma clk_prepare_enable failed\n"); - goto err_phy_init; + return ret; } ret = clk_prepare_enable(pdata->ptpclk); @@ -1399,14 +1327,17 @@ static int xgbe_open(struct net_device *netdev) if (ret) goto err_channels; - /* Initialize the device restart and Tx timestamp work struct */ + INIT_WORK(&pdata->service_work, xgbe_service); INIT_WORK(&pdata->restart_work, xgbe_restart); INIT_WORK(&pdata->tx_tstamp_work, xgbe_tx_tstamp); + xgbe_init_timers(pdata); ret = xgbe_start(pdata); if (ret) goto err_rings; + clear_bit(XGBE_DOWN, &pdata->dev_state); + DBGPR("<--xgbe_open\n"); return 0; @@ -1423,9 +1354,6 @@ err_ptpclk: err_sysclk: clk_disable_unprepare(pdata->sysclk); -err_phy_init: - xgbe_phy_exit(pdata); - return ret; } @@ -1449,8 +1377,7 @@ static int xgbe_close(struct net_device *netdev) clk_disable_unprepare(pdata->ptpclk); clk_disable_unprepare(pdata->sysclk); - /* Release the phy */ - xgbe_phy_exit(pdata); + set_bit(XGBE_DOWN, &pdata->dev_state); DBGPR("<--xgbe_close\n"); @@ -1478,7 +1405,8 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev) ret = NETDEV_TX_OK; if (skb->len == 0) { - netdev_err(netdev, "empty skb received from stack\n"); + netif_err(pdata, tx_err, netdev, + "empty skb received from stack\n"); dev_kfree_skb_any(skb); goto tx_netdev_return; } @@ -1494,7 +1422,8 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev) ret = xgbe_prep_tso(skb, packet); if (ret) { - netdev_err(netdev, "error processing TSO packet\n"); + netif_err(pdata, tx_err, netdev, + "error processing TSO packet\n"); dev_kfree_skb_any(skb); goto tx_netdev_return; } @@ -1513,9 +1442,8 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev) /* Configure required descriptor fields for transmission */ hw_if->dev_xmit(channel); -#ifdef XGMAC_ENABLE_TX_PKT_DUMP - xgbe_print_pkt(netdev, skb, true); -#endif + if (netif_msg_pktdata(pdata)) + xgbe_print_pkt(netdev, skb, true); /* Stop the queue in advance if there may not be enough descriptors */ xgbe_maybe_stop_tx_queue(channel, ring, XGBE_TX_MAX_DESCS); @@ -1710,7 +1638,8 @@ static int xgbe_setup_tc(struct net_device *netdev, u8 tc) (pdata->q2tc_map[queue] == i)) queue++; - DBGPR(" TC%u using TXq%u-%u\n", i, offset, queue - 1); + netif_dbg(pdata, drv, netdev, "TC%u using TXq%u-%u\n", + i, offset, queue - 1); netdev_set_tc_queue(netdev, i, queue - offset, offset); offset = queue; } @@ -1820,9 +1749,10 @@ static void xgbe_rx_refresh(struct xgbe_channel *channel) lower_32_bits(rdata->rdesc_dma)); } -static struct sk_buff *xgbe_create_skb(struct napi_struct *napi, +static struct sk_buff *xgbe_create_skb(struct xgbe_prv_data *pdata, + struct napi_struct *napi, struct xgbe_ring_data *rdata, - unsigned int *len) + unsigned int len) { struct sk_buff *skb; u8 *packet; @@ -1832,14 +1762,31 @@ static struct sk_buff *xgbe_create_skb(struct napi_struct *napi, if (!skb) return NULL; + /* Start with the header buffer which may contain just the header + * or the header plus data + */ + dma_sync_single_for_cpu(pdata->dev, rdata->rx.hdr.dma, + rdata->rx.hdr.dma_len, DMA_FROM_DEVICE); + packet = page_address(rdata->rx.hdr.pa.pages) + rdata->rx.hdr.pa.pages_offset; - copy_len = (rdata->rx.hdr_len) ? rdata->rx.hdr_len : *len; + copy_len = (rdata->rx.hdr_len) ? rdata->rx.hdr_len : len; copy_len = min(rdata->rx.hdr.dma_len, copy_len); skb_copy_to_linear_data(skb, packet, copy_len); skb_put(skb, copy_len); - *len -= copy_len; + len -= copy_len; + if (len) { + /* Add the remaining data as a frag */ + dma_sync_single_for_cpu(pdata->dev, rdata->rx.buf.dma, + rdata->rx.buf.dma_len, DMA_FROM_DEVICE); + + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, + rdata->rx.buf.pa.pages, + rdata->rx.buf.pa.pages_offset, + len, rdata->rx.buf.dma_len); + rdata->rx.buf.pa.pages = NULL; + } return skb; } @@ -1877,9 +1824,8 @@ static int xgbe_tx_poll(struct xgbe_channel *channel) * bit */ dma_rmb(); -#ifdef XGMAC_ENABLE_TX_DESC_DUMP - xgbe_dump_tx_desc(ring, ring->dirty, 1, 0); -#endif + if (netif_msg_tx_done(pdata)) + xgbe_dump_tx_desc(pdata, ring, ring->dirty, 1, 0); if (hw_if->is_last_desc(rdesc)) { tx_packets += rdata->tx.packets; @@ -1922,7 +1868,7 @@ static int xgbe_rx_poll(struct xgbe_channel *channel, int budget) struct sk_buff *skb; struct skb_shared_hwtstamps *hwtstamps; unsigned int incomplete, error, context_next, context; - unsigned int len, put_len, max_len; + unsigned int len, rdesc_len, max_len; unsigned int received = 0; int packet_count = 0; @@ -1932,6 +1878,9 @@ static int xgbe_rx_poll(struct xgbe_channel *channel, int budget) if (!ring) return 0; + incomplete = 0; + context_next = 0; + napi = (pdata->per_channel_irq) ? &channel->napi : &pdata->napi; rdata = XGBE_GET_DESC_DATA(ring, ring->cur); @@ -1941,15 +1890,11 @@ static int xgbe_rx_poll(struct xgbe_channel *channel, int budget) /* First time in loop see if we need to restore state */ if (!received && rdata->state_saved) { - incomplete = rdata->state.incomplete; - context_next = rdata->state.context_next; skb = rdata->state.skb; error = rdata->state.error; len = rdata->state.len; } else { memset(packet, 0, sizeof(*packet)); - incomplete = 0; - context_next = 0; skb = NULL; error = 0; len = 0; @@ -1983,29 +1928,23 @@ read_again: if (error || packet->errors) { if (packet->errors) - DBGPR("Error in received packet\n"); + netif_err(pdata, rx_err, netdev, + "error in received packet\n"); dev_kfree_skb(skb); goto next_packet; } if (!context) { - put_len = rdata->rx.len - len; - len += put_len; - - if (!skb) { - dma_sync_single_for_cpu(pdata->dev, - rdata->rx.hdr.dma, - rdata->rx.hdr.dma_len, - DMA_FROM_DEVICE); - - skb = xgbe_create_skb(napi, rdata, &put_len); - if (!skb) { + /* Length is cumulative, get this descriptor's length */ + rdesc_len = rdata->rx.len - len; + len += rdesc_len; + + if (rdesc_len && !skb) { + skb = xgbe_create_skb(pdata, napi, rdata, + rdesc_len); + if (!skb) error = 1; - goto skip_data; - } - } - - if (put_len) { + } else if (rdesc_len) { dma_sync_single_for_cpu(pdata->dev, rdata->rx.buf.dma, rdata->rx.buf.dma_len, @@ -2014,12 +1953,12 @@ read_again: skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rdata->rx.buf.pa.pages, rdata->rx.buf.pa.pages_offset, - put_len, rdata->rx.buf.dma_len); + rdesc_len, + rdata->rx.buf.dma_len); rdata->rx.buf.pa.pages = NULL; } } -skip_data: if (incomplete || context_next) goto read_again; @@ -2033,14 +1972,14 @@ skip_data: max_len += VLAN_HLEN; if (skb->len > max_len) { - DBGPR("packet length exceeds configured MTU\n"); + netif_err(pdata, rx_err, netdev, + "packet length exceeds configured MTU\n"); dev_kfree_skb(skb); goto next_packet; } -#ifdef XGMAC_ENABLE_RX_PKT_DUMP - xgbe_print_pkt(netdev, skb, false); -#endif + if (netif_msg_pktdata(pdata)) + xgbe_print_pkt(netdev, skb, false); skb_checksum_none_assert(skb); if (XGMAC_GET_BITS(packet->attributes, @@ -2072,7 +2011,6 @@ skip_data: skb_record_rx_queue(skb, channel->queue_index); skb_mark_napi_id(skb, napi); - netdev->last_rx = jiffies; napi_gro_receive(napi, skb); next_packet: @@ -2083,8 +2021,6 @@ next_packet: if (received && (incomplete || context_next)) { rdata = XGBE_GET_DESC_DATA(ring, ring->cur); rdata->state_saved = 1; - rdata->state.incomplete = incomplete; - rdata->state.context_next = context_next; rdata->state.skb = skb; rdata->state.len = len; rdata->state.error = error; @@ -2165,8 +2101,8 @@ static int xgbe_all_poll(struct napi_struct *napi, int budget) return processed; } -void xgbe_dump_tx_desc(struct xgbe_ring *ring, unsigned int idx, - unsigned int count, unsigned int flag) +void xgbe_dump_tx_desc(struct xgbe_prv_data *pdata, struct xgbe_ring *ring, + unsigned int idx, unsigned int count, unsigned int flag) { struct xgbe_ring_data *rdata; struct xgbe_ring_desc *rdesc; @@ -2174,20 +2110,29 @@ void xgbe_dump_tx_desc(struct xgbe_ring *ring, unsigned int idx, while (count--) { rdata = XGBE_GET_DESC_DATA(ring, idx); rdesc = rdata->rdesc; - pr_alert("TX_NORMAL_DESC[%d %s] = %08x:%08x:%08x:%08x\n", idx, - (flag == 1) ? "QUEUED FOR TX" : "TX BY DEVICE", - le32_to_cpu(rdesc->desc0), le32_to_cpu(rdesc->desc1), - le32_to_cpu(rdesc->desc2), le32_to_cpu(rdesc->desc3)); + netdev_dbg(pdata->netdev, + "TX_NORMAL_DESC[%d %s] = %08x:%08x:%08x:%08x\n", idx, + (flag == 1) ? "QUEUED FOR TX" : "TX BY DEVICE", + le32_to_cpu(rdesc->desc0), + le32_to_cpu(rdesc->desc1), + le32_to_cpu(rdesc->desc2), + le32_to_cpu(rdesc->desc3)); idx++; } } -void xgbe_dump_rx_desc(struct xgbe_ring *ring, struct xgbe_ring_desc *desc, +void xgbe_dump_rx_desc(struct xgbe_prv_data *pdata, struct xgbe_ring *ring, unsigned int idx) { - pr_alert("RX_NORMAL_DESC[%d RX BY DEVICE] = %08x:%08x:%08x:%08x\n", idx, - le32_to_cpu(desc->desc0), le32_to_cpu(desc->desc1), - le32_to_cpu(desc->desc2), le32_to_cpu(desc->desc3)); + struct xgbe_ring_data *rdata; + struct xgbe_ring_desc *rdesc; + + rdata = XGBE_GET_DESC_DATA(ring, idx); + rdesc = rdata->rdesc; + netdev_dbg(pdata->netdev, + "RX_NORMAL_DESC[%d RX BY DEVICE] = %08x:%08x:%08x:%08x\n", + idx, le32_to_cpu(rdesc->desc0), le32_to_cpu(rdesc->desc1), + le32_to_cpu(rdesc->desc2), le32_to_cpu(rdesc->desc3)); } void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx) @@ -2197,21 +2142,21 @@ void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx) unsigned char buffer[128]; unsigned int i, j; - netdev_alert(netdev, "\n************** SKB dump ****************\n"); + netdev_dbg(netdev, "\n************** SKB dump ****************\n"); - netdev_alert(netdev, "%s packet of %d bytes\n", - (tx_rx ? "TX" : "RX"), skb->len); + netdev_dbg(netdev, "%s packet of %d bytes\n", + (tx_rx ? "TX" : "RX"), skb->len); - netdev_alert(netdev, "Dst MAC addr: %pM\n", eth->h_dest); - netdev_alert(netdev, "Src MAC addr: %pM\n", eth->h_source); - netdev_alert(netdev, "Protocol: 0x%04hx\n", ntohs(eth->h_proto)); + netdev_dbg(netdev, "Dst MAC addr: %pM\n", eth->h_dest); + netdev_dbg(netdev, "Src MAC addr: %pM\n", eth->h_source); + netdev_dbg(netdev, "Protocol: %#06hx\n", ntohs(eth->h_proto)); for (i = 0, j = 0; i < skb->len;) { j += snprintf(buffer + j, sizeof(buffer) - j, "%02hhx", buf[i++]); if ((i % 32) == 0) { - netdev_alert(netdev, " 0x%04x: %s\n", i - 32, buffer); + netdev_dbg(netdev, " %#06x: %s\n", i - 32, buffer); j = 0; } else if ((i % 16) == 0) { buffer[j++] = ' '; @@ -2221,7 +2166,7 @@ void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx) } } if (i % 32) - netdev_alert(netdev, " 0x%04x: %s\n", i - (i % 32), buffer); + netdev_dbg(netdev, " %#06x: %s\n", i - (i % 32), buffer); - netdev_alert(netdev, "\n************** SKB dump ****************\n"); + netdev_dbg(netdev, "\n************** SKB dump ****************\n"); } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c index 5f149e8ee20f..59e090e95c0e 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c @@ -133,6 +133,12 @@ struct xgbe_stats { offsetof(struct xgbe_prv_data, mmc_stats._var), \ } +#define XGMAC_EXT_STAT(_string, _var) \ + { _string, \ + FIELD_SIZEOF(struct xgbe_ext_stats, _var), \ + offsetof(struct xgbe_prv_data, ext_stats._var), \ + } + static const struct xgbe_stats xgbe_gstring_stats[] = { XGMAC_MMC_STAT("tx_bytes", txoctetcount_gb), XGMAC_MMC_STAT("tx_packets", txframecount_gb), @@ -140,6 +146,7 @@ static const struct xgbe_stats xgbe_gstring_stats[] = { XGMAC_MMC_STAT("tx_broadcast_packets", txbroadcastframes_gb), XGMAC_MMC_STAT("tx_multicast_packets", txmulticastframes_gb), XGMAC_MMC_STAT("tx_vlan_packets", txvlanframes_g), + XGMAC_EXT_STAT("tx_tso_packets", tx_tso_packets), XGMAC_MMC_STAT("tx_64_byte_packets", tx64octets_gb), XGMAC_MMC_STAT("tx_65_to_127_byte_packets", tx65to127octets_gb), XGMAC_MMC_STAT("tx_128_to_255_byte_packets", tx128to255octets_gb), @@ -171,6 +178,7 @@ static const struct xgbe_stats xgbe_gstring_stats[] = { XGMAC_MMC_STAT("rx_fifo_overflow_errors", rxfifooverflow), XGMAC_MMC_STAT("rx_watchdog_errors", rxwatchdogerror), XGMAC_MMC_STAT("rx_pause_frames", rxpauseframes), + XGMAC_EXT_STAT("rx_split_header_packets", rx_split_header_packets), }; #define XGBE_STATS_COUNT ARRAY_SIZE(xgbe_gstring_stats) @@ -239,9 +247,9 @@ static void xgbe_get_pauseparam(struct net_device *netdev, DBGPR("-->xgbe_get_pauseparam\n"); - pause->autoneg = pdata->pause_autoneg; - pause->tx_pause = pdata->tx_pause; - pause->rx_pause = pdata->rx_pause; + pause->autoneg = pdata->phy.pause_autoneg; + pause->tx_pause = pdata->phy.tx_pause; + pause->rx_pause = pdata->phy.rx_pause; DBGPR("<--xgbe_get_pauseparam\n"); } @@ -250,7 +258,6 @@ static int xgbe_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) { struct xgbe_prv_data *pdata = netdev_priv(netdev); - struct phy_device *phydev = pdata->phydev; int ret = 0; DBGPR("-->xgbe_set_pauseparam\n"); @@ -258,21 +265,26 @@ static int xgbe_set_pauseparam(struct net_device *netdev, DBGPR(" autoneg = %d, tx_pause = %d, rx_pause = %d\n", pause->autoneg, pause->tx_pause, pause->rx_pause); - pdata->pause_autoneg = pause->autoneg; - if (pause->autoneg) { - phydev->advertising |= ADVERTISED_Pause; - phydev->advertising |= ADVERTISED_Asym_Pause; + if (pause->autoneg && (pdata->phy.autoneg != AUTONEG_ENABLE)) + return -EINVAL; + + pdata->phy.pause_autoneg = pause->autoneg; + pdata->phy.tx_pause = pause->tx_pause; + pdata->phy.rx_pause = pause->rx_pause; - } else { - phydev->advertising &= ~ADVERTISED_Pause; - phydev->advertising &= ~ADVERTISED_Asym_Pause; + pdata->phy.advertising &= ~ADVERTISED_Pause; + pdata->phy.advertising &= ~ADVERTISED_Asym_Pause; - pdata->tx_pause = pause->tx_pause; - pdata->rx_pause = pause->rx_pause; + if (pause->rx_pause) { + pdata->phy.advertising |= ADVERTISED_Pause; + pdata->phy.advertising |= ADVERTISED_Asym_Pause; } + if (pause->tx_pause) + pdata->phy.advertising ^= ADVERTISED_Asym_Pause; + if (netif_running(netdev)) - ret = phy_start_aneg(phydev); + ret = pdata->phy_if.phy_config_aneg(pdata); DBGPR("<--xgbe_set_pauseparam\n"); @@ -283,36 +295,39 @@ static int xgbe_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { struct xgbe_prv_data *pdata = netdev_priv(netdev); - int ret; DBGPR("-->xgbe_get_settings\n"); - if (!pdata->phydev) - return -ENODEV; + cmd->phy_address = pdata->phy.address; + + cmd->supported = pdata->phy.supported; + cmd->advertising = pdata->phy.advertising; + cmd->lp_advertising = pdata->phy.lp_advertising; + + cmd->autoneg = pdata->phy.autoneg; + ethtool_cmd_speed_set(cmd, pdata->phy.speed); + cmd->duplex = pdata->phy.duplex; - ret = phy_ethtool_gset(pdata->phydev, cmd); + cmd->port = PORT_NONE; + cmd->transceiver = XCVR_INTERNAL; DBGPR("<--xgbe_get_settings\n"); - return ret; + return 0; } static int xgbe_set_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { struct xgbe_prv_data *pdata = netdev_priv(netdev); - struct phy_device *phydev = pdata->phydev; u32 speed; int ret; DBGPR("-->xgbe_set_settings\n"); - if (!pdata->phydev) - return -ENODEV; - speed = ethtool_cmd_speed(cmd); - if (cmd->phy_address != phydev->addr) + if (cmd->phy_address != pdata->phy.address) return -EINVAL; if ((cmd->autoneg != AUTONEG_ENABLE) && @@ -333,23 +348,23 @@ static int xgbe_set_settings(struct net_device *netdev, return -EINVAL; } - cmd->advertising &= phydev->supported; + cmd->advertising &= pdata->phy.supported; if ((cmd->autoneg == AUTONEG_ENABLE) && !cmd->advertising) return -EINVAL; ret = 0; - phydev->autoneg = cmd->autoneg; - phydev->speed = speed; - phydev->duplex = cmd->duplex; - phydev->advertising = cmd->advertising; + pdata->phy.autoneg = cmd->autoneg; + pdata->phy.speed = speed; + pdata->phy.duplex = cmd->duplex; + pdata->phy.advertising = cmd->advertising; if (cmd->autoneg == AUTONEG_ENABLE) - phydev->advertising |= ADVERTISED_Autoneg; + pdata->phy.advertising |= ADVERTISED_Autoneg; else - phydev->advertising &= ~ADVERTISED_Autoneg; + pdata->phy.advertising &= ~ADVERTISED_Autoneg; if (netif_running(netdev)) - ret = phy_start_aneg(phydev); + ret = pdata->phy_if.phy_config_aneg(pdata); DBGPR("<--xgbe_set_settings\n"); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c index 714905384900..fb7c961da49f 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-main.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c @@ -124,9 +124,11 @@ #include <linux/of.h> #include <linux/of_net.h> #include <linux/of_address.h> +#include <linux/of_platform.h> #include <linux/clk.h> #include <linux/property.h> #include <linux/acpi.h> +#include <linux/mdio.h> #include "xgbe.h" #include "xgbe-common.h" @@ -136,6 +138,49 @@ MODULE_LICENSE("Dual BSD/GPL"); MODULE_VERSION(XGBE_DRV_VERSION); MODULE_DESCRIPTION(XGBE_DRV_DESC); +static int debug = -1; +module_param(debug, int, S_IWUSR | S_IRUGO); +MODULE_PARM_DESC(debug, " Network interface message level setting"); + +static const u32 default_msg_level = (NETIF_MSG_LINK | NETIF_MSG_IFDOWN | + NETIF_MSG_IFUP); + +static const u32 xgbe_serdes_blwc[] = { + XGBE_SPEED_1000_BLWC, + XGBE_SPEED_2500_BLWC, + XGBE_SPEED_10000_BLWC, +}; + +static const u32 xgbe_serdes_cdr_rate[] = { + XGBE_SPEED_1000_CDR, + XGBE_SPEED_2500_CDR, + XGBE_SPEED_10000_CDR, +}; + +static const u32 xgbe_serdes_pq_skew[] = { + XGBE_SPEED_1000_PQ, + XGBE_SPEED_2500_PQ, + XGBE_SPEED_10000_PQ, +}; + +static const u32 xgbe_serdes_tx_amp[] = { + XGBE_SPEED_1000_TXAMP, + XGBE_SPEED_2500_TXAMP, + XGBE_SPEED_10000_TXAMP, +}; + +static const u32 xgbe_serdes_dfe_tap_cfg[] = { + XGBE_SPEED_1000_DFE_TAP_CONFIG, + XGBE_SPEED_2500_DFE_TAP_CONFIG, + XGBE_SPEED_10000_DFE_TAP_CONFIG, +}; + +static const u32 xgbe_serdes_dfe_tap_ena[] = { + XGBE_SPEED_1000_DFE_TAP_ENABLE, + XGBE_SPEED_2500_DFE_TAP_ENABLE, + XGBE_SPEED_10000_DFE_TAP_ENABLE, +}; + static void xgbe_default_config(struct xgbe_prv_data *pdata) { DBGPR("-->xgbe_default_config\n"); @@ -153,8 +198,6 @@ static void xgbe_default_config(struct xgbe_prv_data *pdata) pdata->rx_pause = 1; pdata->phy_speed = SPEED_UNKNOWN; pdata->power_down = 0; - pdata->default_autoneg = AUTONEG_ENABLE; - pdata->default_speed = SPEED_10000; DBGPR("<--xgbe_default_config\n"); } @@ -162,6 +205,7 @@ static void xgbe_default_config(struct xgbe_prv_data *pdata) static void xgbe_init_all_fptrs(struct xgbe_prv_data *pdata) { xgbe_init_function_ptrs_dev(&pdata->hw_if); + xgbe_init_function_ptrs_phy(&pdata->phy_if); xgbe_init_function_ptrs_desc(&pdata->desc_if); } @@ -248,23 +292,82 @@ static int xgbe_of_support(struct xgbe_prv_data *pdata) return 0; } + +static struct platform_device *xgbe_of_get_phy_pdev(struct xgbe_prv_data *pdata) +{ + struct device *dev = pdata->dev; + struct device_node *phy_node; + struct platform_device *phy_pdev; + + phy_node = of_parse_phandle(dev->of_node, "phy-handle", 0); + if (phy_node) { + /* Old style device tree: + * The XGBE and PHY resources are separate + */ + phy_pdev = of_find_device_by_node(phy_node); + of_node_put(phy_node); + } else { + /* New style device tree: + * The XGBE and PHY resources are grouped together with + * the PHY resources listed last + */ + get_device(dev); + phy_pdev = pdata->pdev; + } + + return phy_pdev; +} #else /* CONFIG_OF */ static int xgbe_of_support(struct xgbe_prv_data *pdata) { return -EINVAL; } -#endif /*CONFIG_OF */ + +static struct platform_device *xgbe_of_get_phy_pdev(struct xgbe_prv_data *pdata) +{ + return NULL; +} +#endif /* CONFIG_OF */ + +static unsigned int xgbe_resource_count(struct platform_device *pdev, + unsigned int type) +{ + unsigned int count; + int i; + + for (i = 0, count = 0; i < pdev->num_resources; i++) { + struct resource *res = &pdev->resource[i]; + + if (type == resource_type(res)) + count++; + } + + return count; +} + +static struct platform_device *xgbe_get_phy_pdev(struct xgbe_prv_data *pdata) +{ + struct platform_device *phy_pdev; + + if (pdata->use_acpi) { + get_device(pdata->dev); + phy_pdev = pdata->pdev; + } else { + phy_pdev = xgbe_of_get_phy_pdev(pdata); + } + + return phy_pdev; +} static int xgbe_probe(struct platform_device *pdev) { struct xgbe_prv_data *pdata; - struct xgbe_hw_if *hw_if; - struct xgbe_desc_if *desc_if; struct net_device *netdev; - struct device *dev = &pdev->dev; + struct device *dev = &pdev->dev, *phy_dev; + struct platform_device *phy_pdev; struct resource *res; const char *phy_mode; - unsigned int i; + unsigned int i, phy_memnum, phy_irqnum; int ret; DBGPR("--> xgbe_probe\n"); @@ -289,9 +392,36 @@ static int xgbe_probe(struct platform_device *pdev) mutex_init(&pdata->rss_mutex); spin_lock_init(&pdata->tstamp_lock); + pdata->msg_enable = netif_msg_init(debug, default_msg_level); + + set_bit(XGBE_DOWN, &pdata->dev_state); + /* Check if we should use ACPI or DT */ pdata->use_acpi = (!pdata->adev || acpi_disabled) ? 0 : 1; + phy_pdev = xgbe_get_phy_pdev(pdata); + if (!phy_pdev) { + dev_err(dev, "unable to obtain phy device\n"); + ret = -EINVAL; + goto err_phydev; + } + phy_dev = &phy_pdev->dev; + + if (pdev == phy_pdev) { + /* New style device tree or ACPI: + * The XGBE and PHY resources are grouped together with + * the PHY resources listed last + */ + phy_memnum = xgbe_resource_count(pdev, IORESOURCE_MEM) - 3; + phy_irqnum = xgbe_resource_count(pdev, IORESOURCE_IRQ) - 1; + } else { + /* Old style device tree: + * The XGBE and PHY resources are separate + */ + phy_memnum = 0; + phy_irqnum = 0; + } + /* Set and validate the number of descriptors for a ring */ BUILD_BUG_ON_NOT_POWER_OF_2(XGBE_TX_DESC_CNT); pdata->tx_desc_count = XGBE_TX_DESC_CNT; @@ -318,7 +448,8 @@ static int xgbe_probe(struct platform_device *pdev) ret = PTR_ERR(pdata->xgmac_regs); goto err_io; } - DBGPR(" xgmac_regs = %p\n", pdata->xgmac_regs); + if (netif_msg_probe(pdata)) + dev_dbg(dev, "xgmac_regs = %p\n", pdata->xgmac_regs); res = platform_get_resource(pdev, IORESOURCE_MEM, 1); pdata->xpcs_regs = devm_ioremap_resource(dev, res); @@ -327,7 +458,38 @@ static int xgbe_probe(struct platform_device *pdev) ret = PTR_ERR(pdata->xpcs_regs); goto err_io; } - DBGPR(" xpcs_regs = %p\n", pdata->xpcs_regs); + if (netif_msg_probe(pdata)) + dev_dbg(dev, "xpcs_regs = %p\n", pdata->xpcs_regs); + + res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++); + pdata->rxtx_regs = devm_ioremap_resource(dev, res); + if (IS_ERR(pdata->rxtx_regs)) { + dev_err(dev, "rxtx ioremap failed\n"); + ret = PTR_ERR(pdata->rxtx_regs); + goto err_io; + } + if (netif_msg_probe(pdata)) + dev_dbg(dev, "rxtx_regs = %p\n", pdata->rxtx_regs); + + res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++); + pdata->sir0_regs = devm_ioremap_resource(dev, res); + if (IS_ERR(pdata->sir0_regs)) { + dev_err(dev, "sir0 ioremap failed\n"); + ret = PTR_ERR(pdata->sir0_regs); + goto err_io; + } + if (netif_msg_probe(pdata)) + dev_dbg(dev, "sir0_regs = %p\n", pdata->sir0_regs); + + res = platform_get_resource(phy_pdev, IORESOURCE_MEM, phy_memnum++); + pdata->sir1_regs = devm_ioremap_resource(dev, res); + if (IS_ERR(pdata->sir1_regs)) { + dev_err(dev, "sir1 ioremap failed\n"); + ret = PTR_ERR(pdata->sir1_regs); + goto err_io; + } + if (netif_msg_probe(pdata)) + dev_dbg(dev, "sir1_regs = %p\n", pdata->sir1_regs); /* Retrieve the MAC address */ ret = device_property_read_u8_array(dev, XGBE_MAC_ADDR_PROPERTY, @@ -355,6 +517,115 @@ static int xgbe_probe(struct platform_device *pdev) if (device_property_present(dev, XGBE_DMA_IRQS_PROPERTY)) pdata->per_channel_irq = 1; + /* Retrieve the PHY speedset */ + ret = device_property_read_u32(phy_dev, XGBE_SPEEDSET_PROPERTY, + &pdata->speed_set); + if (ret) { + dev_err(dev, "invalid %s property\n", XGBE_SPEEDSET_PROPERTY); + goto err_io; + } + + switch (pdata->speed_set) { + case XGBE_SPEEDSET_1000_10000: + case XGBE_SPEEDSET_2500_10000: + break; + default: + dev_err(dev, "invalid %s property\n", XGBE_SPEEDSET_PROPERTY); + ret = -EINVAL; + goto err_io; + } + + /* Retrieve the PHY configuration properties */ + if (device_property_present(phy_dev, XGBE_BLWC_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_BLWC_PROPERTY, + pdata->serdes_blwc, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_BLWC_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_blwc, xgbe_serdes_blwc, + sizeof(pdata->serdes_blwc)); + } + + if (device_property_present(phy_dev, XGBE_CDR_RATE_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_CDR_RATE_PROPERTY, + pdata->serdes_cdr_rate, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_CDR_RATE_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_cdr_rate, xgbe_serdes_cdr_rate, + sizeof(pdata->serdes_cdr_rate)); + } + + if (device_property_present(phy_dev, XGBE_PQ_SKEW_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_PQ_SKEW_PROPERTY, + pdata->serdes_pq_skew, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_PQ_SKEW_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_pq_skew, xgbe_serdes_pq_skew, + sizeof(pdata->serdes_pq_skew)); + } + + if (device_property_present(phy_dev, XGBE_TX_AMP_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_TX_AMP_PROPERTY, + pdata->serdes_tx_amp, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_TX_AMP_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_tx_amp, xgbe_serdes_tx_amp, + sizeof(pdata->serdes_tx_amp)); + } + + if (device_property_present(phy_dev, XGBE_DFE_CFG_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_DFE_CFG_PROPERTY, + pdata->serdes_dfe_tap_cfg, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_DFE_CFG_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_dfe_tap_cfg, xgbe_serdes_dfe_tap_cfg, + sizeof(pdata->serdes_dfe_tap_cfg)); + } + + if (device_property_present(phy_dev, XGBE_DFE_ENA_PROPERTY)) { + ret = device_property_read_u32_array(phy_dev, + XGBE_DFE_ENA_PROPERTY, + pdata->serdes_dfe_tap_ena, + XGBE_SPEEDS); + if (ret) { + dev_err(dev, "invalid %s property\n", + XGBE_DFE_ENA_PROPERTY); + goto err_io; + } + } else { + memcpy(pdata->serdes_dfe_tap_ena, xgbe_serdes_dfe_tap_ena, + sizeof(pdata->serdes_dfe_tap_ena)); + } + /* Obtain device settings unique to ACPI/OF */ if (pdata->use_acpi) ret = xgbe_acpi_support(pdata); @@ -382,17 +653,23 @@ static int xgbe_probe(struct platform_device *pdev) } pdata->dev_irq = ret; + /* Get the auto-negotiation interrupt */ + ret = platform_get_irq(phy_pdev, phy_irqnum++); + if (ret < 0) { + dev_err(dev, "platform_get_irq phy 0 failed\n"); + goto err_io; + } + pdata->an_irq = ret; + netdev->irq = pdata->dev_irq; netdev->base_addr = (unsigned long)pdata->xgmac_regs; memcpy(netdev->dev_addr, pdata->mac_addr, netdev->addr_len); /* Set all the function pointers */ xgbe_init_all_fptrs(pdata); - hw_if = &pdata->hw_if; - desc_if = &pdata->desc_if; /* Issue software reset to device */ - hw_if->exit(pdata); + pdata->hw_if.exit(pdata); /* Populate the hardware features */ xgbe_get_all_hw_features(pdata); @@ -401,8 +678,6 @@ static int xgbe_probe(struct platform_device *pdev) xgbe_default_config(pdata); /* Set the DMA mask */ - if (!dev->dma_mask) - dev->dma_mask = &dev->coherent_dma_mask; ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(pdata->hw_feat.dma_width)); if (ret) { @@ -447,16 +722,8 @@ static int xgbe_probe(struct platform_device *pdev) XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, TCP4TE, 1); XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, UDP4TE, 1); - /* Prepare to regsiter with MDIO */ - pdata->mii_bus_id = kasprintf(GFP_KERNEL, "%s", pdev->name); - if (!pdata->mii_bus_id) { - dev_err(dev, "failed to allocate mii bus id\n"); - ret = -ENOMEM; - goto err_io; - } - ret = xgbe_mdio_register(pdata); - if (ret) - goto err_bus_id; + /* Call MDIO/PHY initialization routine */ + pdata->phy_if.phy_init(pdata); /* Set device operations */ netdev->netdev_ops = xgbe_get_netdev_ops(); @@ -501,26 +768,52 @@ static int xgbe_probe(struct platform_device *pdev) ret = register_netdev(netdev); if (ret) { dev_err(dev, "net device registration failed\n"); - goto err_reg_netdev; + goto err_io; + } + + /* Create the PHY/ANEG name based on netdev name */ + snprintf(pdata->an_name, sizeof(pdata->an_name) - 1, "%s-pcs", + netdev_name(netdev)); + + /* Create workqueues */ + pdata->dev_workqueue = + create_singlethread_workqueue(netdev_name(netdev)); + if (!pdata->dev_workqueue) { + netdev_err(netdev, "device workqueue creation failed\n"); + ret = -ENOMEM; + goto err_netdev; + } + + pdata->an_workqueue = + create_singlethread_workqueue(pdata->an_name); + if (!pdata->an_workqueue) { + netdev_err(netdev, "phy workqueue creation failed\n"); + ret = -ENOMEM; + goto err_wq; } xgbe_ptp_register(pdata); xgbe_debugfs_init(pdata); + platform_device_put(phy_pdev); + netdev_notice(netdev, "net device enabled\n"); DBGPR("<-- xgbe_probe\n"); return 0; -err_reg_netdev: - xgbe_mdio_unregister(pdata); +err_wq: + destroy_workqueue(pdata->dev_workqueue); -err_bus_id: - kfree(pdata->mii_bus_id); +err_netdev: + unregister_netdev(netdev); err_io: + platform_device_put(phy_pdev); + +err_phydev: free_netdev(netdev); err_alloc: @@ -540,11 +833,13 @@ static int xgbe_remove(struct platform_device *pdev) xgbe_ptp_unregister(pdata); - unregister_netdev(netdev); + flush_workqueue(pdata->an_workqueue); + destroy_workqueue(pdata->an_workqueue); - xgbe_mdio_unregister(pdata); + flush_workqueue(pdata->dev_workqueue); + destroy_workqueue(pdata->dev_workqueue); - kfree(pdata->mii_bus_id); + unregister_netdev(netdev); free_netdev(netdev); @@ -557,16 +852,17 @@ static int xgbe_remove(struct platform_device *pdev) static int xgbe_suspend(struct device *dev) { struct net_device *netdev = dev_get_drvdata(dev); - int ret; + struct xgbe_prv_data *pdata = netdev_priv(netdev); + int ret = 0; DBGPR("-->xgbe_suspend\n"); - if (!netif_running(netdev)) { - DBGPR("<--xgbe_dev_suspend\n"); - return -EINVAL; - } + if (netif_running(netdev)) + ret = xgbe_powerdown(netdev, XGMAC_DRIVER_CONTEXT); - ret = xgbe_powerdown(netdev, XGMAC_DRIVER_CONTEXT); + pdata->lpm_ctrl = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + pdata->lpm_ctrl |= MDIO_CTRL1_LPOWER; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, pdata->lpm_ctrl); DBGPR("<--xgbe_suspend\n"); @@ -576,16 +872,16 @@ static int xgbe_suspend(struct device *dev) static int xgbe_resume(struct device *dev) { struct net_device *netdev = dev_get_drvdata(dev); - int ret; + struct xgbe_prv_data *pdata = netdev_priv(netdev); + int ret = 0; DBGPR("-->xgbe_resume\n"); - if (!netif_running(netdev)) { - DBGPR("<--xgbe_dev_resume\n"); - return -EINVAL; - } + pdata->lpm_ctrl &= ~MDIO_CTRL1_LPOWER; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, pdata->lpm_ctrl); - ret = xgbe_powerup(netdev, XGMAC_DRIVER_CONTEXT); + if (netif_running(netdev)) + ret = xgbe_powerup(netdev, XGMAC_DRIVER_CONTEXT); DBGPR("<--xgbe_resume\n"); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c index 59e267f3f1b7..9088c3a35a20 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c @@ -119,194 +119,1246 @@ #include <linux/mdio.h> #include <linux/phy.h> #include <linux/of.h> +#include <linux/bitops.h> +#include <linux/jiffies.h> #include "xgbe.h" #include "xgbe-common.h" -static int xgbe_mdio_read(struct mii_bus *mii, int prtad, int mmd_reg) +static void xgbe_an_enable_kr_training(struct xgbe_prv_data *pdata) { - struct xgbe_prv_data *pdata = mii->priv; - struct xgbe_hw_if *hw_if = &pdata->hw_if; - int mmd_data; + unsigned int reg; - DBGPR_MDIO("-->xgbe_mdio_read: prtad=%#x mmd_reg=%#x\n", - prtad, mmd_reg); + reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL); - mmd_data = hw_if->read_mmd_regs(pdata, prtad, mmd_reg); + reg |= XGBE_KR_TRAINING_ENABLE; + XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg); +} + +static void xgbe_an_disable_kr_training(struct xgbe_prv_data *pdata) +{ + unsigned int reg; - DBGPR_MDIO("<--xgbe_mdio_read: mmd_data=%#x\n", mmd_data); + reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL); - return mmd_data; + reg &= ~XGBE_KR_TRAINING_ENABLE; + XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, reg); } -static int xgbe_mdio_write(struct mii_bus *mii, int prtad, int mmd_reg, - u16 mmd_val) +static void xgbe_pcs_power_cycle(struct xgbe_prv_data *pdata) { - struct xgbe_prv_data *pdata = mii->priv; - struct xgbe_hw_if *hw_if = &pdata->hw_if; - int mmd_data = mmd_val; + unsigned int reg; - DBGPR_MDIO("-->xgbe_mdio_write: prtad=%#x mmd_reg=%#x mmd_data=%#x\n", - prtad, mmd_reg, mmd_data); + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); - hw_if->write_mmd_regs(pdata, prtad, mmd_reg, mmd_data); + reg |= MDIO_CTRL1_LPOWER; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); - DBGPR_MDIO("<--xgbe_mdio_write\n"); + usleep_range(75, 100); - return 0; + reg &= ~MDIO_CTRL1_LPOWER; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); } -void xgbe_dump_phy_registers(struct xgbe_prv_data *pdata) +static void xgbe_serdes_start_ratechange(struct xgbe_prv_data *pdata) { - struct device *dev = pdata->dev; - struct phy_device *phydev = pdata->mii->phy_map[XGBE_PRTAD]; - int i; - - dev_alert(dev, "\n************* PHY Reg dump **********************\n"); - - dev_alert(dev, "PCS Control Reg (%#04x) = %#04x\n", MDIO_CTRL1, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1)); - dev_alert(dev, "PCS Status Reg (%#04x) = %#04x\n", MDIO_STAT1, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1)); - dev_alert(dev, "Phy Id (PHYS ID 1 %#04x)= %#04x\n", MDIO_DEVID1, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID1)); - dev_alert(dev, "Phy Id (PHYS ID 2 %#04x)= %#04x\n", MDIO_DEVID2, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID2)); - dev_alert(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS1, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS1)); - dev_alert(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS2, - XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS2)); - - dev_alert(dev, "Auto-Neg Control Reg (%#04x) = %#04x\n", MDIO_CTRL1, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1)); - dev_alert(dev, "Auto-Neg Status Reg (%#04x) = %#04x\n", MDIO_STAT1, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_STAT1)); - dev_alert(dev, "Auto-Neg Ad Reg 1 (%#04x) = %#04x\n", - MDIO_AN_ADVERTISE, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE)); - dev_alert(dev, "Auto-Neg Ad Reg 2 (%#04x) = %#04x\n", - MDIO_AN_ADVERTISE + 1, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1)); - dev_alert(dev, "Auto-Neg Ad Reg 3 (%#04x) = %#04x\n", - MDIO_AN_ADVERTISE + 2, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2)); - dev_alert(dev, "Auto-Neg Completion Reg (%#04x) = %#04x\n", - MDIO_AN_COMP_STAT, - XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_COMP_STAT)); - - dev_alert(dev, "MMD Device Mask = %#x\n", - phydev->c45_ids.devices_in_package); - for (i = 0; i < ARRAY_SIZE(phydev->c45_ids.device_ids); i++) - dev_alert(dev, " MMD %d: ID = %#08x\n", i, - phydev->c45_ids.device_ids[i]); - - dev_alert(dev, "\n*************************************************\n"); -} - -int xgbe_mdio_register(struct xgbe_prv_data *pdata) -{ - struct mii_bus *mii; - struct phy_device *phydev; - int ret = 0; - - DBGPR("-->xgbe_mdio_register\n"); - - mii = mdiobus_alloc(); - if (!mii) { - dev_err(pdata->dev, "mdiobus_alloc failed\n"); - return -ENOMEM; + /* Assert Rx and Tx ratechange */ + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, RATECHANGE, 1); +} + +static void xgbe_serdes_complete_ratechange(struct xgbe_prv_data *pdata) +{ + unsigned int wait; + u16 status; + + /* Release Rx and Tx ratechange */ + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, RATECHANGE, 0); + + /* Wait for Rx and Tx ready */ + wait = XGBE_RATECHANGE_COUNT; + while (wait--) { + usleep_range(50, 75); + + status = XSIR0_IOREAD(pdata, SIR0_STATUS); + if (XSIR_GET_BITS(status, SIR0_STATUS, RX_READY) && + XSIR_GET_BITS(status, SIR0_STATUS, TX_READY)) + goto rx_reset; } - /* Register on the MDIO bus (don't probe any PHYs) */ - mii->name = XGBE_PHY_NAME; - mii->read = xgbe_mdio_read; - mii->write = xgbe_mdio_write; - snprintf(mii->id, sizeof(mii->id), "%s", pdata->mii_bus_id); - mii->priv = pdata; - mii->phy_mask = ~0; - mii->parent = pdata->dev; - ret = mdiobus_register(mii); - if (ret) { - dev_err(pdata->dev, "mdiobus_register failed\n"); - goto err_mdiobus_alloc; + netif_dbg(pdata, link, pdata->netdev, "SerDes rx/tx not ready (%#hx)\n", + status); + +rx_reset: + /* Perform Rx reset for the DFE changes */ + XRXTX_IOWRITE_BITS(pdata, RXTX_REG6, RESETB_RXD, 0); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG6, RESETB_RXD, 1); +} + +static void xgbe_xgmii_mode(struct xgbe_prv_data *pdata) +{ + unsigned int reg; + + /* Enable KR training */ + xgbe_an_enable_kr_training(pdata); + + /* Set MAC to 10G speed */ + pdata->hw_if.set_xgmii_speed(pdata); + + /* Set PCS to KR/10G speed */ + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL2); + reg &= ~MDIO_PCS_CTRL2_TYPE; + reg |= MDIO_PCS_CTRL2_10GBR; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL2, reg); + + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + reg &= ~MDIO_CTRL1_SPEEDSEL; + reg |= MDIO_CTRL1_SPEED10G; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); + + xgbe_pcs_power_cycle(pdata); + + /* Set SerDes to 10G speed */ + xgbe_serdes_start_ratechange(pdata); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, DATARATE, XGBE_SPEED_10000_RATE); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, WORDMODE, XGBE_SPEED_10000_WORD); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, PLLSEL, XGBE_SPEED_10000_PLL); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, CDR_RATE, + pdata->serdes_cdr_rate[XGBE_SPEED_10000]); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, TXAMP, + pdata->serdes_tx_amp[XGBE_SPEED_10000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG20, BLWC_ENA, + pdata->serdes_blwc[XGBE_SPEED_10000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG114, PQ_REG, + pdata->serdes_pq_skew[XGBE_SPEED_10000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG129, RXDFE_CONFIG, + pdata->serdes_dfe_tap_cfg[XGBE_SPEED_10000]); + XRXTX_IOWRITE(pdata, RXTX_REG22, + pdata->serdes_dfe_tap_ena[XGBE_SPEED_10000]); + + xgbe_serdes_complete_ratechange(pdata); + + netif_dbg(pdata, link, pdata->netdev, "10GbE KR mode set\n"); +} + +static void xgbe_gmii_2500_mode(struct xgbe_prv_data *pdata) +{ + unsigned int reg; + + /* Disable KR training */ + xgbe_an_disable_kr_training(pdata); + + /* Set MAC to 2.5G speed */ + pdata->hw_if.set_gmii_2500_speed(pdata); + + /* Set PCS to KX/1G speed */ + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL2); + reg &= ~MDIO_PCS_CTRL2_TYPE; + reg |= MDIO_PCS_CTRL2_10GBX; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL2, reg); + + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + reg &= ~MDIO_CTRL1_SPEEDSEL; + reg |= MDIO_CTRL1_SPEED1G; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); + + xgbe_pcs_power_cycle(pdata); + + /* Set SerDes to 2.5G speed */ + xgbe_serdes_start_ratechange(pdata); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, DATARATE, XGBE_SPEED_2500_RATE); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, WORDMODE, XGBE_SPEED_2500_WORD); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, PLLSEL, XGBE_SPEED_2500_PLL); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, CDR_RATE, + pdata->serdes_cdr_rate[XGBE_SPEED_2500]); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, TXAMP, + pdata->serdes_tx_amp[XGBE_SPEED_2500]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG20, BLWC_ENA, + pdata->serdes_blwc[XGBE_SPEED_2500]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG114, PQ_REG, + pdata->serdes_pq_skew[XGBE_SPEED_2500]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG129, RXDFE_CONFIG, + pdata->serdes_dfe_tap_cfg[XGBE_SPEED_2500]); + XRXTX_IOWRITE(pdata, RXTX_REG22, + pdata->serdes_dfe_tap_ena[XGBE_SPEED_2500]); + + xgbe_serdes_complete_ratechange(pdata); + + netif_dbg(pdata, link, pdata->netdev, "2.5GbE KX mode set\n"); +} + +static void xgbe_gmii_mode(struct xgbe_prv_data *pdata) +{ + unsigned int reg; + + /* Disable KR training */ + xgbe_an_disable_kr_training(pdata); + + /* Set MAC to 1G speed */ + pdata->hw_if.set_gmii_speed(pdata); + + /* Set PCS to KX/1G speed */ + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL2); + reg &= ~MDIO_PCS_CTRL2_TYPE; + reg |= MDIO_PCS_CTRL2_10GBX; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL2, reg); + + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + reg &= ~MDIO_CTRL1_SPEEDSEL; + reg |= MDIO_CTRL1_SPEED1G; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); + + xgbe_pcs_power_cycle(pdata); + + /* Set SerDes to 1G speed */ + xgbe_serdes_start_ratechange(pdata); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, DATARATE, XGBE_SPEED_1000_RATE); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, WORDMODE, XGBE_SPEED_1000_WORD); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, PLLSEL, XGBE_SPEED_1000_PLL); + + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, CDR_RATE, + pdata->serdes_cdr_rate[XGBE_SPEED_1000]); + XSIR1_IOWRITE_BITS(pdata, SIR1_SPEED, TXAMP, + pdata->serdes_tx_amp[XGBE_SPEED_1000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG20, BLWC_ENA, + pdata->serdes_blwc[XGBE_SPEED_1000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG114, PQ_REG, + pdata->serdes_pq_skew[XGBE_SPEED_1000]); + XRXTX_IOWRITE_BITS(pdata, RXTX_REG129, RXDFE_CONFIG, + pdata->serdes_dfe_tap_cfg[XGBE_SPEED_1000]); + XRXTX_IOWRITE(pdata, RXTX_REG22, + pdata->serdes_dfe_tap_ena[XGBE_SPEED_1000]); + + xgbe_serdes_complete_ratechange(pdata); + + netif_dbg(pdata, link, pdata->netdev, "1GbE KX mode set\n"); +} + +static void xgbe_cur_mode(struct xgbe_prv_data *pdata, + enum xgbe_mode *mode) +{ + unsigned int reg; + + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL2); + if ((reg & MDIO_PCS_CTRL2_TYPE) == MDIO_PCS_CTRL2_10GBR) + *mode = XGBE_MODE_KR; + else + *mode = XGBE_MODE_KX; +} + +static bool xgbe_in_kr_mode(struct xgbe_prv_data *pdata) +{ + enum xgbe_mode mode; + + xgbe_cur_mode(pdata, &mode); + + return (mode == XGBE_MODE_KR); +} + +static void xgbe_switch_mode(struct xgbe_prv_data *pdata) +{ + /* If we are in KR switch to KX, and vice-versa */ + if (xgbe_in_kr_mode(pdata)) { + if (pdata->speed_set == XGBE_SPEEDSET_1000_10000) + xgbe_gmii_mode(pdata); + else + xgbe_gmii_2500_mode(pdata); + } else { + xgbe_xgmii_mode(pdata); } - DBGPR(" mdiobus_register succeeded for %s\n", pdata->mii_bus_id); - - /* Probe the PCS using Clause 45 */ - phydev = get_phy_device(mii, XGBE_PRTAD, true); - if (IS_ERR(phydev) || !phydev || - !phydev->c45_ids.device_ids[MDIO_MMD_PCS]) { - dev_err(pdata->dev, "get_phy_device failed\n"); - ret = phydev ? PTR_ERR(phydev) : -ENOLINK; - goto err_mdiobus_register; +} + +static void xgbe_set_mode(struct xgbe_prv_data *pdata, + enum xgbe_mode mode) +{ + enum xgbe_mode cur_mode; + + xgbe_cur_mode(pdata, &cur_mode); + if (mode != cur_mode) + xgbe_switch_mode(pdata); +} + +static bool xgbe_use_xgmii_mode(struct xgbe_prv_data *pdata) +{ + if (pdata->phy.autoneg == AUTONEG_ENABLE) { + if (pdata->phy.advertising & ADVERTISED_10000baseKR_Full) + return true; + } else { + if (pdata->phy.speed == SPEED_10000) + return true; } - request_module(MDIO_MODULE_PREFIX MDIO_ID_FMT, - MDIO_ID_ARGS(phydev->c45_ids.device_ids[MDIO_MMD_PCS])); - ret = phy_device_register(phydev); - if (ret) { - dev_err(pdata->dev, "phy_device_register failed\n"); - goto err_phy_device; + return false; +} + +static bool xgbe_use_gmii_2500_mode(struct xgbe_prv_data *pdata) +{ + if (pdata->phy.autoneg == AUTONEG_ENABLE) { + if (pdata->phy.advertising & ADVERTISED_2500baseX_Full) + return true; + } else { + if (pdata->phy.speed == SPEED_2500) + return true; } - if (!phydev->dev.driver) { - dev_err(pdata->dev, "phy driver probe failed\n"); - ret = -EIO; - goto err_phy_device; + + return false; +} + +static bool xgbe_use_gmii_mode(struct xgbe_prv_data *pdata) +{ + if (pdata->phy.autoneg == AUTONEG_ENABLE) { + if (pdata->phy.advertising & ADVERTISED_1000baseKX_Full) + return true; + } else { + if (pdata->phy.speed == SPEED_1000) + return true; } - /* Add a reference to the PHY driver so it can't be unloaded */ - pdata->phy_module = phydev->dev.driver->owner; - if (!try_module_get(pdata->phy_module)) { - dev_err(pdata->dev, "try_module_get failed\n"); - ret = -EIO; - goto err_phy_device; + return false; +} + +static void xgbe_set_an(struct xgbe_prv_data *pdata, bool enable, bool restart) +{ + unsigned int reg; + + reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1); + reg &= ~MDIO_AN_CTRL1_ENABLE; + + if (enable) + reg |= MDIO_AN_CTRL1_ENABLE; + + if (restart) + reg |= MDIO_AN_CTRL1_RESTART; + + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_CTRL1, reg); +} + +static void xgbe_restart_an(struct xgbe_prv_data *pdata) +{ + xgbe_set_an(pdata, true, true); + + netif_dbg(pdata, link, pdata->netdev, "AN enabled/restarted\n"); +} + +static void xgbe_disable_an(struct xgbe_prv_data *pdata) +{ + xgbe_set_an(pdata, false, false); + + netif_dbg(pdata, link, pdata->netdev, "AN disabled\n"); +} + +static enum xgbe_an xgbe_an_tx_training(struct xgbe_prv_data *pdata, + enum xgbe_rx *state) +{ + unsigned int ad_reg, lp_reg, reg; + + *state = XGBE_RX_COMPLETE; + + /* If we're not in KR mode then we're done */ + if (!xgbe_in_kr_mode(pdata)) + return XGBE_AN_PAGE_RECEIVED; + + /* Enable/Disable FEC */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 2); + + reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_FECCTRL); + reg &= ~(MDIO_PMA_10GBR_FECABLE_ABLE | MDIO_PMA_10GBR_FECABLE_ERRABLE); + if ((ad_reg & 0xc000) && (lp_reg & 0xc000)) + reg |= pdata->fec_ability; + + XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_FECCTRL, reg); + + /* Start KR training */ + reg = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL); + if (reg & XGBE_KR_TRAINING_ENABLE) { + XSIR0_IOWRITE_BITS(pdata, SIR0_KR_RT_1, RESET, 1); + + reg |= XGBE_KR_TRAINING_START; + XMDIO_WRITE(pdata, MDIO_MMD_PMAPMD, MDIO_PMA_10GBR_PMD_CTRL, + reg); + + XSIR0_IOWRITE_BITS(pdata, SIR0_KR_RT_1, RESET, 0); + + netif_dbg(pdata, link, pdata->netdev, + "KR training initiated\n"); } - pdata->mii = mii; - pdata->mdio_mmd = MDIO_MMD_PCS; + return XGBE_AN_PAGE_RECEIVED; +} + +static enum xgbe_an xgbe_an_tx_xnp(struct xgbe_prv_data *pdata, + enum xgbe_rx *state) +{ + u16 msg; + + *state = XGBE_RX_XNP; + + msg = XGBE_XNP_MCF_NULL_MESSAGE; + msg |= XGBE_XNP_MP_FORMATTED; + + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_XNP + 2, 0); + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_XNP + 1, 0); + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_XNP, msg); + + return XGBE_AN_PAGE_RECEIVED; +} + +static enum xgbe_an xgbe_an_rx_bpa(struct xgbe_prv_data *pdata, + enum xgbe_rx *state) +{ + unsigned int link_support; + unsigned int reg, ad_reg, lp_reg; + + /* Read Base Ability register 2 first */ + reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 1); + + /* Check for a supported mode, otherwise restart in a different one */ + link_support = xgbe_in_kr_mode(pdata) ? 0x80 : 0x20; + if (!(reg & link_support)) + return XGBE_AN_INCOMPAT_LINK; + + /* Check Extended Next Page support */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA); + + return ((ad_reg & XGBE_XNP_NP_EXCHANGE) || + (lp_reg & XGBE_XNP_NP_EXCHANGE)) + ? xgbe_an_tx_xnp(pdata, state) + : xgbe_an_tx_training(pdata, state); +} + +static enum xgbe_an xgbe_an_rx_xnp(struct xgbe_prv_data *pdata, + enum xgbe_rx *state) +{ + unsigned int ad_reg, lp_reg; + + /* Check Extended Next Page support */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_XNP); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPX); + + return ((ad_reg & XGBE_XNP_NP_EXCHANGE) || + (lp_reg & XGBE_XNP_NP_EXCHANGE)) + ? xgbe_an_tx_xnp(pdata, state) + : xgbe_an_tx_training(pdata, state); +} + +static enum xgbe_an xgbe_an_page_received(struct xgbe_prv_data *pdata) +{ + enum xgbe_rx *state; + unsigned long an_timeout; + enum xgbe_an ret; + + if (!pdata->an_start) { + pdata->an_start = jiffies; + } else { + an_timeout = pdata->an_start + + msecs_to_jiffies(XGBE_AN_MS_TIMEOUT); + if (time_after(jiffies, an_timeout)) { + /* Auto-negotiation timed out, reset state */ + pdata->kr_state = XGBE_RX_BPA; + pdata->kx_state = XGBE_RX_BPA; + + pdata->an_start = jiffies; + + netif_dbg(pdata, link, pdata->netdev, + "AN timed out, resetting state\n"); + } + } - phydev->autoneg = pdata->default_autoneg; - if (phydev->autoneg == AUTONEG_DISABLE) { - phydev->speed = pdata->default_speed; - phydev->duplex = DUPLEX_FULL; + state = xgbe_in_kr_mode(pdata) ? &pdata->kr_state + : &pdata->kx_state; - phydev->advertising &= ~ADVERTISED_Autoneg; + switch (*state) { + case XGBE_RX_BPA: + ret = xgbe_an_rx_bpa(pdata, state); + break; + + case XGBE_RX_XNP: + ret = xgbe_an_rx_xnp(pdata, state); + break; + + default: + ret = XGBE_AN_ERROR; + } + + return ret; +} + +static enum xgbe_an xgbe_an_incompat_link(struct xgbe_prv_data *pdata) +{ + /* Be sure we aren't looping trying to negotiate */ + if (xgbe_in_kr_mode(pdata)) { + pdata->kr_state = XGBE_RX_ERROR; + + if (!(pdata->phy.advertising & ADVERTISED_1000baseKX_Full) && + !(pdata->phy.advertising & ADVERTISED_2500baseX_Full)) + return XGBE_AN_NO_LINK; + + if (pdata->kx_state != XGBE_RX_BPA) + return XGBE_AN_NO_LINK; + } else { + pdata->kx_state = XGBE_RX_ERROR; + + if (!(pdata->phy.advertising & ADVERTISED_10000baseKR_Full)) + return XGBE_AN_NO_LINK; + + if (pdata->kr_state != XGBE_RX_BPA) + return XGBE_AN_NO_LINK; + } + + xgbe_disable_an(pdata); + + xgbe_switch_mode(pdata); + + xgbe_restart_an(pdata); + + return XGBE_AN_INCOMPAT_LINK; +} + +static irqreturn_t xgbe_an_isr(int irq, void *data) +{ + struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + + netif_dbg(pdata, intr, pdata->netdev, "AN interrupt received\n"); + + /* Interrupt reason must be read and cleared outside of IRQ context */ + disable_irq_nosync(pdata->an_irq); + + queue_work(pdata->an_workqueue, &pdata->an_irq_work); + + return IRQ_HANDLED; +} + +static void xgbe_an_irq_work(struct work_struct *work) +{ + struct xgbe_prv_data *pdata = container_of(work, + struct xgbe_prv_data, + an_irq_work); + + /* Avoid a race between enabling the IRQ and exiting the work by + * waiting for the work to finish and then queueing it + */ + flush_work(&pdata->an_work); + queue_work(pdata->an_workqueue, &pdata->an_work); +} + +static const char *xgbe_state_as_string(enum xgbe_an state) +{ + switch (state) { + case XGBE_AN_READY: + return "Ready"; + case XGBE_AN_PAGE_RECEIVED: + return "Page-Received"; + case XGBE_AN_INCOMPAT_LINK: + return "Incompatible-Link"; + case XGBE_AN_COMPLETE: + return "Complete"; + case XGBE_AN_NO_LINK: + return "No-Link"; + case XGBE_AN_ERROR: + return "Error"; + default: + return "Undefined"; + } +} + +static void xgbe_an_state_machine(struct work_struct *work) +{ + struct xgbe_prv_data *pdata = container_of(work, + struct xgbe_prv_data, + an_work); + enum xgbe_an cur_state = pdata->an_state; + unsigned int int_reg, int_mask; + + mutex_lock(&pdata->an_mutex); + + /* Read the interrupt */ + int_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_INT); + if (!int_reg) + goto out; + +next_int: + if (int_reg & XGBE_AN_PG_RCV) { + pdata->an_state = XGBE_AN_PAGE_RECEIVED; + int_mask = XGBE_AN_PG_RCV; + } else if (int_reg & XGBE_AN_INC_LINK) { + pdata->an_state = XGBE_AN_INCOMPAT_LINK; + int_mask = XGBE_AN_INC_LINK; + } else if (int_reg & XGBE_AN_INT_CMPLT) { + pdata->an_state = XGBE_AN_COMPLETE; + int_mask = XGBE_AN_INT_CMPLT; + } else { + pdata->an_state = XGBE_AN_ERROR; + int_mask = 0; } - pdata->phydev = phydev; + /* Clear the interrupt to be processed */ + int_reg &= ~int_mask; + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, int_reg); + + pdata->an_result = pdata->an_state; + +again: + netif_dbg(pdata, link, pdata->netdev, "AN %s\n", + xgbe_state_as_string(pdata->an_state)); + + cur_state = pdata->an_state; + + switch (pdata->an_state) { + case XGBE_AN_READY: + pdata->an_supported = 0; + break; + + case XGBE_AN_PAGE_RECEIVED: + pdata->an_state = xgbe_an_page_received(pdata); + pdata->an_supported++; + break; + + case XGBE_AN_INCOMPAT_LINK: + pdata->an_supported = 0; + pdata->parallel_detect = 0; + pdata->an_state = xgbe_an_incompat_link(pdata); + break; - DBGPHY_REGS(pdata); + case XGBE_AN_COMPLETE: + pdata->parallel_detect = pdata->an_supported ? 0 : 1; + netif_dbg(pdata, link, pdata->netdev, "%s successful\n", + pdata->an_supported ? "Auto negotiation" + : "Parallel detection"); + break; - DBGPR("<--xgbe_mdio_register\n"); + case XGBE_AN_NO_LINK: + break; + + default: + pdata->an_state = XGBE_AN_ERROR; + } + + if (pdata->an_state == XGBE_AN_NO_LINK) { + int_reg = 0; + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0); + } else if (pdata->an_state == XGBE_AN_ERROR) { + netdev_err(pdata->netdev, + "error during auto-negotiation, state=%u\n", + cur_state); + + int_reg = 0; + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0); + } + + if (pdata->an_state >= XGBE_AN_COMPLETE) { + pdata->an_result = pdata->an_state; + pdata->an_state = XGBE_AN_READY; + pdata->kr_state = XGBE_RX_BPA; + pdata->kx_state = XGBE_RX_BPA; + pdata->an_start = 0; + + netif_dbg(pdata, link, pdata->netdev, "AN result: %s\n", + xgbe_state_as_string(pdata->an_result)); + } + + if (cur_state != pdata->an_state) + goto again; + + if (int_reg) + goto next_int; + +out: + enable_irq(pdata->an_irq); + + mutex_unlock(&pdata->an_mutex); +} + +static void xgbe_an_init(struct xgbe_prv_data *pdata) +{ + unsigned int reg; + + /* Set up Advertisement register 3 first */ + reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2); + if (pdata->phy.advertising & ADVERTISED_10000baseR_FEC) + reg |= 0xc000; + else + reg &= ~0xc000; + + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2, reg); + + /* Set up Advertisement register 2 next */ + reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1); + if (pdata->phy.advertising & ADVERTISED_10000baseKR_Full) + reg |= 0x80; + else + reg &= ~0x80; + + if ((pdata->phy.advertising & ADVERTISED_1000baseKX_Full) || + (pdata->phy.advertising & ADVERTISED_2500baseX_Full)) + reg |= 0x20; + else + reg &= ~0x20; + + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1, reg); + + /* Set up Advertisement register 1 last */ + reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE); + if (pdata->phy.advertising & ADVERTISED_Pause) + reg |= 0x400; + else + reg &= ~0x400; + + if (pdata->phy.advertising & ADVERTISED_Asym_Pause) + reg |= 0x800; + else + reg &= ~0x800; + + /* We don't intend to perform XNP */ + reg &= ~XGBE_XNP_NP_EXCHANGE; + + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE, reg); + + netif_dbg(pdata, link, pdata->netdev, "AN initialized\n"); +} + +static const char *xgbe_phy_fc_string(struct xgbe_prv_data *pdata) +{ + if (pdata->tx_pause && pdata->rx_pause) + return "rx/tx"; + else if (pdata->rx_pause) + return "rx"; + else if (pdata->tx_pause) + return "tx"; + else + return "off"; +} + +static const char *xgbe_phy_speed_string(int speed) +{ + switch (speed) { + case SPEED_1000: + return "1Gbps"; + case SPEED_2500: + return "2.5Gbps"; + case SPEED_10000: + return "10Gbps"; + case SPEED_UNKNOWN: + return "Unknown"; + default: + return "Unsupported"; + } +} + +static void xgbe_phy_print_status(struct xgbe_prv_data *pdata) +{ + if (pdata->phy.link) + netdev_info(pdata->netdev, + "Link is Up - %s/%s - flow control %s\n", + xgbe_phy_speed_string(pdata->phy.speed), + pdata->phy.duplex == DUPLEX_FULL ? "Full" : "Half", + xgbe_phy_fc_string(pdata)); + else + netdev_info(pdata->netdev, "Link is Down\n"); +} + +static void xgbe_phy_adjust_link(struct xgbe_prv_data *pdata) +{ + int new_state = 0; + + if (pdata->phy.link) { + /* Flow control support */ + pdata->pause_autoneg = pdata->phy.pause_autoneg; + + if (pdata->tx_pause != pdata->phy.tx_pause) { + new_state = 1; + pdata->hw_if.config_tx_flow_control(pdata); + pdata->tx_pause = pdata->phy.tx_pause; + } + + if (pdata->rx_pause != pdata->phy.rx_pause) { + new_state = 1; + pdata->hw_if.config_rx_flow_control(pdata); + pdata->rx_pause = pdata->phy.rx_pause; + } + + /* Speed support */ + if (pdata->phy_speed != pdata->phy.speed) { + new_state = 1; + pdata->phy_speed = pdata->phy.speed; + } + + if (pdata->phy_link != pdata->phy.link) { + new_state = 1; + pdata->phy_link = pdata->phy.link; + } + } else if (pdata->phy_link) { + new_state = 1; + pdata->phy_link = 0; + pdata->phy_speed = SPEED_UNKNOWN; + } + + if (new_state && netif_msg_link(pdata)) + xgbe_phy_print_status(pdata); +} + +static int xgbe_phy_config_fixed(struct xgbe_prv_data *pdata) +{ + netif_dbg(pdata, link, pdata->netdev, "fixed PHY configuration\n"); + + /* Disable auto-negotiation */ + xgbe_disable_an(pdata); + + /* Validate/Set specified speed */ + switch (pdata->phy.speed) { + case SPEED_10000: + xgbe_set_mode(pdata, XGBE_MODE_KR); + break; + + case SPEED_2500: + case SPEED_1000: + xgbe_set_mode(pdata, XGBE_MODE_KX); + break; + + default: + return -EINVAL; + } + + /* Validate duplex mode */ + if (pdata->phy.duplex != DUPLEX_FULL) + return -EINVAL; + + return 0; +} + +static int __xgbe_phy_config_aneg(struct xgbe_prv_data *pdata) +{ + set_bit(XGBE_LINK_INIT, &pdata->dev_state); + pdata->link_check = jiffies; + + if (pdata->phy.autoneg != AUTONEG_ENABLE) + return xgbe_phy_config_fixed(pdata); + + netif_dbg(pdata, link, pdata->netdev, "AN PHY configuration\n"); + + /* Disable auto-negotiation interrupt */ + disable_irq(pdata->an_irq); + + /* Start auto-negotiation in a supported mode */ + if (pdata->phy.advertising & ADVERTISED_10000baseKR_Full) { + xgbe_set_mode(pdata, XGBE_MODE_KR); + } else if ((pdata->phy.advertising & ADVERTISED_1000baseKX_Full) || + (pdata->phy.advertising & ADVERTISED_2500baseX_Full)) { + xgbe_set_mode(pdata, XGBE_MODE_KX); + } else { + enable_irq(pdata->an_irq); + return -EINVAL; + } + + /* Disable and stop any in progress auto-negotiation */ + xgbe_disable_an(pdata); + + /* Clear any auto-negotitation interrupts */ + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0); + + pdata->an_result = XGBE_AN_READY; + pdata->an_state = XGBE_AN_READY; + pdata->kr_state = XGBE_RX_BPA; + pdata->kx_state = XGBE_RX_BPA; + + /* Re-enable auto-negotiation interrupt */ + enable_irq(pdata->an_irq); + + /* Set up advertisement registers based on current settings */ + xgbe_an_init(pdata); + + /* Enable and start auto-negotiation */ + xgbe_restart_an(pdata); return 0; +} + +static int xgbe_phy_config_aneg(struct xgbe_prv_data *pdata) +{ + int ret; + + mutex_lock(&pdata->an_mutex); + + ret = __xgbe_phy_config_aneg(pdata); + if (ret) + set_bit(XGBE_LINK_ERR, &pdata->dev_state); + else + clear_bit(XGBE_LINK_ERR, &pdata->dev_state); + + mutex_unlock(&pdata->an_mutex); + + return ret; +} + +static bool xgbe_phy_aneg_done(struct xgbe_prv_data *pdata) +{ + return (pdata->an_result == XGBE_AN_COMPLETE); +} + +static void xgbe_check_link_timeout(struct xgbe_prv_data *pdata) +{ + unsigned long link_timeout; + + link_timeout = pdata->link_check + (XGBE_LINK_TIMEOUT * HZ); + if (time_after(jiffies, link_timeout)) { + netif_dbg(pdata, link, pdata->netdev, "AN link timeout\n"); + xgbe_phy_config_aneg(pdata); + } +} + +static void xgbe_phy_status_force(struct xgbe_prv_data *pdata) +{ + if (xgbe_in_kr_mode(pdata)) { + pdata->phy.speed = SPEED_10000; + } else { + switch (pdata->speed_set) { + case XGBE_SPEEDSET_1000_10000: + pdata->phy.speed = SPEED_1000; + break; + + case XGBE_SPEEDSET_2500_10000: + pdata->phy.speed = SPEED_2500; + break; + } + } + pdata->phy.duplex = DUPLEX_FULL; +} + +static void xgbe_phy_status_aneg(struct xgbe_prv_data *pdata) +{ + unsigned int ad_reg, lp_reg; + + pdata->phy.lp_advertising = 0; + + if ((pdata->phy.autoneg != AUTONEG_ENABLE) || pdata->parallel_detect) + return xgbe_phy_status_force(pdata); + + pdata->phy.lp_advertising |= ADVERTISED_Autoneg; + pdata->phy.lp_advertising |= ADVERTISED_Backplane; + + /* Compare Advertisement and Link Partner register 1 */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA); + if (lp_reg & 0x400) + pdata->phy.lp_advertising |= ADVERTISED_Pause; + if (lp_reg & 0x800) + pdata->phy.lp_advertising |= ADVERTISED_Asym_Pause; + + if (pdata->phy.pause_autoneg) { + /* Set flow control based on auto-negotiation result */ + pdata->phy.tx_pause = 0; + pdata->phy.rx_pause = 0; + + if (ad_reg & lp_reg & 0x400) { + pdata->phy.tx_pause = 1; + pdata->phy.rx_pause = 1; + } else if (ad_reg & lp_reg & 0x800) { + if (ad_reg & 0x400) + pdata->phy.rx_pause = 1; + else if (lp_reg & 0x400) + pdata->phy.tx_pause = 1; + } + } + + /* Compare Advertisement and Link Partner register 2 */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 1); + if (lp_reg & 0x80) + pdata->phy.lp_advertising |= ADVERTISED_10000baseKR_Full; + if (lp_reg & 0x20) { + switch (pdata->speed_set) { + case XGBE_SPEEDSET_1000_10000: + pdata->phy.lp_advertising |= ADVERTISED_1000baseKX_Full; + break; + case XGBE_SPEEDSET_2500_10000: + pdata->phy.lp_advertising |= ADVERTISED_2500baseX_Full; + break; + } + } + + ad_reg &= lp_reg; + if (ad_reg & 0x80) { + pdata->phy.speed = SPEED_10000; + xgbe_set_mode(pdata, XGBE_MODE_KR); + } else if (ad_reg & 0x20) { + switch (pdata->speed_set) { + case XGBE_SPEEDSET_1000_10000: + pdata->phy.speed = SPEED_1000; + break; + + case XGBE_SPEEDSET_2500_10000: + pdata->phy.speed = SPEED_2500; + break; + } + + xgbe_set_mode(pdata, XGBE_MODE_KX); + } else { + pdata->phy.speed = SPEED_UNKNOWN; + } + + /* Compare Advertisement and Link Partner register 3 */ + ad_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2); + lp_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_LPA + 2); + if (lp_reg & 0xc000) + pdata->phy.lp_advertising |= ADVERTISED_10000baseR_FEC; + + pdata->phy.duplex = DUPLEX_FULL; +} + +static void xgbe_phy_status(struct xgbe_prv_data *pdata) +{ + unsigned int reg, link_aneg; + + if (test_bit(XGBE_LINK_ERR, &pdata->dev_state)) { + if (test_and_clear_bit(XGBE_LINK, &pdata->dev_state)) + netif_carrier_off(pdata->netdev); + + pdata->phy.link = 0; + goto adjust_link; + } + + link_aneg = (pdata->phy.autoneg == AUTONEG_ENABLE); + + /* Get the link status. Link status is latched low, so read + * once to clear and then read again to get current state + */ + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); + pdata->phy.link = (reg & MDIO_STAT1_LSTATUS) ? 1 : 0; + + if (pdata->phy.link) { + if (link_aneg && !xgbe_phy_aneg_done(pdata)) { + xgbe_check_link_timeout(pdata); + return; + } + + xgbe_phy_status_aneg(pdata); + + if (test_bit(XGBE_LINK_INIT, &pdata->dev_state)) + clear_bit(XGBE_LINK_INIT, &pdata->dev_state); + + if (!test_bit(XGBE_LINK, &pdata->dev_state)) { + set_bit(XGBE_LINK, &pdata->dev_state); + netif_carrier_on(pdata->netdev); + } + } else { + if (test_bit(XGBE_LINK_INIT, &pdata->dev_state)) { + xgbe_check_link_timeout(pdata); + + if (link_aneg) + return; + } + + xgbe_phy_status_aneg(pdata); + + if (test_bit(XGBE_LINK, &pdata->dev_state)) { + clear_bit(XGBE_LINK, &pdata->dev_state); + netif_carrier_off(pdata->netdev); + } + } + +adjust_link: + xgbe_phy_adjust_link(pdata); +} + +static void xgbe_phy_stop(struct xgbe_prv_data *pdata) +{ + netif_dbg(pdata, link, pdata->netdev, "stopping PHY\n"); + + /* Disable auto-negotiation */ + xgbe_disable_an(pdata); + + /* Disable auto-negotiation interrupts */ + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INTMASK, 0); + + devm_free_irq(pdata->dev, pdata->an_irq, pdata); -err_phy_device: - phy_device_free(phydev); + pdata->phy.link = 0; + if (test_and_clear_bit(XGBE_LINK, &pdata->dev_state)) + netif_carrier_off(pdata->netdev); -err_mdiobus_register: - mdiobus_unregister(mii); + xgbe_phy_adjust_link(pdata); +} + +static int xgbe_phy_start(struct xgbe_prv_data *pdata) +{ + struct net_device *netdev = pdata->netdev; + int ret; + + netif_dbg(pdata, link, pdata->netdev, "starting PHY\n"); + + ret = devm_request_irq(pdata->dev, pdata->an_irq, + xgbe_an_isr, 0, pdata->an_name, + pdata); + if (ret) { + netdev_err(netdev, "phy irq request failed\n"); + return ret; + } + + /* Set initial mode - call the mode setting routines + * directly to insure we are properly configured + */ + if (xgbe_use_xgmii_mode(pdata)) { + xgbe_xgmii_mode(pdata); + } else if (xgbe_use_gmii_mode(pdata)) { + xgbe_gmii_mode(pdata); + } else if (xgbe_use_gmii_2500_mode(pdata)) { + xgbe_gmii_2500_mode(pdata); + } else { + ret = -EINVAL; + goto err_irq; + } + + /* Set up advertisement registers based on current settings */ + xgbe_an_init(pdata); + + /* Enable auto-negotiation interrupts */ + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INTMASK, 0x07); -err_mdiobus_alloc: - mdiobus_free(mii); + return xgbe_phy_config_aneg(pdata); + +err_irq: + devm_free_irq(pdata->dev, pdata->an_irq, pdata); return ret; } -void xgbe_mdio_unregister(struct xgbe_prv_data *pdata) +static int xgbe_phy_reset(struct xgbe_prv_data *pdata) +{ + unsigned int count, reg; + + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + reg |= MDIO_CTRL1_RESET; + XMDIO_WRITE(pdata, MDIO_MMD_PCS, MDIO_CTRL1, reg); + + count = 50; + do { + msleep(20); + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1); + } while ((reg & MDIO_CTRL1_RESET) && --count); + + if (reg & MDIO_CTRL1_RESET) + return -ETIMEDOUT; + + /* Disable auto-negotiation for now */ + xgbe_disable_an(pdata); + + /* Clear auto-negotiation interrupts */ + XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0); + + return 0; +} + +static void xgbe_dump_phy_registers(struct xgbe_prv_data *pdata) { - DBGPR("-->xgbe_mdio_unregister\n"); + struct device *dev = pdata->dev; + + dev_dbg(dev, "\n************* PHY Reg dump **********************\n"); + + dev_dbg(dev, "PCS Control Reg (%#04x) = %#04x\n", MDIO_CTRL1, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_CTRL1)); + dev_dbg(dev, "PCS Status Reg (%#04x) = %#04x\n", MDIO_STAT1, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1)); + dev_dbg(dev, "Phy Id (PHYS ID 1 %#04x)= %#04x\n", MDIO_DEVID1, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID1)); + dev_dbg(dev, "Phy Id (PHYS ID 2 %#04x)= %#04x\n", MDIO_DEVID2, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVID2)); + dev_dbg(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS1, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS1)); + dev_dbg(dev, "Devices in Package (%#04x)= %#04x\n", MDIO_DEVS2, + XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_DEVS2)); + + dev_dbg(dev, "Auto-Neg Control Reg (%#04x) = %#04x\n", MDIO_CTRL1, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1)); + dev_dbg(dev, "Auto-Neg Status Reg (%#04x) = %#04x\n", MDIO_STAT1, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_STAT1)); + dev_dbg(dev, "Auto-Neg Ad Reg 1 (%#04x) = %#04x\n", + MDIO_AN_ADVERTISE, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE)); + dev_dbg(dev, "Auto-Neg Ad Reg 2 (%#04x) = %#04x\n", + MDIO_AN_ADVERTISE + 1, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 1)); + dev_dbg(dev, "Auto-Neg Ad Reg 3 (%#04x) = %#04x\n", + MDIO_AN_ADVERTISE + 2, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_ADVERTISE + 2)); + dev_dbg(dev, "Auto-Neg Completion Reg (%#04x) = %#04x\n", + MDIO_AN_COMP_STAT, + XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_COMP_STAT)); + + dev_dbg(dev, "\n*************************************************\n"); +} + +static void xgbe_phy_init(struct xgbe_prv_data *pdata) +{ + mutex_init(&pdata->an_mutex); + INIT_WORK(&pdata->an_irq_work, xgbe_an_irq_work); + INIT_WORK(&pdata->an_work, xgbe_an_state_machine); + pdata->mdio_mmd = MDIO_MMD_PCS; + + /* Initialize supported features */ + pdata->phy.supported = SUPPORTED_Autoneg; + pdata->phy.supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; + pdata->phy.supported |= SUPPORTED_Backplane; + pdata->phy.supported |= SUPPORTED_10000baseKR_Full; + switch (pdata->speed_set) { + case XGBE_SPEEDSET_1000_10000: + pdata->phy.supported |= SUPPORTED_1000baseKX_Full; + break; + case XGBE_SPEEDSET_2500_10000: + pdata->phy.supported |= SUPPORTED_2500baseX_Full; + break; + } - pdata->phydev = NULL; + pdata->fec_ability = XMDIO_READ(pdata, MDIO_MMD_PMAPMD, + MDIO_PMA_10GBR_FECABLE); + pdata->fec_ability &= (MDIO_PMA_10GBR_FECABLE_ABLE | + MDIO_PMA_10GBR_FECABLE_ERRABLE); + if (pdata->fec_ability & MDIO_PMA_10GBR_FECABLE_ABLE) + pdata->phy.supported |= SUPPORTED_10000baseR_FEC; - module_put(pdata->phy_module); - pdata->phy_module = NULL; + pdata->phy.advertising = pdata->phy.supported; - mdiobus_unregister(pdata->mii); - pdata->mii->priv = NULL; + pdata->phy.address = 0; + + pdata->phy.autoneg = AUTONEG_ENABLE; + pdata->phy.speed = SPEED_UNKNOWN; + pdata->phy.duplex = DUPLEX_UNKNOWN; + + pdata->phy.link = 0; + + pdata->phy.pause_autoneg = pdata->pause_autoneg; + pdata->phy.tx_pause = pdata->tx_pause; + pdata->phy.rx_pause = pdata->rx_pause; + + /* Fix up Flow Control advertising */ + pdata->phy.advertising &= ~ADVERTISED_Pause; + pdata->phy.advertising &= ~ADVERTISED_Asym_Pause; + + if (pdata->rx_pause) { + pdata->phy.advertising |= ADVERTISED_Pause; + pdata->phy.advertising |= ADVERTISED_Asym_Pause; + } + + if (pdata->tx_pause) + pdata->phy.advertising ^= ADVERTISED_Asym_Pause; + + if (netif_msg_drv(pdata)) + xgbe_dump_phy_registers(pdata); +} + +void xgbe_init_function_ptrs_phy(struct xgbe_phy_if *phy_if) +{ + phy_if->phy_init = xgbe_phy_init; - mdiobus_free(pdata->mii); - pdata->mii = NULL; + phy_if->phy_reset = xgbe_phy_reset; + phy_if->phy_start = xgbe_phy_start; + phy_if->phy_stop = xgbe_phy_stop; - DBGPR("<--xgbe_mdio_unregister\n"); + phy_if->phy_status = xgbe_phy_status; + phy_if->phy_config_aneg = xgbe_phy_config_aneg; } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index e62dfa2deab6..63d72a140053 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -129,7 +129,7 @@ #include <net/dcbnl.h> #define XGBE_DRV_NAME "amd-xgbe" -#define XGBE_DRV_VERSION "1.0.0-a" +#define XGBE_DRV_VERSION "1.0.2" #define XGBE_DRV_DESC "AMD 10 Gigabit Ethernet Driver" /* Descriptor related defines */ @@ -178,14 +178,17 @@ #define XGMAC_JUMBO_PACKET_MTU 9000 #define XGMAC_MAX_JUMBO_PACKET 9018 -/* MDIO bus phy name */ -#define XGBE_PHY_NAME "amd_xgbe_phy" -#define XGBE_PRTAD 0 - /* Common property names */ #define XGBE_MAC_ADDR_PROPERTY "mac-address" #define XGBE_PHY_MODE_PROPERTY "phy-mode" #define XGBE_DMA_IRQS_PROPERTY "amd,per-channel-interrupt" +#define XGBE_SPEEDSET_PROPERTY "amd,speed-set" +#define XGBE_BLWC_PROPERTY "amd,serdes-blwc" +#define XGBE_CDR_RATE_PROPERTY "amd,serdes-cdr-rate" +#define XGBE_PQ_SKEW_PROPERTY "amd,serdes-pq-skew" +#define XGBE_TX_AMP_PROPERTY "amd,serdes-tx-amp" +#define XGBE_DFE_CFG_PROPERTY "amd,serdes-dfe-tap-config" +#define XGBE_DFE_ENA_PROPERTY "amd,serdes-dfe-tap-enable" /* Device-tree clock names */ #define XGBE_DMA_CLOCK "dma_clk" @@ -241,6 +244,49 @@ #define XGBE_RSS_LOOKUP_TABLE_TYPE 0 #define XGBE_RSS_HASH_KEY_TYPE 1 +/* Auto-negotiation */ +#define XGBE_AN_MS_TIMEOUT 500 +#define XGBE_LINK_TIMEOUT 10 + +#define XGBE_AN_INT_CMPLT 0x01 +#define XGBE_AN_INC_LINK 0x02 +#define XGBE_AN_PG_RCV 0x04 +#define XGBE_AN_INT_MASK 0x07 + +/* Rate-change complete wait/retry count */ +#define XGBE_RATECHANGE_COUNT 500 + +/* Default SerDes settings */ +#define XGBE_SPEED_10000_BLWC 0 +#define XGBE_SPEED_10000_CDR 0x7 +#define XGBE_SPEED_10000_PLL 0x1 +#define XGBE_SPEED_10000_PQ 0x12 +#define XGBE_SPEED_10000_RATE 0x0 +#define XGBE_SPEED_10000_TXAMP 0xa +#define XGBE_SPEED_10000_WORD 0x7 +#define XGBE_SPEED_10000_DFE_TAP_CONFIG 0x1 +#define XGBE_SPEED_10000_DFE_TAP_ENABLE 0x7f + +#define XGBE_SPEED_2500_BLWC 1 +#define XGBE_SPEED_2500_CDR 0x2 +#define XGBE_SPEED_2500_PLL 0x0 +#define XGBE_SPEED_2500_PQ 0xa +#define XGBE_SPEED_2500_RATE 0x1 +#define XGBE_SPEED_2500_TXAMP 0xf +#define XGBE_SPEED_2500_WORD 0x1 +#define XGBE_SPEED_2500_DFE_TAP_CONFIG 0x3 +#define XGBE_SPEED_2500_DFE_TAP_ENABLE 0x0 + +#define XGBE_SPEED_1000_BLWC 1 +#define XGBE_SPEED_1000_CDR 0x2 +#define XGBE_SPEED_1000_PLL 0x0 +#define XGBE_SPEED_1000_PQ 0xa +#define XGBE_SPEED_1000_RATE 0x3 +#define XGBE_SPEED_1000_TXAMP 0xf +#define XGBE_SPEED_1000_WORD 0x1 +#define XGBE_SPEED_1000_DFE_TAP_CONFIG 0x3 +#define XGBE_SPEED_1000_DFE_TAP_ENABLE 0x0 + struct xgbe_prv_data; struct xgbe_packet_data { @@ -334,8 +380,6 @@ struct xgbe_ring_data { */ unsigned int state_saved; struct { - unsigned int incomplete; - unsigned int context_next; struct sk_buff *skb; unsigned int len; unsigned int error; @@ -414,6 +458,13 @@ struct xgbe_channel { struct xgbe_ring *rx_ring; } ____cacheline_aligned; +enum xgbe_state { + XGBE_DOWN, + XGBE_LINK, + XGBE_LINK_INIT, + XGBE_LINK_ERR, +}; + enum xgbe_int { XGMAC_INT_DMA_CH_SR_TI, XGMAC_INT_DMA_CH_SR_TPS, @@ -445,6 +496,57 @@ enum xgbe_mtl_fifo_size { XGMAC_MTL_FIFO_SIZE_256K = 0x3ff, }; +enum xgbe_speed { + XGBE_SPEED_1000 = 0, + XGBE_SPEED_2500, + XGBE_SPEED_10000, + XGBE_SPEEDS, +}; + +enum xgbe_an { + XGBE_AN_READY = 0, + XGBE_AN_PAGE_RECEIVED, + XGBE_AN_INCOMPAT_LINK, + XGBE_AN_COMPLETE, + XGBE_AN_NO_LINK, + XGBE_AN_ERROR, +}; + +enum xgbe_rx { + XGBE_RX_BPA = 0, + XGBE_RX_XNP, + XGBE_RX_COMPLETE, + XGBE_RX_ERROR, +}; + +enum xgbe_mode { + XGBE_MODE_KR = 0, + XGBE_MODE_KX, +}; + +enum xgbe_speedset { + XGBE_SPEEDSET_1000_10000 = 0, + XGBE_SPEEDSET_2500_10000, +}; + +struct xgbe_phy { + u32 supported; + u32 advertising; + u32 lp_advertising; + + int address; + + int autoneg; + int speed; + int duplex; + + int link; + + int pause_autoneg; + int tx_pause; + int rx_pause; +}; + struct xgbe_mmc_stats { /* Tx Stats */ u64 txoctetcount_gb; @@ -492,6 +594,11 @@ struct xgbe_mmc_stats { u64 rxwatchdogerror; }; +struct xgbe_ext_stats { + u64 tx_tso_packets; + u64 rx_split_header_packets; +}; + struct xgbe_hw_if { int (*tx_complete)(struct xgbe_ring_desc *); @@ -591,6 +698,20 @@ struct xgbe_hw_if { int (*set_rss_lookup_table)(struct xgbe_prv_data *, const u32 *); }; +struct xgbe_phy_if { + /* For initial PHY setup */ + void (*phy_init)(struct xgbe_prv_data *); + + /* For PHY support when setting device up/down */ + int (*phy_reset)(struct xgbe_prv_data *); + int (*phy_start)(struct xgbe_prv_data *); + void (*phy_stop)(struct xgbe_prv_data *); + + /* For PHY support while device is up */ + void (*phy_status)(struct xgbe_prv_data *); + int (*phy_config_aneg)(struct xgbe_prv_data *); +}; + struct xgbe_desc_if { int (*alloc_ring_resources)(struct xgbe_prv_data *); void (*free_ring_resources)(struct xgbe_prv_data *); @@ -660,6 +781,9 @@ struct xgbe_prv_data { /* XGMAC/XPCS related mmio registers */ void __iomem *xgmac_regs; /* XGMAC CSRs */ void __iomem *xpcs_regs; /* XPCS MMD registers */ + void __iomem *rxtx_regs; /* SerDes Rx/Tx CSRs */ + void __iomem *sir0_regs; /* SerDes integration registers (1/2) */ + void __iomem *sir1_regs; /* SerDes integration registers (2/2) */ /* Overall device lock */ spinlock_t lock; @@ -670,10 +794,14 @@ struct xgbe_prv_data { /* RSS addressing mutex */ struct mutex rss_mutex; + /* Flags representing xgbe_state */ + unsigned long dev_state; + int dev_irq; unsigned int per_channel_irq; struct xgbe_hw_if hw_if; + struct xgbe_phy_if phy_if; struct xgbe_desc_if desc_if; /* AXI DMA settings */ @@ -682,6 +810,11 @@ struct xgbe_prv_data { unsigned int arcache; unsigned int awcache; + /* Service routine support */ + struct workqueue_struct *dev_workqueue; + struct work_struct service_work; + struct timer_list service_timer; + /* Rings for Tx/Rx on a DMA channel */ struct xgbe_channel *channel; unsigned int channel_count; @@ -729,27 +862,12 @@ struct xgbe_prv_data { u32 rss_table[XGBE_RSS_MAX_TABLE_SIZE]; u32 rss_options; - /* MDIO settings */ - struct module *phy_module; - char *mii_bus_id; - struct mii_bus *mii; - int mdio_mmd; - struct phy_device *phydev; - int default_autoneg; - int default_speed; - - /* Current PHY settings */ - phy_interface_t phy_mode; - int phy_link; - int phy_speed; - unsigned int phy_tx_pause; - unsigned int phy_rx_pause; - /* Netdev related settings */ unsigned char mac_addr[ETH_ALEN]; netdev_features_t netdev_features; struct napi_struct napi; struct xgbe_mmc_stats mmc_stats; + struct xgbe_ext_stats ext_stats; /* Filtering support */ unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; @@ -787,6 +905,54 @@ struct xgbe_prv_data { /* Keeps track of power mode */ unsigned int power_down; + /* Network interface message level setting */ + u32 msg_enable; + + /* Current PHY settings */ + phy_interface_t phy_mode; + int phy_link; + int phy_speed; + + /* MDIO/PHY related settings */ + struct xgbe_phy phy; + int mdio_mmd; + unsigned long link_check; + + char an_name[IFNAMSIZ + 32]; + struct workqueue_struct *an_workqueue; + + int an_irq; + struct work_struct an_irq_work; + + unsigned int speed_set; + + /* SerDes UEFI configurable settings. + * Switching between modes/speeds requires new values for some + * SerDes settings. The values can be supplied as device + * properties in array format. The first array entry is for + * 1GbE, second for 2.5GbE and third for 10GbE + */ + u32 serdes_blwc[XGBE_SPEEDS]; + u32 serdes_cdr_rate[XGBE_SPEEDS]; + u32 serdes_pq_skew[XGBE_SPEEDS]; + u32 serdes_tx_amp[XGBE_SPEEDS]; + u32 serdes_dfe_tap_cfg[XGBE_SPEEDS]; + u32 serdes_dfe_tap_ena[XGBE_SPEEDS]; + + /* Auto-negotiation state machine support */ + struct mutex an_mutex; + enum xgbe_an an_result; + enum xgbe_an an_state; + enum xgbe_rx kr_state; + enum xgbe_rx kx_state; + struct work_struct an_work; + unsigned int an_supported; + unsigned int parallel_detect; + unsigned int fec_ability; + unsigned long an_start; + + unsigned int lpm_ctrl; /* CTRL1 for resume */ + #ifdef CONFIG_DEBUG_FS struct dentry *xgbe_debugfs; @@ -800,6 +966,7 @@ struct xgbe_prv_data { /* Function prototypes*/ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *); +void xgbe_init_function_ptrs_phy(struct xgbe_phy_if *); void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *); struct net_device_ops *xgbe_get_netdev_ops(void); struct ethtool_ops *xgbe_get_ethtool_ops(void); @@ -807,14 +974,11 @@ struct ethtool_ops *xgbe_get_ethtool_ops(void); const struct dcbnl_rtnl_ops *xgbe_get_dcbnl_ops(void); #endif -int xgbe_mdio_register(struct xgbe_prv_data *); -void xgbe_mdio_unregister(struct xgbe_prv_data *); -void xgbe_dump_phy_registers(struct xgbe_prv_data *); void xgbe_ptp_register(struct xgbe_prv_data *); void xgbe_ptp_unregister(struct xgbe_prv_data *); -void xgbe_dump_tx_desc(struct xgbe_ring *, unsigned int, unsigned int, - unsigned int); -void xgbe_dump_rx_desc(struct xgbe_ring *, struct xgbe_ring_desc *, +void xgbe_dump_tx_desc(struct xgbe_prv_data *, struct xgbe_ring *, + unsigned int, unsigned int, unsigned int); +void xgbe_dump_rx_desc(struct xgbe_prv_data *, struct xgbe_ring *, unsigned int); void xgbe_print_pkt(struct net_device *, struct sk_buff *, bool); void xgbe_get_all_hw_features(struct xgbe_prv_data *); @@ -831,18 +995,6 @@ static inline void xgbe_debugfs_init(struct xgbe_prv_data *pdata) {} static inline void xgbe_debugfs_exit(struct xgbe_prv_data *pdata) {} #endif /* CONFIG_DEBUG_FS */ -/* NOTE: Uncomment for TX and RX DESCRIPTOR DUMP in KERNEL LOG */ -#if 0 -#define XGMAC_ENABLE_TX_DESC_DUMP -#define XGMAC_ENABLE_RX_DESC_DUMP -#endif - -/* NOTE: Uncomment for TX and RX PACKET DUMP in KERNEL LOG */ -#if 0 -#define XGMAC_ENABLE_TX_PKT_DUMP -#define XGMAC_ENABLE_RX_PKT_DUMP -#endif - /* NOTE: Uncomment for function trace log messages in KERNEL LOG */ #if 0 #define YDEBUG @@ -852,10 +1004,8 @@ static inline void xgbe_debugfs_exit(struct xgbe_prv_data *pdata) {} /* For debug prints */ #ifdef YDEBUG #define DBGPR(x...) pr_alert(x) -#define DBGPHY_REGS(x...) xgbe_dump_phy_registers(x) #else #define DBGPR(x...) do { } while (0) -#define DBGPHY_REGS(x...) do { } while (0) #endif #ifdef YDEBUG_MDIO diff --git a/drivers/net/ethernet/apm/xgene/Makefile b/drivers/net/ethernet/apm/xgene/Makefile index 68be565548c0..700b5abe5de5 100644 --- a/drivers/net/ethernet/apm/xgene/Makefile +++ b/drivers/net/ethernet/apm/xgene/Makefile @@ -3,5 +3,5 @@ # xgene-enet-objs := xgene_enet_hw.o xgene_enet_sgmac.o xgene_enet_xgmac.o \ - xgene_enet_main.o xgene_enet_ethtool.o + xgene_enet_main.o xgene_enet_ring2.o xgene_enet_ethtool.o obj-$(CONFIG_NET_XGENE) += xgene-enet.o diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index b927021c6c40..25873d142b95 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -87,10 +87,11 @@ static void xgene_enet_ring_rd32(struct xgene_enet_desc_ring *ring, static void xgene_enet_write_ring_state(struct xgene_enet_desc_ring *ring) { + struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev); int i; xgene_enet_ring_wr32(ring, CSR_RING_CONFIG, ring->num); - for (i = 0; i < NUM_RING_CONFIG; i++) { + for (i = 0; i < pdata->ring_ops->num_ring_config; i++) { xgene_enet_ring_wr32(ring, CSR_RING_WR_BASE + (i * 4), ring->state[i]); } @@ -98,7 +99,7 @@ static void xgene_enet_write_ring_state(struct xgene_enet_desc_ring *ring) static void xgene_enet_clr_ring_state(struct xgene_enet_desc_ring *ring) { - memset(ring->state, 0, sizeof(u32) * NUM_RING_CONFIG); + memset(ring->state, 0, sizeof(ring->state)); xgene_enet_write_ring_state(ring); } @@ -141,8 +142,8 @@ static void xgene_enet_clr_desc_ring_id(struct xgene_enet_desc_ring *ring) xgene_enet_ring_wr32(ring, CSR_RING_ID_BUF, 0); } -struct xgene_enet_desc_ring *xgene_enet_setup_ring( - struct xgene_enet_desc_ring *ring) +static struct xgene_enet_desc_ring *xgene_enet_setup_ring( + struct xgene_enet_desc_ring *ring) { u32 size = ring->size; u32 i, data; @@ -168,7 +169,7 @@ struct xgene_enet_desc_ring *xgene_enet_setup_ring( return ring; } -void xgene_enet_clear_ring(struct xgene_enet_desc_ring *ring) +static void xgene_enet_clear_ring(struct xgene_enet_desc_ring *ring) { u32 data; bool is_bufpool; @@ -186,6 +187,22 @@ out: xgene_enet_clr_ring_state(ring); } +static void xgene_enet_wr_cmd(struct xgene_enet_desc_ring *ring, int count) +{ + iowrite32(count, ring->cmd); +} + +static u32 xgene_enet_ring_len(struct xgene_enet_desc_ring *ring) +{ + u32 __iomem *cmd_base = ring->cmd_base; + u32 ring_state, num_msgs; + + ring_state = ioread32(&cmd_base[1]); + num_msgs = GET_VAL(NUMMSGSINQ, ring_state); + + return num_msgs; +} + void xgene_enet_parse_error(struct xgene_enet_desc_ring *ring, struct xgene_enet_pdata *pdata, enum xgene_enet_err_code status) @@ -803,3 +820,12 @@ struct xgene_port_ops xgene_gport_ops = { .cle_bypass = xgene_enet_cle_bypass, .shutdown = xgene_gport_shutdown, }; + +struct xgene_ring_ops xgene_ring1_ops = { + .num_ring_config = NUM_RING_CONFIG, + .num_ring_id_shift = 6, + .setup = xgene_enet_setup_ring, + .clear = xgene_enet_clear_ring, + .wr_cmd = xgene_enet_wr_cmd, + .len = xgene_enet_ring_len, +}; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h index d9bc89d69266..541bed056012 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h @@ -26,6 +26,7 @@ struct xgene_enet_pdata; struct xgene_enet_stats; +struct xgene_enet_desc_ring; /* clears and then set bits */ static inline void xgene_set_bits(u32 *dst, u32 val, u32 start, u32 len) @@ -101,8 +102,8 @@ enum xgene_enet_rm { #define BLOCK_ETH_CSR_OFFSET 0x2000 #define BLOCK_ETH_RING_IF_OFFSET 0x9000 +#define BLOCK_ETH_CLKRST_CSR_OFFSET 0xc000 #define BLOCK_ETH_DIAG_CSR_OFFSET 0xD000 - #define BLOCK_ETH_MAC_OFFSET 0x0000 #define BLOCK_ETH_MAC_CSR_OFFSET 0x2800 @@ -261,6 +262,7 @@ enum xgene_enet_ring_type { enum xgene_ring_owner { RING_OWNER_ETH0, + RING_OWNER_ETH1, RING_OWNER_CPU = 15, RING_OWNER_INVALID }; @@ -314,9 +316,6 @@ static inline u16 xgene_enet_get_numslots(u16 id, u32 size) size / WORK_DESC_SIZE; } -struct xgene_enet_desc_ring *xgene_enet_setup_ring( - struct xgene_enet_desc_ring *ring); -void xgene_enet_clear_ring(struct xgene_enet_desc_ring *ring); void xgene_enet_parse_error(struct xgene_enet_desc_ring *ring, struct xgene_enet_pdata *pdata, enum xgene_enet_err_code status); @@ -327,5 +326,6 @@ bool xgene_ring_mgr_init(struct xgene_enet_pdata *p); extern struct xgene_mac_ops xgene_gmac_ops; extern struct xgene_port_ops xgene_gport_ops; +extern struct xgene_ring_ops xgene_ring1_ops; #endif /* __XGENE_ENET_HW_H__ */ diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 40d3530d7f30..1bb317532f75 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -28,6 +28,8 @@ #define RES_RING_CSR 1 #define RES_RING_CMD 2 +static const struct of_device_id xgene_enet_of_match[]; + static void xgene_enet_init_bufpool(struct xgene_enet_desc_ring *buf_pool) { struct xgene_enet_raw_desc16 *raw_desc; @@ -48,6 +50,7 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool, { struct sk_buff *skb; struct xgene_enet_raw_desc16 *raw_desc; + struct xgene_enet_pdata *pdata; struct net_device *ndev; struct device *dev; dma_addr_t dma_addr; @@ -58,6 +61,7 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool, ndev = buf_pool->ndev; dev = ndev_to_dev(buf_pool->ndev); + pdata = netdev_priv(ndev); bufdatalen = BUF_LEN_CODE_2K | (SKB_BUFFER_SIZE & GENMASK(11, 0)); len = XGENE_ENET_MAX_MTU; @@ -82,7 +86,7 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool, tail = (tail + 1) & slots; } - iowrite32(nbuf, buf_pool->cmd); + pdata->ring_ops->wr_cmd(buf_pool, nbuf); buf_pool->tail = tail; return 0; @@ -102,26 +106,16 @@ static u8 xgene_enet_hdr_len(const void *data) return (eth->h_proto == htons(ETH_P_8021Q)) ? VLAN_ETH_HLEN : ETH_HLEN; } -static u32 xgene_enet_ring_len(struct xgene_enet_desc_ring *ring) -{ - u32 __iomem *cmd_base = ring->cmd_base; - u32 ring_state, num_msgs; - - ring_state = ioread32(&cmd_base[1]); - num_msgs = ring_state & CREATE_MASK(NUMMSGSINQ_POS, NUMMSGSINQ_LEN); - - return num_msgs >> NUMMSGSINQ_POS; -} - static void xgene_enet_delete_bufpool(struct xgene_enet_desc_ring *buf_pool) { + struct xgene_enet_pdata *pdata = netdev_priv(buf_pool->ndev); struct xgene_enet_raw_desc16 *raw_desc; u32 slots = buf_pool->slots - 1; u32 tail = buf_pool->tail; u32 userinfo; int i, len; - len = xgene_enet_ring_len(buf_pool); + len = pdata->ring_ops->len(buf_pool); for (i = 0; i < len; i++) { tail = (tail - 1) & slots; raw_desc = &buf_pool->raw_desc16[tail]; @@ -131,7 +125,7 @@ static void xgene_enet_delete_bufpool(struct xgene_enet_desc_ring *buf_pool) dev_kfree_skb_any(buf_pool->rx_skb[userinfo]); } - iowrite32(-len, buf_pool->cmd); + pdata->ring_ops->wr_cmd(buf_pool, -len); buf_pool->tail = tail; } @@ -263,8 +257,8 @@ static netdev_tx_t xgene_enet_start_xmit(struct sk_buff *skb, struct xgene_enet_desc_ring *cp_ring = tx_ring->cp_ring; u32 tx_level, cq_level; - tx_level = xgene_enet_ring_len(tx_ring); - cq_level = xgene_enet_ring_len(cp_ring); + tx_level = pdata->ring_ops->len(tx_ring); + cq_level = pdata->ring_ops->len(cp_ring); if (unlikely(tx_level > pdata->tx_qcnt_hi || cq_level > pdata->cp_qcnt_hi)) { netif_stop_queue(ndev); @@ -276,7 +270,7 @@ static netdev_tx_t xgene_enet_start_xmit(struct sk_buff *skb, return NETDEV_TX_OK; } - iowrite32(1, tx_ring->cmd); + pdata->ring_ops->wr_cmd(tx_ring, 1); skb_tx_timestamp(skb); tx_ring->tail = (tx_ring->tail + 1) & (tx_ring->slots - 1); @@ -389,11 +383,11 @@ static int xgene_enet_process_ring(struct xgene_enet_desc_ring *ring, } while (--budget); if (likely(count)) { - iowrite32(-count, ring->cmd); + pdata->ring_ops->wr_cmd(ring, -count); ring->head = head; if (netif_queue_stopped(ring->ndev)) { - if (xgene_enet_ring_len(ring) < pdata->cp_qcnt_low) + if (pdata->ring_ops->len(ring) < pdata->cp_qcnt_low) netif_wake_queue(ring->ndev); } } @@ -510,6 +504,7 @@ static int xgene_enet_open(struct net_device *ndev) else schedule_delayed_work(&pdata->link_work, PHY_POLL_LINK_OFF); + netif_carrier_off(ndev); netif_start_queue(ndev); return ret; @@ -545,7 +540,7 @@ static void xgene_enet_delete_ring(struct xgene_enet_desc_ring *ring) pdata = netdev_priv(ring->ndev); dev = ndev_to_dev(ring->ndev); - xgene_enet_clear_ring(ring); + pdata->ring_ops->clear(ring); dma_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); } @@ -598,15 +593,17 @@ static int xgene_enet_get_ring_size(struct device *dev, static void xgene_enet_free_desc_ring(struct xgene_enet_desc_ring *ring) { + struct xgene_enet_pdata *pdata; struct device *dev; if (!ring) return; dev = ndev_to_dev(ring->ndev); + pdata = netdev_priv(ring->ndev); if (ring->desc_addr) { - xgene_enet_clear_ring(ring); + pdata->ring_ops->clear(ring); dma_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); } devm_kfree(dev, ring); @@ -637,6 +634,25 @@ static void xgene_enet_free_desc_rings(struct xgene_enet_pdata *pdata) } } +static bool is_irq_mbox_required(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring) +{ + if ((pdata->enet_id == XGENE_ENET2) && + (xgene_enet_ring_owner(ring->id) == RING_OWNER_CPU)) { + return true; + } + + return false; +} + +static void __iomem *xgene_enet_ring_cmd_base(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring) +{ + u8 num_ring_id_shift = pdata->ring_ops->num_ring_id_shift; + + return pdata->ring_cmd_addr + (ring->num << num_ring_id_shift); +} + static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring( struct net_device *ndev, u32 ring_num, enum xgene_enet_ring_cfgsize cfgsize, u32 ring_id) @@ -668,9 +684,20 @@ static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring( } ring->size = size; - ring->cmd_base = pdata->ring_cmd_addr + (ring->num << 6); + if (is_irq_mbox_required(pdata, ring)) { + ring->irq_mbox_addr = dma_zalloc_coherent(dev, INTR_MBOX_SIZE, + &ring->irq_mbox_dma, GFP_KERNEL); + if (!ring->irq_mbox_addr) { + dma_free_coherent(dev, size, ring->desc_addr, + ring->dma); + devm_kfree(dev, ring); + return NULL; + } + } + + ring->cmd_base = xgene_enet_ring_cmd_base(pdata, ring); ring->cmd = ring->cmd_base + INC_DEC_CMD_ADDR; - ring = xgene_enet_setup_ring(ring); + ring = pdata->ring_ops->setup(ring); netdev_dbg(ndev, "ring info: num=%d size=%d id=%d slots=%d\n", ring->num, ring->size, ring->id, ring->slots); @@ -682,12 +709,34 @@ static u16 xgene_enet_get_ring_id(enum xgene_ring_owner owner, u8 bufnum) return (owner << 6) | (bufnum & GENMASK(5, 0)); } +static enum xgene_ring_owner xgene_derive_ring_owner(struct xgene_enet_pdata *p) +{ + enum xgene_ring_owner owner; + + if (p->enet_id == XGENE_ENET1) { + switch (p->phy_mode) { + case PHY_INTERFACE_MODE_SGMII: + owner = RING_OWNER_ETH0; + break; + default: + owner = (!p->port_id) ? RING_OWNER_ETH0 : + RING_OWNER_ETH1; + break; + } + } else { + owner = (!p->port_id) ? RING_OWNER_ETH0 : RING_OWNER_ETH1; + } + + return owner; +} + static int xgene_enet_create_desc_rings(struct net_device *ndev) { struct xgene_enet_pdata *pdata = netdev_priv(ndev); struct device *dev = ndev_to_dev(ndev); struct xgene_enet_desc_ring *rx_ring, *tx_ring, *cp_ring; struct xgene_enet_desc_ring *buf_pool = NULL; + enum xgene_ring_owner owner; u8 cpu_bufnum = pdata->cpu_bufnum; u8 eth_bufnum = pdata->eth_bufnum; u8 bp_bufnum = pdata->bp_bufnum; @@ -696,6 +745,7 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) int ret; /* allocate rx descriptor ring */ + owner = xgene_derive_ring_owner(pdata); ring_id = xgene_enet_get_ring_id(RING_OWNER_CPU, cpu_bufnum++); rx_ring = xgene_enet_create_desc_ring(ndev, ring_num++, RING_CFGSIZE_16KB, ring_id); @@ -705,7 +755,8 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) } /* allocate buffer pool for receiving packets */ - ring_id = xgene_enet_get_ring_id(RING_OWNER_ETH0, bp_bufnum++); + owner = xgene_derive_ring_owner(pdata); + ring_id = xgene_enet_get_ring_id(owner, bp_bufnum++); buf_pool = xgene_enet_create_desc_ring(ndev, ring_num++, RING_CFGSIZE_2KB, ring_id); if (!buf_pool) { @@ -734,7 +785,8 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) pdata->rx_ring = rx_ring; /* allocate tx descriptor ring */ - ring_id = xgene_enet_get_ring_id(RING_OWNER_ETH0, eth_bufnum++); + owner = xgene_derive_ring_owner(pdata); + ring_id = xgene_enet_get_ring_id(owner, eth_bufnum++); tx_ring = xgene_enet_create_desc_ring(ndev, ring_num++, RING_CFGSIZE_16KB, ring_id); if (!tx_ring) { @@ -824,14 +876,21 @@ static int xgene_get_port_id(struct device *dev, struct xgene_enet_pdata *pdata) int ret; ret = device_property_read_u32(dev, "port-id", &id); - if (!ret && id > 1) { - dev_err(dev, "Incorrect port-id specified\n"); - return -ENODEV; - } - pdata->port_id = id; + switch (ret) { + case -EINVAL: + pdata->port_id = 0; + ret = 0; + break; + case 0: + pdata->port_id = id & BIT(0); + break; + default: + dev_err(dev, "Incorrect port-id specified: errno: %d\n", ret); + break; + } - return 0; + return ret; } static int xgene_get_mac_address(struct device *dev, @@ -876,6 +935,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) struct device *dev; struct resource *res; void __iomem *base_addr; + u32 offset; int ret; pdev = pdata->pdev; @@ -962,14 +1022,20 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) pdata->clk = NULL; } - base_addr = pdata->base_addr - (pdata->port_id * MAC_OFFSET); + if (pdata->phy_mode != PHY_INTERFACE_MODE_XGMII) + base_addr = pdata->base_addr - (pdata->port_id * MAC_OFFSET); + else + base_addr = pdata->base_addr; pdata->eth_csr_addr = base_addr + BLOCK_ETH_CSR_OFFSET; pdata->eth_ring_if_addr = base_addr + BLOCK_ETH_RING_IF_OFFSET; pdata->eth_diag_csr_addr = base_addr + BLOCK_ETH_DIAG_CSR_OFFSET; if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII || pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) { pdata->mcx_mac_addr = pdata->base_addr + BLOCK_ETH_MAC_OFFSET; - pdata->mcx_mac_csr_addr = base_addr + BLOCK_ETH_MAC_CSR_OFFSET; + offset = (pdata->enet_id == XGENE_ENET1) ? + BLOCK_ETH_MAC_CSR_OFFSET : + X2_BLOCK_ETH_MAC_CSR_OFFSET; + pdata->mcx_mac_csr_addr = base_addr + offset; } else { pdata->mcx_mac_addr = base_addr + BLOCK_AXG_MAC_OFFSET; pdata->mcx_mac_csr_addr = base_addr + BLOCK_AXG_MAC_CSR_OFFSET; @@ -1034,23 +1100,44 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata) break; } - switch (pdata->port_id) { - case 0: - pdata->cpu_bufnum = START_CPU_BUFNUM_0; - pdata->eth_bufnum = START_ETH_BUFNUM_0; - pdata->bp_bufnum = START_BP_BUFNUM_0; - pdata->ring_num = START_RING_NUM_0; - break; - case 1: - pdata->cpu_bufnum = START_CPU_BUFNUM_1; - pdata->eth_bufnum = START_ETH_BUFNUM_1; - pdata->bp_bufnum = START_BP_BUFNUM_1; - pdata->ring_num = START_RING_NUM_1; - break; - default: - break; + if (pdata->enet_id == XGENE_ENET1) { + switch (pdata->port_id) { + case 0: + pdata->cpu_bufnum = START_CPU_BUFNUM_0; + pdata->eth_bufnum = START_ETH_BUFNUM_0; + pdata->bp_bufnum = START_BP_BUFNUM_0; + pdata->ring_num = START_RING_NUM_0; + break; + case 1: + pdata->cpu_bufnum = START_CPU_BUFNUM_1; + pdata->eth_bufnum = START_ETH_BUFNUM_1; + pdata->bp_bufnum = START_BP_BUFNUM_1; + pdata->ring_num = START_RING_NUM_1; + break; + default: + break; + } + pdata->ring_ops = &xgene_ring1_ops; + } else { + switch (pdata->port_id) { + case 0: + pdata->cpu_bufnum = X2_START_CPU_BUFNUM_0; + pdata->eth_bufnum = X2_START_ETH_BUFNUM_0; + pdata->bp_bufnum = X2_START_BP_BUFNUM_0; + pdata->ring_num = X2_START_RING_NUM_0; + break; + case 1: + pdata->cpu_bufnum = X2_START_CPU_BUFNUM_1; + pdata->eth_bufnum = X2_START_ETH_BUFNUM_1; + pdata->bp_bufnum = X2_START_BP_BUFNUM_1; + pdata->ring_num = X2_START_RING_NUM_1; + break; + default: + break; + } + pdata->rm = RM0; + pdata->ring_ops = &xgene_ring2_ops; } - } static void xgene_enet_napi_add(struct xgene_enet_pdata *pdata) @@ -1086,6 +1173,9 @@ static int xgene_enet_probe(struct platform_device *pdev) struct xgene_enet_pdata *pdata; struct device *dev = &pdev->dev; struct xgene_mac_ops *mac_ops; +#ifdef CONFIG_OF + const struct of_device_id *of_id; +#endif int ret; ndev = alloc_etherdev(sizeof(struct xgene_enet_pdata)); @@ -1104,6 +1194,17 @@ static int xgene_enet_probe(struct platform_device *pdev) NETIF_F_GSO | NETIF_F_GRO; +#ifdef CONFIG_OF + of_id = of_match_device(xgene_enet_of_match, &pdev->dev); + if (of_id) { + pdata->enet_id = (enum xgene_enet_id)of_id->data; + if (!pdata->enet_id) { + free_netdev(ndev); + return -ENODEV; + } + } +#endif + ret = xgene_enet_get_resources(pdata); if (ret) goto err; @@ -1175,9 +1276,11 @@ MODULE_DEVICE_TABLE(acpi, xgene_enet_acpi_match); #ifdef CONFIG_OF static const struct of_device_id xgene_enet_of_match[] = { - {.compatible = "apm,xgene-enet",}, - {.compatible = "apm,xgene1-sgenet",}, - {.compatible = "apm,xgene1-xgenet",}, + {.compatible = "apm,xgene-enet", .data = (void *)XGENE_ENET1}, + {.compatible = "apm,xgene1-sgenet", .data = (void *)XGENE_ENET1}, + {.compatible = "apm,xgene1-xgenet", .data = (void *)XGENE_ENET1}, + {.compatible = "apm,xgene2-sgenet", .data = (void *)XGENE_ENET2}, + {.compatible = "apm,xgene2-xgenet", .data = (void *)XGENE_ENET2}, {}, }; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index 8f3d232b09bc..1c85fc87703a 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -35,6 +35,7 @@ #include <linux/if_vlan.h> #include <linux/phy.h> #include "xgene_enet_hw.h" +#include "xgene_enet_ring2.h" #define XGENE_DRV_VERSION "v1.0" #define XGENE_ENET_MAX_MTU 1536 @@ -51,12 +52,26 @@ #define START_BP_BUFNUM_1 0x2A #define START_RING_NUM_1 264 +#define X2_START_CPU_BUFNUM_0 0 +#define X2_START_ETH_BUFNUM_0 0 +#define X2_START_BP_BUFNUM_0 0x20 +#define X2_START_RING_NUM_0 0 +#define X2_START_CPU_BUFNUM_1 0xc +#define X2_START_ETH_BUFNUM_1 0 +#define X2_START_BP_BUFNUM_1 0x20 +#define X2_START_RING_NUM_1 256 + #define IRQ_ID_SIZE 16 #define XGENE_MAX_TXC_RINGS 1 #define PHY_POLL_LINK_ON (10 * HZ) #define PHY_POLL_LINK_OFF (PHY_POLL_LINK_ON / 5) +enum xgene_enet_id { + XGENE_ENET1 = 1, + XGENE_ENET2 +}; + /* software context of a descriptor ring */ struct xgene_enet_desc_ring { struct net_device *ndev; @@ -68,10 +83,12 @@ struct xgene_enet_desc_ring { u16 irq; char irq_name[IRQ_ID_SIZE]; u32 size; - u32 state[NUM_RING_CONFIG]; + u32 state[X2_NUM_RING_CONFIG]; void __iomem *cmd_base; void __iomem *cmd; dma_addr_t dma; + dma_addr_t irq_mbox_dma; + void *irq_mbox_addr; u16 dst_ring_num; u8 nbufpool; struct sk_buff *(*rx_skb); @@ -105,6 +122,15 @@ struct xgene_port_ops { void (*shutdown)(struct xgene_enet_pdata *pdata); }; +struct xgene_ring_ops { + u8 num_ring_config; + u8 num_ring_id_shift; + struct xgene_enet_desc_ring * (*setup)(struct xgene_enet_desc_ring *); + void (*clear)(struct xgene_enet_desc_ring *); + void (*wr_cmd)(struct xgene_enet_desc_ring *, int); + u32 (*len)(struct xgene_enet_desc_ring *); +}; + /* ethernet private data */ struct xgene_enet_pdata { struct net_device *ndev; @@ -113,6 +139,7 @@ struct xgene_enet_pdata { int phy_speed; struct clk *clk; struct platform_device *pdev; + enum xgene_enet_id enet_id; struct xgene_enet_desc_ring *tx_ring; struct xgene_enet_desc_ring *rx_ring; char *dev_name; @@ -136,6 +163,7 @@ struct xgene_enet_pdata { struct rtnl_link_stats64 stats; struct xgene_mac_ops *mac_ops; struct xgene_port_ops *port_ops; + struct xgene_ring_ops *ring_ops; struct delayed_work link_work; u32 port_id; u8 cpu_bufnum; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c new file mode 100644 index 000000000000..0b6896bb351e --- /dev/null +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c @@ -0,0 +1,200 @@ +/* Applied Micro X-Gene SoC Ethernet Driver + * + * Copyright (c) 2015, Applied Micro Circuits Corporation + * Author: Iyappan Subramanian <isubramanian@apm.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This program is distributed in the hope that 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, see <http://www.gnu.org/licenses/>. + */ + +#include "xgene_enet_main.h" +#include "xgene_enet_hw.h" +#include "xgene_enet_ring2.h" + +static void xgene_enet_ring_init(struct xgene_enet_desc_ring *ring) +{ + u32 *ring_cfg = ring->state; + u64 addr = ring->dma; + + if (xgene_enet_ring_owner(ring->id) == RING_OWNER_CPU) { + ring_cfg[0] |= SET_VAL(X2_INTLINE, ring->id & RING_BUFNUM_MASK); + ring_cfg[3] |= SET_BIT(X2_DEQINTEN); + } + ring_cfg[0] |= SET_VAL(X2_CFGCRID, 1); + + addr >>= 8; + ring_cfg[2] |= QCOHERENT | SET_VAL(RINGADDRL, addr); + + addr >>= 27; + ring_cfg[3] |= SET_VAL(RINGSIZE, ring->cfgsize) + | ACCEPTLERR + | SET_VAL(RINGADDRH, addr); + ring_cfg[4] |= SET_VAL(X2_SELTHRSH, 1); + ring_cfg[5] |= SET_BIT(X2_QBASE_AM) | SET_BIT(X2_MSG_AM); +} + +static void xgene_enet_ring_set_type(struct xgene_enet_desc_ring *ring) +{ + u32 *ring_cfg = ring->state; + bool is_bufpool; + u32 val; + + is_bufpool = xgene_enet_is_bufpool(ring->id); + val = (is_bufpool) ? RING_BUFPOOL : RING_REGULAR; + ring_cfg[4] |= SET_VAL(X2_RINGTYPE, val); + if (is_bufpool) + ring_cfg[3] |= SET_VAL(RINGMODE, BUFPOOL_MODE); +} + +static void xgene_enet_ring_set_recombbuf(struct xgene_enet_desc_ring *ring) +{ + u32 *ring_cfg = ring->state; + + ring_cfg[3] |= RECOMBBUF; + ring_cfg[4] |= SET_VAL(X2_RECOMTIMEOUT, 0x7); +} + +static void xgene_enet_ring_wr32(struct xgene_enet_desc_ring *ring, + u32 offset, u32 data) +{ + struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev); + + iowrite32(data, pdata->ring_csr_addr + offset); +} + +static void xgene_enet_write_ring_state(struct xgene_enet_desc_ring *ring) +{ + struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev); + int i; + + xgene_enet_ring_wr32(ring, CSR_RING_CONFIG, ring->num); + for (i = 0; i < pdata->ring_ops->num_ring_config; i++) { + xgene_enet_ring_wr32(ring, CSR_RING_WR_BASE + (i * 4), + ring->state[i]); + } +} + +static void xgene_enet_clr_ring_state(struct xgene_enet_desc_ring *ring) +{ + memset(ring->state, 0, sizeof(ring->state)); + xgene_enet_write_ring_state(ring); +} + +static void xgene_enet_set_ring_state(struct xgene_enet_desc_ring *ring) +{ + enum xgene_ring_owner owner; + + xgene_enet_ring_set_type(ring); + + owner = xgene_enet_ring_owner(ring->id); + if (owner == RING_OWNER_ETH0 || owner == RING_OWNER_ETH1) + xgene_enet_ring_set_recombbuf(ring); + + xgene_enet_ring_init(ring); + xgene_enet_write_ring_state(ring); +} + +static void xgene_enet_set_ring_id(struct xgene_enet_desc_ring *ring) +{ + u32 ring_id_val, ring_id_buf; + bool is_bufpool; + + if (xgene_enet_ring_owner(ring->id) == RING_OWNER_CPU) + return; + + is_bufpool = xgene_enet_is_bufpool(ring->id); + + ring_id_val = ring->id & GENMASK(9, 0); + ring_id_val |= OVERWRITE; + + ring_id_buf = (ring->num << 9) & GENMASK(18, 9); + ring_id_buf |= PREFETCH_BUF_EN; + if (is_bufpool) + ring_id_buf |= IS_BUFFER_POOL; + + xgene_enet_ring_wr32(ring, CSR_RING_ID, ring_id_val); + xgene_enet_ring_wr32(ring, CSR_RING_ID_BUF, ring_id_buf); +} + +static void xgene_enet_clr_desc_ring_id(struct xgene_enet_desc_ring *ring) +{ + u32 ring_id; + + ring_id = ring->id | OVERWRITE; + xgene_enet_ring_wr32(ring, CSR_RING_ID, ring_id); + xgene_enet_ring_wr32(ring, CSR_RING_ID_BUF, 0); +} + +static struct xgene_enet_desc_ring *xgene_enet_setup_ring( + struct xgene_enet_desc_ring *ring) +{ + bool is_bufpool; + u32 addr, i; + + xgene_enet_clr_ring_state(ring); + xgene_enet_set_ring_state(ring); + xgene_enet_set_ring_id(ring); + + ring->slots = xgene_enet_get_numslots(ring->id, ring->size); + + is_bufpool = xgene_enet_is_bufpool(ring->id); + if (is_bufpool || xgene_enet_ring_owner(ring->id) != RING_OWNER_CPU) + return ring; + + addr = CSR_VMID0_INTR_MBOX + (4 * (ring->id & RING_BUFNUM_MASK)); + xgene_enet_ring_wr32(ring, addr, ring->irq_mbox_dma >> 10); + + for (i = 0; i < ring->slots; i++) + xgene_enet_mark_desc_slot_empty(&ring->raw_desc[i]); + + return ring; +} + +static void xgene_enet_clear_ring(struct xgene_enet_desc_ring *ring) +{ + xgene_enet_clr_desc_ring_id(ring); + xgene_enet_clr_ring_state(ring); +} + +static void xgene_enet_wr_cmd(struct xgene_enet_desc_ring *ring, int count) +{ + u32 data = 0; + + if (xgene_enet_ring_owner(ring->id) == RING_OWNER_CPU) { + data = SET_VAL(X2_INTLINE, ring->id & RING_BUFNUM_MASK) | + INTR_CLEAR; + } + data |= (count & GENMASK(16, 0)); + + iowrite32(data, ring->cmd); +} + +static u32 xgene_enet_ring_len(struct xgene_enet_desc_ring *ring) +{ + u32 __iomem *cmd_base = ring->cmd_base; + u32 ring_state, num_msgs; + + ring_state = ioread32(&cmd_base[1]); + num_msgs = GET_VAL(X2_NUMMSGSINQ, ring_state); + + return num_msgs; +} + +struct xgene_ring_ops xgene_ring2_ops = { + .num_ring_config = X2_NUM_RING_CONFIG, + .num_ring_id_shift = 13, + .setup = xgene_enet_setup_ring, + .clear = xgene_enet_clear_ring, + .wr_cmd = xgene_enet_wr_cmd, + .len = xgene_enet_ring_len, +}; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h new file mode 100644 index 000000000000..8b235db23c42 --- /dev/null +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.h @@ -0,0 +1,49 @@ +/* Applied Micro X-Gene SoC Ethernet Driver + * + * Copyright (c) 2015, Applied Micro Circuits Corporation + * Author: Iyappan Subramanian <isubramanian@apm.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This program is distributed in the hope that 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, see <http://www.gnu.org/licenses/>. + */ + +#ifndef __XGENE_ENET_RING2_H__ +#define __XGENE_ENET_RING2_H__ + +#include "xgene_enet_main.h" + +#define X2_NUM_RING_CONFIG 6 + +#define INTR_MBOX_SIZE 1024 +#define CSR_VMID0_INTR_MBOX 0x0270 +#define INTR_CLEAR BIT(23) + +#define X2_MSG_AM_POS 10 +#define X2_QBASE_AM_POS 11 +#define X2_INTLINE_POS 24 +#define X2_INTLINE_LEN 5 +#define X2_CFGCRID_POS 29 +#define X2_CFGCRID_LEN 3 +#define X2_SELTHRSH_POS 7 +#define X2_SELTHRSH_LEN 3 +#define X2_RINGTYPE_POS 23 +#define X2_RINGTYPE_LEN 2 +#define X2_DEQINTEN_POS 29 +#define X2_RECOMTIMEOUT_POS 0 +#define X2_RECOMTIMEOUT_LEN 7 +#define X2_NUMMSGSINQ_POS 0 +#define X2_NUMMSGSINQ_LEN 17 + +extern struct xgene_ring_ops xgene_ring2_ops; + +#endif /* __XGENE_ENET_RING2_H__ */ diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c index f27fb6f2a93b..ff240b3cb2b8 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c @@ -21,6 +21,7 @@ #include "xgene_enet_main.h" #include "xgene_enet_hw.h" #include "xgene_enet_sgmac.h" +#include "xgene_enet_xgmac.h" static void xgene_enet_wr_csr(struct xgene_enet_pdata *p, u32 offset, u32 val) { @@ -39,6 +40,14 @@ static void xgene_enet_wr_diag_csr(struct xgene_enet_pdata *p, iowrite32(val, p->eth_diag_csr_addr + offset); } +static void xgene_enet_wr_mcx_csr(struct xgene_enet_pdata *pdata, + u32 offset, u32 val) +{ + void __iomem *addr = pdata->mcx_mac_csr_addr + offset; + + iowrite32(val, addr); +} + static bool xgene_enet_wr_indirect(struct xgene_indirect_ctl *ctl, u32 wr_addr, u32 wr_data) { @@ -140,8 +149,9 @@ static int xgene_enet_ecc_init(struct xgene_enet_pdata *p) static void xgene_enet_config_ring_if_assoc(struct xgene_enet_pdata *p) { - u32 val = 0xffffffff; + u32 val; + val = (p->enet_id == XGENE_ENET1) ? 0xffffffff : 0; xgene_enet_wr_ring_if(p, ENET_CFGSSQMIWQASSOC_ADDR, val); xgene_enet_wr_ring_if(p, ENET_CFGSSQMIFPQASSOC_ADDR, val); } @@ -227,6 +237,8 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) { u32 data, loop = 10; u32 offset = p->port_id * 4; + u32 enet_spare_cfg_reg, rsif_config_reg; + u32 cfg_bypass_reg, rx_dv_gate_reg; xgene_sgmac_reset(p); @@ -239,7 +251,7 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) SGMII_STATUS_ADDR >> 2); if ((data & AUTO_NEG_COMPLETE) && (data & LINK_STATUS)) break; - usleep_range(10, 20); + usleep_range(1000, 2000); } if (!(data & AUTO_NEG_COMPLETE) || !(data & LINK_STATUS)) netdev_err(p->ndev, "Auto-negotiation failed\n"); @@ -249,33 +261,38 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) xgene_enet_wr_mac(p, MAC_CONFIG_2_ADDR, data | FULL_DUPLEX2); xgene_enet_wr_mac(p, INTERFACE_CONTROL_ADDR, ENET_GHD_MODE); - data = xgene_enet_rd_csr(p, ENET_SPARE_CFG_REG_ADDR); + if (p->enet_id == XGENE_ENET1) { + enet_spare_cfg_reg = ENET_SPARE_CFG_REG_ADDR; + rsif_config_reg = RSIF_CONFIG_REG_ADDR; + cfg_bypass_reg = CFG_BYPASS_ADDR; + rx_dv_gate_reg = SG_RX_DV_GATE_REG_0_ADDR; + } else { + enet_spare_cfg_reg = XG_ENET_SPARE_CFG_REG_ADDR; + rsif_config_reg = XG_RSIF_CONFIG_REG_ADDR; + cfg_bypass_reg = XG_CFG_BYPASS_ADDR; + rx_dv_gate_reg = XG_MCX_RX_DV_GATE_REG_0_ADDR; + } + + data = xgene_enet_rd_csr(p, enet_spare_cfg_reg); data |= MPA_IDLE_WITH_QMI_EMPTY; - xgene_enet_wr_csr(p, ENET_SPARE_CFG_REG_ADDR, data); + xgene_enet_wr_csr(p, enet_spare_cfg_reg, data); xgene_sgmac_set_mac_addr(p); - data = xgene_enet_rd_csr(p, DEBUG_REG_ADDR); - data |= CFG_BYPASS_UNISEC_TX | CFG_BYPASS_UNISEC_RX; - xgene_enet_wr_csr(p, DEBUG_REG_ADDR, data); - /* Adjust MDC clock frequency */ data = xgene_enet_rd_mac(p, MII_MGMT_CONFIG_ADDR); MGMT_CLOCK_SEL_SET(&data, 7); xgene_enet_wr_mac(p, MII_MGMT_CONFIG_ADDR, data); /* Enable drop if bufpool not available */ - data = xgene_enet_rd_csr(p, RSIF_CONFIG_REG_ADDR); + data = xgene_enet_rd_csr(p, rsif_config_reg); data |= CFG_RSIF_FPBUFF_TIMEOUT_EN; - xgene_enet_wr_csr(p, RSIF_CONFIG_REG_ADDR, data); - - /* Rtype should be copied from FP */ - xgene_enet_wr_csr(p, RSIF_RAM_DBG_REG0_ADDR, 0); + xgene_enet_wr_csr(p, rsif_config_reg, data); /* Bypass traffic gating */ - xgene_enet_wr_csr(p, CFG_LINK_AGGR_RESUME_0_ADDR + offset, TX_PORT0); - xgene_enet_wr_csr(p, CFG_BYPASS_ADDR, RESUME_TX); - xgene_enet_wr_csr(p, SG_RX_DV_GATE_REG_0_ADDR + offset, RESUME_RX0); + xgene_enet_wr_csr(p, XG_ENET_SPARE_CFG_REG_1_ADDR, 0x84); + xgene_enet_wr_csr(p, cfg_bypass_reg, RESUME_TX); + xgene_enet_wr_mcx_csr(p, rx_dv_gate_reg + offset, RESUME_RX0); } static void xgene_sgmac_rxtx(struct xgene_enet_pdata *p, u32 bits, bool set) @@ -331,14 +348,23 @@ static void xgene_enet_cle_bypass(struct xgene_enet_pdata *p, u32 dst_ring_num, u16 bufpool_id) { u32 data, fpsel; + u32 cle_bypass_reg0, cle_bypass_reg1; u32 offset = p->port_id * MAC_OFFSET; + if (p->enet_id == XGENE_ENET1) { + cle_bypass_reg0 = CLE_BYPASS_REG0_0_ADDR; + cle_bypass_reg1 = CLE_BYPASS_REG1_0_ADDR; + } else { + cle_bypass_reg0 = XCLE_BYPASS_REG0_ADDR; + cle_bypass_reg1 = XCLE_BYPASS_REG1_ADDR; + } + data = CFG_CLE_BYPASS_EN0; - xgene_enet_wr_csr(p, CLE_BYPASS_REG0_0_ADDR + offset, data); + xgene_enet_wr_csr(p, cle_bypass_reg0 + offset, data); fpsel = xgene_enet_ring_bufnum(bufpool_id) - 0x20; data = CFG_CLE_DSTQID0(dst_ring_num) | CFG_CLE_FPSEL0(fpsel); - xgene_enet_wr_csr(p, CLE_BYPASS_REG1_0_ADDR + offset, data); + xgene_enet_wr_csr(p, cle_bypass_reg1 + offset, data); } static void xgene_enet_shutdown(struct xgene_enet_pdata *p) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c index a18a9d1f1143..27ba2fe3fca6 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c @@ -122,7 +122,6 @@ static bool xgene_enet_rd_indirect(void __iomem *addr, void __iomem *rd, return true; } - static void xgene_enet_rd_mac(struct xgene_enet_pdata *pdata, u32 rd_addr, u32 *rd_data) { diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h index 5a5296a6d1df..bf0a99435737 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h @@ -21,9 +21,28 @@ #ifndef __XGENE_ENET_XGMAC_H__ #define __XGENE_ENET_XGMAC_H__ +#define X2_BLOCK_ETH_MAC_CSR_OFFSET 0x3000 #define BLOCK_AXG_MAC_OFFSET 0x0800 #define BLOCK_AXG_MAC_CSR_OFFSET 0x2000 +#define XGENET_CONFIG_REG_ADDR 0x20 +#define XGENET_SRST_ADDR 0x00 +#define XGENET_CLKEN_ADDR 0x08 + +#define CSR_CLK BIT(0) +#define XGENET_CLK BIT(1) +#define PCS_CLK BIT(3) +#define AN_REF_CLK BIT(4) +#define AN_CLK BIT(5) +#define AD_CLK BIT(6) + +#define CSR_RST BIT(0) +#define XGENET_RST BIT(1) +#define PCS_RST BIT(3) +#define AN_REF_RST BIT(4) +#define AN_RST BIT(5) +#define AD_RST BIT(6) + #define AXGMAC_CONFIG_0 0x0000 #define AXGMAC_CONFIG_1 0x0004 #define HSTMACRST BIT(31) @@ -38,6 +57,7 @@ #define HSTMACADR_MSW_ADDR 0x0014 #define HSTMAXFRAME_LENGTH_ADDR 0x0020 +#define XG_MCX_RX_DV_GATE_REG_0_ADDR 0x0004 #define XG_RSIF_CONFIG_REG_ADDR 0x00a0 #define XCLE_BYPASS_REG0_ADDR 0x0160 #define XCLE_BYPASS_REG1_ADDR 0x0164 diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 783543ad1fcf..909ad7a0d480 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -456,6 +456,67 @@ static int bcm_sysport_set_wol(struct net_device *dev, return 0; } +static int bcm_sysport_get_coalesce(struct net_device *dev, + struct ethtool_coalesce *ec) +{ + struct bcm_sysport_priv *priv = netdev_priv(dev); + u32 reg; + + reg = tdma_readl(priv, TDMA_DESC_RING_INTR_CONTROL(0)); + + ec->tx_coalesce_usecs = (reg >> RING_TIMEOUT_SHIFT) * 8192 / 1000; + ec->tx_max_coalesced_frames = reg & RING_INTR_THRESH_MASK; + + reg = rdma_readl(priv, RDMA_MBDONE_INTR); + + ec->rx_coalesce_usecs = (reg >> RDMA_TIMEOUT_SHIFT) * 8192 / 1000; + ec->rx_max_coalesced_frames = reg & RDMA_INTR_THRESH_MASK; + + return 0; +} + +static int bcm_sysport_set_coalesce(struct net_device *dev, + struct ethtool_coalesce *ec) +{ + struct bcm_sysport_priv *priv = netdev_priv(dev); + unsigned int i; + u32 reg; + + /* Base system clock is 125Mhz, DMA timeout is this reference clock + * divided by 1024, which yield roughly 8.192 us, our maximum value has + * to fit in the RING_TIMEOUT_MASK (16 bits). + */ + if (ec->tx_max_coalesced_frames > RING_INTR_THRESH_MASK || + ec->tx_coalesce_usecs > (RING_TIMEOUT_MASK * 8) + 1 || + ec->rx_max_coalesced_frames > RDMA_INTR_THRESH_MASK || + ec->rx_coalesce_usecs > (RDMA_TIMEOUT_MASK * 8) + 1) + return -EINVAL; + + if ((ec->tx_coalesce_usecs == 0 && ec->tx_max_coalesced_frames == 0) || + (ec->rx_coalesce_usecs == 0 && ec->rx_max_coalesced_frames == 0)) + return -EINVAL; + + for (i = 0; i < dev->num_tx_queues; i++) { + reg = tdma_readl(priv, TDMA_DESC_RING_INTR_CONTROL(i)); + reg &= ~(RING_INTR_THRESH_MASK | + RING_TIMEOUT_MASK << RING_TIMEOUT_SHIFT); + reg |= ec->tx_max_coalesced_frames; + reg |= DIV_ROUND_UP(ec->tx_coalesce_usecs * 1000, 8192) << + RING_TIMEOUT_SHIFT; + tdma_writel(priv, reg, TDMA_DESC_RING_INTR_CONTROL(i)); + } + + reg = rdma_readl(priv, RDMA_MBDONE_INTR); + reg &= ~(RDMA_INTR_THRESH_MASK | + RDMA_TIMEOUT_MASK << RDMA_TIMEOUT_SHIFT); + reg |= ec->rx_max_coalesced_frames; + reg |= DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000, 8192) << + RDMA_TIMEOUT_SHIFT; + rdma_writel(priv, reg, RDMA_MBDONE_INTR); + + return 0; +} + static void bcm_sysport_free_cb(struct bcm_sysport_cb *cb) { dev_kfree_skb_any(cb->skb); @@ -463,67 +524,70 @@ static void bcm_sysport_free_cb(struct bcm_sysport_cb *cb) dma_unmap_addr_set(cb, dma_addr, 0); } -static int bcm_sysport_rx_refill(struct bcm_sysport_priv *priv, - struct bcm_sysport_cb *cb) +static struct sk_buff *bcm_sysport_rx_refill(struct bcm_sysport_priv *priv, + struct bcm_sysport_cb *cb) { struct device *kdev = &priv->pdev->dev; struct net_device *ndev = priv->netdev; + struct sk_buff *skb, *rx_skb; dma_addr_t mapping; - int ret; - cb->skb = netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH); - if (!cb->skb) { + /* Allocate a new SKB for a new packet */ + skb = netdev_alloc_skb(priv->netdev, RX_BUF_LENGTH); + if (!skb) { + priv->mib.alloc_rx_buff_failed++; netif_err(priv, rx_err, ndev, "SKB alloc failed\n"); - return -ENOMEM; + return NULL; } - mapping = dma_map_single(kdev, cb->skb->data, + mapping = dma_map_single(kdev, skb->data, RX_BUF_LENGTH, DMA_FROM_DEVICE); - ret = dma_mapping_error(kdev, mapping); - if (ret) { + if (dma_mapping_error(kdev, mapping)) { priv->mib.rx_dma_failed++; - bcm_sysport_free_cb(cb); + dev_kfree_skb_any(skb); netif_err(priv, rx_err, ndev, "DMA mapping failure\n"); - return ret; + return NULL; } - dma_unmap_addr_set(cb, dma_addr, mapping); - dma_desc_set_addr(priv, priv->rx_bd_assign_ptr, mapping); + /* Grab the current SKB on the ring */ + rx_skb = cb->skb; + if (likely(rx_skb)) + dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr), + RX_BUF_LENGTH, DMA_FROM_DEVICE); - priv->rx_bd_assign_index++; - priv->rx_bd_assign_index &= (priv->num_rx_bds - 1); - priv->rx_bd_assign_ptr = priv->rx_bds + - (priv->rx_bd_assign_index * DESC_SIZE); + /* Put the new SKB on the ring */ + cb->skb = skb; + dma_unmap_addr_set(cb, dma_addr, mapping); + dma_desc_set_addr(priv, cb->bd_addr, mapping); netif_dbg(priv, rx_status, ndev, "RX refill\n"); - return 0; + /* Return the current SKB to the caller */ + return rx_skb; } static int bcm_sysport_alloc_rx_bufs(struct bcm_sysport_priv *priv) { struct bcm_sysport_cb *cb; - int ret = 0; + struct sk_buff *skb; unsigned int i; for (i = 0; i < priv->num_rx_bds; i++) { - cb = &priv->rx_cbs[priv->rx_bd_assign_index]; - if (cb->skb) - continue; - - ret = bcm_sysport_rx_refill(priv, cb); - if (ret) - break; + cb = &priv->rx_cbs[i]; + skb = bcm_sysport_rx_refill(priv, cb); + if (skb) + dev_kfree_skb(skb); + if (!cb->skb) + return -ENOMEM; } - return ret; + return 0; } /* Poll the hardware for up to budget packets to process */ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, unsigned int budget) { - struct device *kdev = &priv->pdev->dev; struct net_device *ndev = priv->netdev; unsigned int processed = 0, to_process; struct bcm_sysport_cb *cb; @@ -531,7 +595,6 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, unsigned int p_index; u16 len, status; struct bcm_rsb *rsb; - int ret; /* Determine how much we should process since last call */ p_index = rdma_readl(priv, RDMA_PROD_INDEX); @@ -549,13 +612,8 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, while ((processed < to_process) && (processed < budget)) { cb = &priv->rx_cbs[priv->rx_read_ptr]; - skb = cb->skb; + skb = bcm_sysport_rx_refill(priv, cb); - processed++; - priv->rx_read_ptr++; - - if (priv->rx_read_ptr == priv->num_rx_bds) - priv->rx_read_ptr = 0; /* We do not have a backing SKB, so we do not a corresponding * DMA mapping for this incoming packet since @@ -566,12 +624,9 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, netif_err(priv, rx_err, ndev, "out of memory!\n"); ndev->stats.rx_dropped++; ndev->stats.rx_errors++; - goto refill; + goto next; } - dma_unmap_single(kdev, dma_unmap_addr(cb, dma_addr), - RX_BUF_LENGTH, DMA_FROM_DEVICE); - /* Extract the Receive Status Block prepended */ rsb = (struct bcm_rsb *)skb->data; len = (rsb->rx_status_len >> DESC_LEN_SHIFT) & DESC_LEN_MASK; @@ -583,12 +638,20 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, p_index, priv->rx_c_index, priv->rx_read_ptr, len, status); + if (unlikely(len > RX_BUF_LENGTH)) { + netif_err(priv, rx_status, ndev, "oversized packet\n"); + ndev->stats.rx_length_errors++; + ndev->stats.rx_errors++; + dev_kfree_skb_any(skb); + goto next; + } + if (unlikely(!(status & DESC_EOP) || !(status & DESC_SOP))) { netif_err(priv, rx_status, ndev, "fragmented packet!\n"); ndev->stats.rx_dropped++; ndev->stats.rx_errors++; - bcm_sysport_free_cb(cb); - goto refill; + dev_kfree_skb_any(skb); + goto next; } if (unlikely(status & (RX_STATUS_ERR | RX_STATUS_OVFLOW))) { @@ -597,8 +660,8 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, ndev->stats.rx_over_errors++; ndev->stats.rx_dropped++; ndev->stats.rx_errors++; - bcm_sysport_free_cb(cb); - goto refill; + dev_kfree_skb_any(skb); + goto next; } skb_put(skb, len); @@ -625,10 +688,12 @@ static unsigned int bcm_sysport_desc_rx(struct bcm_sysport_priv *priv, ndev->stats.rx_bytes += len; napi_gro_receive(&priv->napi, skb); -refill: - ret = bcm_sysport_rx_refill(priv, cb); - if (ret) - priv->mib.alloc_rx_buff_failed++; +next: + processed++; + priv->rx_read_ptr++; + + if (priv->rx_read_ptr == priv->num_rx_bds) + priv->rx_read_ptr = 0; } return processed; @@ -1269,14 +1334,14 @@ static inline int tdma_enable_set(struct bcm_sysport_priv *priv, static int bcm_sysport_init_rx_ring(struct bcm_sysport_priv *priv) { + struct bcm_sysport_cb *cb; u32 reg; int ret; + int i; /* Initialize SW view of the RX ring */ priv->num_rx_bds = NUM_RX_DESC; priv->rx_bds = priv->base + SYS_PORT_RDMA_OFFSET; - priv->rx_bd_assign_ptr = priv->rx_bds; - priv->rx_bd_assign_index = 0; priv->rx_c_index = 0; priv->rx_read_ptr = 0; priv->rx_cbs = kcalloc(priv->num_rx_bds, sizeof(struct bcm_sysport_cb), @@ -1286,6 +1351,11 @@ static int bcm_sysport_init_rx_ring(struct bcm_sysport_priv *priv) return -ENOMEM; } + for (i = 0; i < priv->num_rx_bds; i++) { + cb = priv->rx_cbs + i; + cb->bd_addr = priv->rx_bds + i * DESC_SIZE; + } + ret = bcm_sysport_alloc_rx_bufs(priv); if (ret) { netif_err(priv, hw, priv->netdev, "SKB allocation failed\n"); @@ -1641,6 +1711,8 @@ static struct ethtool_ops bcm_sysport_ethtool_ops = { .get_sset_count = bcm_sysport_get_sset_count, .get_wol = bcm_sysport_get_wol, .set_wol = bcm_sysport_set_wol, + .get_coalesce = bcm_sysport_get_coalesce, + .set_coalesce = bcm_sysport_set_coalesce, }; static const struct net_device_ops bcm_sysport_netdev_ops = { diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h index e2c043eabbf3..f28bf545d7f4 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.h +++ b/drivers/net/ethernet/broadcom/bcmsysport.h @@ -292,7 +292,7 @@ struct bcm_rsb { #define RDMA_END_ADDR_LO 0x102c #define RDMA_MBDONE_INTR 0x1030 -#define RDMA_INTR_THRESH_MASK 0xff +#define RDMA_INTR_THRESH_MASK 0x1ff #define RDMA_TIMEOUT_SHIFT 16 #define RDMA_TIMEOUT_MASK 0xffff @@ -663,8 +663,6 @@ struct bcm_sysport_priv { /* Receive queue */ void __iomem *rx_bds; - void __iomem *rx_bd_assign_ptr; - unsigned int rx_bd_assign_index; struct bcm_sysport_cb *rx_cbs; unsigned int num_rx_bds; unsigned int rx_read_ptr; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 1f82a04ce01a..7a4aaa3c01b6 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -357,6 +357,7 @@ struct sw_tx_bd { struct sw_rx_page { struct page *page; DEFINE_DMA_UNMAP_ADDR(mapping); + unsigned int offset; }; union db_prod { @@ -381,9 +382,10 @@ union db_prod { #define PAGES_PER_SGE_SHIFT 0 #define PAGES_PER_SGE (1 << PAGES_PER_SGE_SHIFT) -#define SGE_PAGE_SIZE PAGE_SIZE -#define SGE_PAGE_SHIFT PAGE_SHIFT -#define SGE_PAGE_ALIGN(addr) PAGE_ALIGN((typeof(PAGE_SIZE))(addr)) +#define SGE_PAGE_SHIFT 12 +#define SGE_PAGE_SIZE (1 << SGE_PAGE_SHIFT) +#define SGE_PAGE_MASK (~(SGE_PAGE_SIZE - 1)) +#define SGE_PAGE_ALIGN(addr) (((addr) + SGE_PAGE_SIZE - 1) & SGE_PAGE_MASK) #define SGE_PAGES (SGE_PAGE_SIZE * PAGES_PER_SGE) #define TPA_AGG_SIZE min_t(u32, (min_t(u32, 8, MAX_SKB_FRAGS) * \ SGE_PAGES), 0xffff) @@ -526,6 +528,12 @@ enum bnx2x_tpa_mode_t { TPA_MODE_GRO }; +struct bnx2x_alloc_pool { + struct page *page; + dma_addr_t dma; + unsigned int offset; +}; + struct bnx2x_fastpath { struct bnx2x *bp; /* parent */ @@ -599,6 +607,8 @@ struct bnx2x_fastpath { 4 (for the digits and to make it DWORD aligned) */ #define FP_NAME_SIZE (sizeof(((struct net_device *)0)->name) + 8) char name[FP_NAME_SIZE]; + + struct bnx2x_alloc_pool page_pool; }; #define bnx2x_fp(bp, nr, var) ((bp)->fp[(nr)].var) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index ec56a9b65dc3..e2a65334708d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -544,30 +544,49 @@ static void bnx2x_set_gro_params(struct sk_buff *skb, u16 parsing_flags, static int bnx2x_alloc_rx_sge(struct bnx2x *bp, struct bnx2x_fastpath *fp, u16 index, gfp_t gfp_mask) { - struct page *page = alloc_pages(gfp_mask, PAGES_PER_SGE_SHIFT); struct sw_rx_page *sw_buf = &fp->rx_page_ring[index]; struct eth_rx_sge *sge = &fp->rx_sge_ring[index]; + struct bnx2x_alloc_pool *pool = &fp->page_pool; dma_addr_t mapping; - if (unlikely(page == NULL)) { - BNX2X_ERR("Can't alloc sge\n"); - return -ENOMEM; - } + if (!pool->page || (PAGE_SIZE - pool->offset) < SGE_PAGE_SIZE) { - mapping = dma_map_page(&bp->pdev->dev, page, 0, - SGE_PAGES, DMA_FROM_DEVICE); - if (unlikely(dma_mapping_error(&bp->pdev->dev, mapping))) { - __free_pages(page, PAGES_PER_SGE_SHIFT); - BNX2X_ERR("Can't map sge\n"); - return -ENOMEM; + /* put page reference used by the memory pool, since we + * won't be using this page as the mempool anymore. + */ + if (pool->page) + put_page(pool->page); + + pool->page = alloc_pages(gfp_mask, PAGES_PER_SGE_SHIFT); + if (unlikely(!pool->page)) { + BNX2X_ERR("Can't alloc sge\n"); + return -ENOMEM; + } + + pool->dma = dma_map_page(&bp->pdev->dev, pool->page, 0, + PAGE_SIZE, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(&bp->pdev->dev, + pool->dma))) { + __free_pages(pool->page, PAGES_PER_SGE_SHIFT); + pool->page = NULL; + BNX2X_ERR("Can't map sge\n"); + return -ENOMEM; + } + pool->offset = 0; } - sw_buf->page = page; + get_page(pool->page); + sw_buf->page = pool->page; + sw_buf->offset = pool->offset; + + mapping = pool->dma + sw_buf->offset; dma_unmap_addr_set(sw_buf, mapping, mapping); sge->addr_hi = cpu_to_le32(U64_HI(mapping)); sge->addr_lo = cpu_to_le32(U64_LO(mapping)); + pool->offset += SGE_PAGE_SIZE; + return 0; } @@ -629,20 +648,22 @@ static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct bnx2x_fastpath *fp, return err; } - /* Unmap the page as we're going to pass it to the stack */ - dma_unmap_page(&bp->pdev->dev, - dma_unmap_addr(&old_rx_pg, mapping), - SGE_PAGES, DMA_FROM_DEVICE); + dma_unmap_single(&bp->pdev->dev, + dma_unmap_addr(&old_rx_pg, mapping), + SGE_PAGE_SIZE, DMA_FROM_DEVICE); /* Add one frag and update the appropriate fields in the skb */ if (fp->mode == TPA_MODE_LRO) - skb_fill_page_desc(skb, j, old_rx_pg.page, 0, frag_len); + skb_fill_page_desc(skb, j, old_rx_pg.page, + old_rx_pg.offset, frag_len); else { /* GRO */ int rem; int offset = 0; for (rem = frag_len; rem > 0; rem -= gro_size) { int len = rem > gro_size ? gro_size : rem; skb_fill_page_desc(skb, frag_id++, - old_rx_pg.page, offset, len); + old_rx_pg.page, + old_rx_pg.offset + offset, + len); if (offset) get_page(old_rx_pg.page); offset += len; @@ -662,7 +683,7 @@ static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct bnx2x_fastpath *fp, static void bnx2x_frag_free(const struct bnx2x_fastpath *fp, void *data) { if (fp->rx_frag_size) - put_page(virt_to_head_page(data)); + skb_free_frag(data); else kfree(data); } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index d7a71758e876..2b30081ec26d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -804,9 +804,13 @@ static inline void bnx2x_free_rx_sge(struct bnx2x *bp, if (!page) return; - dma_unmap_page(&bp->pdev->dev, dma_unmap_addr(sw_buf, mapping), - SGE_PAGES, DMA_FROM_DEVICE); - __free_pages(page, PAGES_PER_SGE_SHIFT); + /* Since many fragments can share the same page, make sure to + * only unmap and free the page once. + */ + dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(sw_buf, mapping), + SGE_PAGE_SIZE, DMA_FROM_DEVICE); + + put_page(page); sw_buf->page = NULL; sge->addr_hi = 0; @@ -964,6 +968,25 @@ static inline void bnx2x_set_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid, ((u8 *)fw_lo)[1] = mac[4]; } +static inline void bnx2x_free_rx_mem_pool(struct bnx2x *bp, + struct bnx2x_alloc_pool *pool) +{ + if (!pool->page) + return; + + /* Page was not fully fragmented. Unmap unused space */ + if (pool->offset < PAGE_SIZE) { + dma_addr_t dma = pool->dma + pool->offset; + int size = PAGE_SIZE - pool->offset; + + dma_unmap_single(&bp->pdev->dev, dma, size, DMA_FROM_DEVICE); + } + + put_page(pool->page); + + pool->page = NULL; +} + static inline void bnx2x_free_rx_sge_range(struct bnx2x *bp, struct bnx2x_fastpath *fp, int last) { @@ -974,6 +997,8 @@ static inline void bnx2x_free_rx_sge_range(struct bnx2x *bp, for (i = 0; i < last; i++) bnx2x_free_rx_sge(bp, fp, i); + + bnx2x_free_rx_mem_pool(bp, &fp->page_pool); } static inline void bnx2x_set_next_page_rx_bd(struct bnx2x_fastpath *fp) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 069952fa5d64..73c934cf6c61 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -6618,7 +6618,7 @@ static void tg3_tx(struct tg3_napi *tnapi) static void tg3_frag_free(bool is_frag, void *data) { if (is_frag) - put_page(virt_to_head_page(data)); + skb_free_frag(data); else kfree(data); } diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index fc646a41d548..740d04fd2223 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -54,6 +54,8 @@ #define MACB_MAX_TX_LEN ((unsigned int)((1 << MACB_TX_FRMLEN_SIZE) - 1)) #define GEM_MAX_TX_LEN ((unsigned int)((1 << GEM_TX_FRMLEN_SIZE) - 1)) +#define GEM_MTU_MIN_SIZE 68 + /* * Graceful stop timeouts in us. We should allow up to * 1 frame time (10 Mbits/s, full-duplex, ignoring collisions) @@ -785,7 +787,7 @@ static int gem_rx(struct macb *bp, int budget) } /* now everything is ready for receiving packet */ bp->rx_skbuff[entry] = NULL; - len = MACB_BFEXT(RX_FRMLEN, ctrl); + len = ctrl & bp->rx_frm_len_mask; netdev_vdbg(bp->dev, "gem_rx %u (len %u)\n", entry, len); @@ -831,7 +833,7 @@ static int macb_rx_frame(struct macb *bp, unsigned int first_frag, struct macb_dma_desc *desc; desc = macb_rx_desc(bp, last_frag); - len = MACB_BFEXT(RX_FRMLEN, desc->ctrl); + len = desc->ctrl & bp->rx_frm_len_mask; netdev_vdbg(bp->dev, "macb_rx_frame frags %u - %u (len %u)\n", macb_rx_ring_wrap(first_frag), @@ -1651,7 +1653,10 @@ static void macb_init_hw(struct macb *bp) config |= MACB_BF(RBOF, NET_IP_ALIGN); /* Make eth data aligned */ config |= MACB_BIT(PAE); /* PAuse Enable */ config |= MACB_BIT(DRFCS); /* Discard Rx FCS */ - config |= MACB_BIT(BIG); /* Receive oversized frames */ + if (bp->caps & MACB_CAPS_JUMBO) + config |= MACB_BIT(JFRAME); /* Enable jumbo frames */ + else + config |= MACB_BIT(BIG); /* Receive oversized frames */ if (bp->dev->flags & IFF_PROMISC) config |= MACB_BIT(CAF); /* Copy All Frames */ else if (macb_is_gem(bp) && bp->dev->features & NETIF_F_RXCSUM) @@ -1660,8 +1665,13 @@ static void macb_init_hw(struct macb *bp) config |= MACB_BIT(NBC); /* No BroadCast */ config |= macb_dbw(bp); macb_writel(bp, NCFGR, config); + if ((bp->caps & MACB_CAPS_JUMBO) && bp->jumbo_max_len) + gem_writel(bp, JML, bp->jumbo_max_len); bp->speed = SPEED_10; bp->duplex = DUPLEX_HALF; + bp->rx_frm_len_mask = MACB_RX_FRMLEN_MASK; + if (bp->caps & MACB_CAPS_JUMBO) + bp->rx_frm_len_mask = MACB_RX_JFRMLEN_MASK; macb_configure_dma(bp); @@ -1865,6 +1875,26 @@ static int macb_close(struct net_device *dev) return 0; } +static int macb_change_mtu(struct net_device *dev, int new_mtu) +{ + struct macb *bp = netdev_priv(dev); + u32 max_mtu; + + if (netif_running(dev)) + return -EBUSY; + + max_mtu = ETH_DATA_LEN; + if (bp->caps & MACB_CAPS_JUMBO) + max_mtu = gem_readl(bp, JML) - ETH_HLEN - ETH_FCS_LEN; + + if ((new_mtu > max_mtu) || (new_mtu < GEM_MTU_MIN_SIZE)) + return -EINVAL; + + dev->mtu = new_mtu; + + return 0; +} + static void gem_update_stats(struct macb *bp) { int i; @@ -2141,7 +2171,7 @@ static const struct net_device_ops macb_netdev_ops = { .ndo_get_stats = macb_get_stats, .ndo_do_ioctl = macb_ioctl, .ndo_validate_addr = eth_validate_addr, - .ndo_change_mtu = eth_change_mtu, + .ndo_change_mtu = macb_change_mtu, .ndo_set_mac_address = eth_mac_addr, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = macb_poll_controller, @@ -2702,6 +2732,16 @@ static const struct macb_config emac_config = { .init = at91ether_init, }; + +static const struct macb_config zynqmp_config = { + .caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE | + MACB_CAPS_JUMBO, + .dma_burst_length = 16, + .clk_init = macb_clk_init, + .init = macb_init, + .jumbo_max_len = 10240, +}; + static const struct macb_config zynq_config = { .caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_NO_GIGABIT_HALF, @@ -2720,6 +2760,7 @@ static const struct of_device_id macb_dt_ids[] = { { .compatible = "atmel,sama5d4-gem", .data = &sama5d4_config }, { .compatible = "cdns,at91rm9200-emac", .data = &emac_config }, { .compatible = "cdns,emac", .data = &emac_config }, + { .compatible = "cdns,zynqmp-gem", .data = &zynqmp_config}, { .compatible = "cdns,zynq-gem", .data = &zynq_config }, { /* sentinel */ } }; @@ -2789,6 +2830,10 @@ static int macb_probe(struct platform_device *pdev) bp->pclk = pclk; bp->hclk = hclk; bp->tx_clk = tx_clk; + if (macb_config->jumbo_max_len) { + bp->jumbo_max_len = macb_config->jumbo_max_len; + } + spin_lock_init(&bp->lock); /* setup capabilities */ diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 24b1d9bcd865..d74655993d4b 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -71,6 +71,7 @@ #define GEM_NCFGR 0x0004 /* Network Config */ #define GEM_USRIO 0x000c /* User IO */ #define GEM_DMACFG 0x0010 /* DMA Configuration */ +#define GEM_JML 0x0048 /* Jumbo Max Length */ #define GEM_HRB 0x0080 /* Hash Bottom */ #define GEM_HRT 0x0084 /* Hash Top */ #define GEM_SA1B 0x0088 /* Specific1 Bottom */ @@ -398,6 +399,7 @@ #define MACB_CAPS_GIGABIT_MODE_AVAILABLE 0x20000000 #define MACB_CAPS_SG_DISABLED 0x40000000 #define MACB_CAPS_MACB_IS_GEM 0x80000000 +#define MACB_CAPS_JUMBO 0x00000008 /* Bit manipulation macros */ #define MACB_BIT(name) \ @@ -515,6 +517,9 @@ struct macb_dma_desc { #define MACB_RX_BROADCAST_OFFSET 31 #define MACB_RX_BROADCAST_SIZE 1 +#define MACB_RX_FRMLEN_MASK 0xFFF +#define MACB_RX_JFRMLEN_MASK 0x3FFF + /* RX checksum offload disabled: bit 24 clear in NCFGR */ #define GEM_RX_TYPEID_MATCH_OFFSET 22 #define GEM_RX_TYPEID_MATCH_SIZE 2 @@ -758,6 +763,7 @@ struct macb_config { int (*clk_init)(struct platform_device *pdev, struct clk **pclk, struct clk **hclk, struct clk **tx_clk); int (*init)(struct platform_device *pdev); + int jumbo_max_len; }; struct macb_queue { @@ -827,6 +833,9 @@ struct macb { unsigned int max_tx_length; u64 ethtool_stats[GEM_STATS_LEN]; + + unsigned int rx_frm_len_mask; + unsigned int jumbo_max_len; }; static inline bool macb_is_gem(struct macb *bp) diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig new file mode 100644 index 000000000000..fc3d8e3ee807 --- /dev/null +++ b/drivers/net/ethernet/cavium/Kconfig @@ -0,0 +1,40 @@ +# +# Cavium ethernet device configuration +# + +config NET_VENDOR_CAVIUM + tristate "Cavium ethernet drivers" + depends on PCI && 64BIT + ---help--- + Enable support for the Cavium ThunderX Network Interface + Controller (NIC). The NIC provides the controller and DMA + engines to move network traffic to/from the memory. The NIC + works closely with TNS, BGX and SerDes to implement the + functions replacing and virtualizing those of a typical + standalone PCIe NIC chip. + + If you have a Cavium Thunder board, say Y. + +if NET_VENDOR_CAVIUM + +config THUNDER_NIC_PF + tristate "Thunder Physical function driver" + default NET_VENDOR_CAVIUM + select THUNDER_NIC_BGX + ---help--- + This driver supports Thunder's NIC physical function. + +config THUNDER_NIC_VF + tristate "Thunder Virtual function driver" + default NET_VENDOR_CAVIUM + ---help--- + This driver supports Thunder's NIC virtual function + +config THUNDER_NIC_BGX + tristate "Thunder MAC interface driver (BGX)" + default NET_VENDOR_CAVIUM + ---help--- + This driver supports programming and controlling of MAC + interface from NIC physical function driver. + +endif # NET_VENDOR_CAVIUM diff --git a/drivers/net/ethernet/cavium/Makefile b/drivers/net/ethernet/cavium/Makefile new file mode 100644 index 000000000000..7aac4780d050 --- /dev/null +++ b/drivers/net/ethernet/cavium/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the Cavium ethernet device drivers. +# + +obj-$(CONFIG_NET_VENDOR_CAVIUM) += thunder/ diff --git a/drivers/net/ethernet/cavium/thunder/Makefile b/drivers/net/ethernet/cavium/thunder/Makefile new file mode 100644 index 000000000000..5c4615ccaa14 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/Makefile @@ -0,0 +1,11 @@ +# +# Makefile for Cavium's Thunder ethernet device +# + +obj-$(CONFIG_THUNDER_NIC_BGX) += thunder_bgx.o +obj-$(CONFIG_THUNDER_NIC_PF) += nicpf.o +obj-$(CONFIG_THUNDER_NIC_VF) += nicvf.o + +nicpf-y := nic_main.o +nicvf-y := nicvf_main.o nicvf_queues.o +nicvf-y += nicvf_ethtool.o diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h new file mode 100644 index 000000000000..9b0be527909b --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nic.h @@ -0,0 +1,414 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#ifndef NIC_H +#define NIC_H + +#include <linux/netdevice.h> +#include <linux/interrupt.h> +#include "thunder_bgx.h" + +/* PCI device IDs */ +#define PCI_DEVICE_ID_THUNDER_NIC_PF 0xA01E +#define PCI_DEVICE_ID_THUNDER_PASS1_NIC_VF 0x0011 +#define PCI_DEVICE_ID_THUNDER_NIC_VF 0xA034 +#define PCI_DEVICE_ID_THUNDER_BGX 0xA026 + +/* PCI BAR nos */ +#define PCI_CFG_REG_BAR_NUM 0 +#define PCI_MSIX_REG_BAR_NUM 4 + +/* NIC SRIOV VF count */ +#define MAX_NUM_VFS_SUPPORTED 128 +#define DEFAULT_NUM_VF_ENABLED 8 + +#define NIC_TNS_BYPASS_MODE 0 +#define NIC_TNS_MODE 1 + +/* NIC priv flags */ +#define NIC_SRIOV_ENABLED BIT(0) + +/* Min/Max packet size */ +#define NIC_HW_MIN_FRS 64 +#define NIC_HW_MAX_FRS 9200 /* 9216 max packet including FCS */ + +/* Max pkinds */ +#define NIC_MAX_PKIND 16 + +/* Rx Channels */ +/* Receive channel configuration in TNS bypass mode + * Below is configuration in TNS bypass mode + * BGX0-LMAC0-CHAN0 - VNIC CHAN0 + * BGX0-LMAC1-CHAN0 - VNIC CHAN16 + * ... + * BGX1-LMAC0-CHAN0 - VNIC CHAN128 + * ... + * BGX1-LMAC3-CHAN0 - VNIC CHAN174 + */ +#define NIC_INTF_COUNT 2 /* Interfaces btw VNIC and TNS/BGX */ +#define NIC_CHANS_PER_INF 128 +#define NIC_MAX_CHANS (NIC_INTF_COUNT * NIC_CHANS_PER_INF) +#define NIC_CPI_COUNT 2048 /* No of channel parse indices */ + +/* TNS bypass mode: 1-1 mapping between VNIC and BGX:LMAC */ +#define NIC_MAX_BGX MAX_BGX_PER_CN88XX +#define NIC_CPI_PER_BGX (NIC_CPI_COUNT / NIC_MAX_BGX) +#define NIC_MAX_CPI_PER_LMAC 64 /* Max when CPI_ALG is IP diffserv */ +#define NIC_RSSI_PER_BGX (NIC_RSSI_COUNT / NIC_MAX_BGX) + +/* Tx scheduling */ +#define NIC_MAX_TL4 1024 +#define NIC_MAX_TL4_SHAPERS 256 /* 1 shaper for 4 TL4s */ +#define NIC_MAX_TL3 256 +#define NIC_MAX_TL3_SHAPERS 64 /* 1 shaper for 4 TL3s */ +#define NIC_MAX_TL2 64 +#define NIC_MAX_TL2_SHAPERS 2 /* 1 shaper for 32 TL2s */ +#define NIC_MAX_TL1 2 + +/* TNS bypass mode */ +#define NIC_TL2_PER_BGX 32 +#define NIC_TL4_PER_BGX (NIC_MAX_TL4 / NIC_MAX_BGX) +#define NIC_TL4_PER_LMAC (NIC_MAX_TL4 / NIC_CHANS_PER_INF) + +/* NIC VF Interrupts */ +#define NICVF_INTR_CQ 0 +#define NICVF_INTR_SQ 1 +#define NICVF_INTR_RBDR 2 +#define NICVF_INTR_PKT_DROP 3 +#define NICVF_INTR_TCP_TIMER 4 +#define NICVF_INTR_MBOX 5 +#define NICVF_INTR_QS_ERR 6 + +#define NICVF_INTR_CQ_SHIFT 0 +#define NICVF_INTR_SQ_SHIFT 8 +#define NICVF_INTR_RBDR_SHIFT 16 +#define NICVF_INTR_PKT_DROP_SHIFT 20 +#define NICVF_INTR_TCP_TIMER_SHIFT 21 +#define NICVF_INTR_MBOX_SHIFT 22 +#define NICVF_INTR_QS_ERR_SHIFT 23 + +#define NICVF_INTR_CQ_MASK (0xFF << NICVF_INTR_CQ_SHIFT) +#define NICVF_INTR_SQ_MASK (0xFF << NICVF_INTR_SQ_SHIFT) +#define NICVF_INTR_RBDR_MASK (0x03 << NICVF_INTR_RBDR_SHIFT) +#define NICVF_INTR_PKT_DROP_MASK BIT(NICVF_INTR_PKT_DROP_SHIFT) +#define NICVF_INTR_TCP_TIMER_MASK BIT(NICVF_INTR_TCP_TIMER_SHIFT) +#define NICVF_INTR_MBOX_MASK BIT(NICVF_INTR_MBOX_SHIFT) +#define NICVF_INTR_QS_ERR_MASK BIT(NICVF_INTR_QS_ERR_SHIFT) + +/* MSI-X interrupts */ +#define NIC_PF_MSIX_VECTORS 10 +#define NIC_VF_MSIX_VECTORS 20 + +#define NIC_PF_INTR_ID_ECC0_SBE 0 +#define NIC_PF_INTR_ID_ECC0_DBE 1 +#define NIC_PF_INTR_ID_ECC1_SBE 2 +#define NIC_PF_INTR_ID_ECC1_DBE 3 +#define NIC_PF_INTR_ID_ECC2_SBE 4 +#define NIC_PF_INTR_ID_ECC2_DBE 5 +#define NIC_PF_INTR_ID_ECC3_SBE 6 +#define NIC_PF_INTR_ID_ECC3_DBE 7 +#define NIC_PF_INTR_ID_MBOX0 8 +#define NIC_PF_INTR_ID_MBOX1 9 + +/* Global timer for CQ timer thresh interrupts + * Calculated for SCLK of 700Mhz + * value written should be a 1/16th of what is expected + * + * 1 tick per 0.05usec = value of 2.2 + * This 10% would be covered in CQ timer thresh value + */ +#define NICPF_CLK_PER_INT_TICK 2 + +struct nicvf_cq_poll { + u8 cq_idx; /* Completion queue index */ + struct napi_struct napi; +}; + +#define NIC_RSSI_COUNT 4096 /* Total no of RSS indices */ +#define NIC_MAX_RSS_HASH_BITS 8 +#define NIC_MAX_RSS_IDR_TBL_SIZE (1 << NIC_MAX_RSS_HASH_BITS) +#define RSS_HASH_KEY_SIZE 5 /* 320 bit key */ + +struct nicvf_rss_info { + bool enable; +#define RSS_L2_EXTENDED_HASH_ENA BIT(0) +#define RSS_IP_HASH_ENA BIT(1) +#define RSS_TCP_HASH_ENA BIT(2) +#define RSS_TCP_SYN_DIS BIT(3) +#define RSS_UDP_HASH_ENA BIT(4) +#define RSS_L4_EXTENDED_HASH_ENA BIT(5) +#define RSS_ROCE_ENA BIT(6) +#define RSS_L3_BI_DIRECTION_ENA BIT(7) +#define RSS_L4_BI_DIRECTION_ENA BIT(8) + u64 cfg; + u8 hash_bits; + u16 rss_size; + u8 ind_tbl[NIC_MAX_RSS_IDR_TBL_SIZE]; + u64 key[RSS_HASH_KEY_SIZE]; +} ____cacheline_aligned_in_smp; + +enum rx_stats_reg_offset { + RX_OCTS = 0x0, + RX_UCAST = 0x1, + RX_BCAST = 0x2, + RX_MCAST = 0x3, + RX_RED = 0x4, + RX_RED_OCTS = 0x5, + RX_ORUN = 0x6, + RX_ORUN_OCTS = 0x7, + RX_FCS = 0x8, + RX_L2ERR = 0x9, + RX_DRP_BCAST = 0xa, + RX_DRP_MCAST = 0xb, + RX_DRP_L3BCAST = 0xc, + RX_DRP_L3MCAST = 0xd, + RX_STATS_ENUM_LAST, +}; + +enum tx_stats_reg_offset { + TX_OCTS = 0x0, + TX_UCAST = 0x1, + TX_BCAST = 0x2, + TX_MCAST = 0x3, + TX_DROP = 0x4, + TX_STATS_ENUM_LAST, +}; + +struct nicvf_hw_stats { + u64 rx_bytes_ok; + u64 rx_ucast_frames_ok; + u64 rx_bcast_frames_ok; + u64 rx_mcast_frames_ok; + u64 rx_fcs_errors; + u64 rx_l2_errors; + u64 rx_drop_red; + u64 rx_drop_red_bytes; + u64 rx_drop_overrun; + u64 rx_drop_overrun_bytes; + u64 rx_drop_bcast; + u64 rx_drop_mcast; + u64 rx_drop_l3_bcast; + u64 rx_drop_l3_mcast; + u64 tx_bytes_ok; + u64 tx_ucast_frames_ok; + u64 tx_bcast_frames_ok; + u64 tx_mcast_frames_ok; + u64 tx_drops; +}; + +struct nicvf_drv_stats { + /* Rx */ + u64 rx_frames_ok; + u64 rx_frames_64; + u64 rx_frames_127; + u64 rx_frames_255; + u64 rx_frames_511; + u64 rx_frames_1023; + u64 rx_frames_1518; + u64 rx_frames_jumbo; + u64 rx_drops; + /* Tx */ + u64 tx_frames_ok; + u64 tx_drops; + u64 tx_busy; + u64 tx_tso; +}; + +struct nicvf { + struct net_device *netdev; + struct pci_dev *pdev; + u8 vf_id; + u8 node; + u8 tns_mode; + u16 mtu; + struct queue_set *qs; + void __iomem *reg_base; + bool link_up; + u8 duplex; + u32 speed; + struct page *rb_page; + u32 rb_page_offset; + bool rb_alloc_fail; + bool rb_work_scheduled; + struct delayed_work rbdr_work; + struct tasklet_struct rbdr_task; + struct tasklet_struct qs_err_task; + struct tasklet_struct cq_task; + struct nicvf_cq_poll *napi[8]; + struct nicvf_rss_info rss_info; + u8 cpi_alg; + /* Interrupt coalescing settings */ + u32 cq_coalesce_usecs; + + u32 msg_enable; + struct nicvf_hw_stats stats; + struct nicvf_drv_stats drv_stats; + struct bgx_stats bgx_stats; + struct work_struct reset_task; + + /* MSI-X */ + bool msix_enabled; + u8 num_vec; + struct msix_entry msix_entries[NIC_VF_MSIX_VECTORS]; + char irq_name[NIC_VF_MSIX_VECTORS][20]; + bool irq_allocated[NIC_VF_MSIX_VECTORS]; + + bool pf_ready_to_rcv_msg; + bool pf_acked; + bool pf_nacked; + bool bgx_stats_acked; +} ____cacheline_aligned_in_smp; + +/* PF <--> VF Mailbox communication + * Eight 64bit registers are shared between PF and VF. + * Separate set for each VF. + * Writing '1' into last register mbx7 means end of message. + */ + +/* PF <--> VF mailbox communication */ +#define NIC_PF_VF_MAILBOX_SIZE 2 +#define NIC_MBOX_MSG_TIMEOUT 2000 /* ms */ + +/* Mailbox message types */ +#define NIC_MBOX_MSG_READY 0x01 /* Is PF ready to rcv msgs */ +#define NIC_MBOX_MSG_ACK 0x02 /* ACK the message received */ +#define NIC_MBOX_MSG_NACK 0x03 /* NACK the message received */ +#define NIC_MBOX_MSG_QS_CFG 0x04 /* Configure Qset */ +#define NIC_MBOX_MSG_RQ_CFG 0x05 /* Configure receive queue */ +#define NIC_MBOX_MSG_SQ_CFG 0x06 /* Configure Send queue */ +#define NIC_MBOX_MSG_RQ_DROP_CFG 0x07 /* Configure receive queue */ +#define NIC_MBOX_MSG_SET_MAC 0x08 /* Add MAC ID to DMAC filter */ +#define NIC_MBOX_MSG_SET_MAX_FRS 0x09 /* Set max frame size */ +#define NIC_MBOX_MSG_CPI_CFG 0x0A /* Config CPI, RSSI */ +#define NIC_MBOX_MSG_RSS_SIZE 0x0B /* Get RSS indir_tbl size */ +#define NIC_MBOX_MSG_RSS_CFG 0x0C /* Config RSS table */ +#define NIC_MBOX_MSG_RSS_CFG_CONT 0x0D /* RSS config continuation */ +#define NIC_MBOX_MSG_RQ_BP_CFG 0x0E /* RQ backpressure config */ +#define NIC_MBOX_MSG_RQ_SW_SYNC 0x0F /* Flush inflight pkts to RQ */ +#define NIC_MBOX_MSG_BGX_STATS 0x10 /* Get stats from BGX */ +#define NIC_MBOX_MSG_BGX_LINK_CHANGE 0x11 /* BGX:LMAC link status */ +#define NIC_MBOX_MSG_CFG_DONE 0x12 /* VF configuration done */ +#define NIC_MBOX_MSG_SHUTDOWN 0x13 /* VF is being shutdown */ + +struct nic_cfg_msg { + u8 msg; + u8 vf_id; + u8 tns_mode; + u8 node_id; + u64 mac_addr; +}; + +/* Qset configuration */ +struct qs_cfg_msg { + u8 msg; + u8 num; + u64 cfg; +}; + +/* Receive queue configuration */ +struct rq_cfg_msg { + u8 msg; + u8 qs_num; + u8 rq_num; + u64 cfg; +}; + +/* Send queue configuration */ +struct sq_cfg_msg { + u8 msg; + u8 qs_num; + u8 sq_num; + u64 cfg; +}; + +/* Set VF's MAC address */ +struct set_mac_msg { + u8 msg; + u8 vf_id; + u64 addr; +}; + +/* Set Maximum frame size */ +struct set_frs_msg { + u8 msg; + u8 vf_id; + u16 max_frs; +}; + +/* Set CPI algorithm type */ +struct cpi_cfg_msg { + u8 msg; + u8 vf_id; + u8 rq_cnt; + u8 cpi_alg; +}; + +/* Get RSS table size */ +struct rss_sz_msg { + u8 msg; + u8 vf_id; + u16 ind_tbl_size; +}; + +/* Set RSS configuration */ +struct rss_cfg_msg { + u8 msg; + u8 vf_id; + u8 hash_bits; + u8 tbl_len; + u8 tbl_offset; +#define RSS_IND_TBL_LEN_PER_MBX_MSG 8 + u8 ind_tbl[RSS_IND_TBL_LEN_PER_MBX_MSG]; +}; + +struct bgx_stats_msg { + u8 msg; + u8 vf_id; + u8 rx; + u8 idx; + u64 stats; +}; + +/* Physical interface link status */ +struct bgx_link_status { + u8 msg; + u8 link_up; + u8 duplex; + u32 speed; +}; + +/* 128 bit shared memory between PF and each VF */ +union nic_mbx { + struct { u8 msg; } msg; + struct nic_cfg_msg nic_cfg; + struct qs_cfg_msg qs; + struct rq_cfg_msg rq; + struct sq_cfg_msg sq; + struct set_mac_msg mac; + struct set_frs_msg frs; + struct cpi_cfg_msg cpi_cfg; + struct rss_sz_msg rss_size; + struct rss_cfg_msg rss_cfg; + struct bgx_stats_msg bgx_stats; + struct bgx_link_status link_status; +}; + +int nicvf_set_real_num_queues(struct net_device *netdev, + int tx_queues, int rx_queues); +int nicvf_open(struct net_device *netdev); +int nicvf_stop(struct net_device *netdev); +int nicvf_send_msg_to_pf(struct nicvf *vf, union nic_mbx *mbx); +void nicvf_config_cpi(struct nicvf *nic); +void nicvf_config_rss(struct nicvf *nic); +void nicvf_set_rss_key(struct nicvf *nic); +void nicvf_free_skb(struct nicvf *nic, struct sk_buff *skb); +void nicvf_set_ethtool_ops(struct net_device *netdev); +void nicvf_update_stats(struct nicvf *nic); +void nicvf_update_lmac_stats(struct nicvf *nic); + +#endif /* NIC_H */ diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c new file mode 100644 index 000000000000..0f1f58b54bf1 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nic_main.c @@ -0,0 +1,940 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <linux/etherdevice.h> +#include <linux/of.h> + +#include "nic_reg.h" +#include "nic.h" +#include "q_struct.h" +#include "thunder_bgx.h" + +#define DRV_NAME "thunder-nic" +#define DRV_VERSION "1.0" + +struct nicpf { + struct pci_dev *pdev; + u8 rev_id; +#define NIC_NODE_ID_MASK 0x300000000000 +#define NIC_NODE_ID(x) ((x & NODE_ID_MASK) >> 44) + u8 node; + unsigned int flags; + u8 num_vf_en; /* No of VF enabled */ + bool vf_enabled[MAX_NUM_VFS_SUPPORTED]; + void __iomem *reg_base; /* Register start address */ + struct pkind_cfg pkind; +#define NIC_SET_VF_LMAC_MAP(bgx, lmac) (((bgx & 0xF) << 4) | (lmac & 0xF)) +#define NIC_GET_BGX_FROM_VF_LMAC_MAP(map) ((map >> 4) & 0xF) +#define NIC_GET_LMAC_FROM_VF_LMAC_MAP(map) (map & 0xF) + u8 vf_lmac_map[MAX_LMAC]; + struct delayed_work dwork; + struct workqueue_struct *check_link; + u8 link[MAX_LMAC]; + u8 duplex[MAX_LMAC]; + u32 speed[MAX_LMAC]; + u16 cpi_base[MAX_NUM_VFS_SUPPORTED]; + u16 rss_ind_tbl_size; + bool mbx_lock[MAX_NUM_VFS_SUPPORTED]; + + /* MSI-X */ + bool msix_enabled; + u8 num_vec; + struct msix_entry msix_entries[NIC_PF_MSIX_VECTORS]; + bool irq_allocated[NIC_PF_MSIX_VECTORS]; +}; + +/* Supported devices */ +static const struct pci_device_id nic_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_NIC_PF) }, + { 0, } /* end of table */ +}; + +MODULE_AUTHOR("Sunil Goutham"); +MODULE_DESCRIPTION("Cavium Thunder NIC Physical Function Driver"); +MODULE_LICENSE("GPL v2"); +MODULE_VERSION(DRV_VERSION); +MODULE_DEVICE_TABLE(pci, nic_id_table); + +/* The Cavium ThunderX network controller can *only* be found in SoCs + * containing the ThunderX ARM64 CPU implementation. All accesses to the device + * registers on this platform are implicitly strongly ordered with respect + * to memory accesses. So writeq_relaxed() and readq_relaxed() are safe to use + * with no memory barriers in this driver. The readq()/writeq() functions add + * explicit ordering operation which in this case are redundant, and only + * add overhead. + */ + +/* Register read/write APIs */ +static void nic_reg_write(struct nicpf *nic, u64 offset, u64 val) +{ + writeq_relaxed(val, nic->reg_base + offset); +} + +static u64 nic_reg_read(struct nicpf *nic, u64 offset) +{ + return readq_relaxed(nic->reg_base + offset); +} + +/* PF -> VF mailbox communication APIs */ +static void nic_enable_mbx_intr(struct nicpf *nic) +{ + /* Enable mailbox interrupt for all 128 VFs */ + nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S, ~0ull); + nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S + sizeof(u64), ~0ull); +} + +static void nic_clear_mbx_intr(struct nicpf *nic, int vf, int mbx_reg) +{ + nic_reg_write(nic, NIC_PF_MAILBOX_INT + (mbx_reg << 3), BIT_ULL(vf)); +} + +static u64 nic_get_mbx_addr(int vf) +{ + return NIC_PF_VF_0_127_MAILBOX_0_1 + (vf << NIC_VF_NUM_SHIFT); +} + +/* Send a mailbox message to VF + * @vf: vf to which this message to be sent + * @mbx: Message to be sent + */ +static void nic_send_msg_to_vf(struct nicpf *nic, int vf, union nic_mbx *mbx) +{ + void __iomem *mbx_addr = nic->reg_base + nic_get_mbx_addr(vf); + u64 *msg = (u64 *)mbx; + + /* In first revision HW, mbox interrupt is triggerred + * when PF writes to MBOX(1), in next revisions when + * PF writes to MBOX(0) + */ + if (nic->rev_id == 0) { + /* see the comment for nic_reg_write()/nic_reg_read() + * functions above + */ + writeq_relaxed(msg[0], mbx_addr); + writeq_relaxed(msg[1], mbx_addr + 8); + } else { + writeq_relaxed(msg[1], mbx_addr + 8); + writeq_relaxed(msg[0], mbx_addr); + } +} + +/* Responds to VF's READY message with VF's + * ID, node, MAC address e.t.c + * @vf: VF which sent READY message + */ +static void nic_mbx_send_ready(struct nicpf *nic, int vf) +{ + union nic_mbx mbx = {}; + int bgx_idx, lmac; + const char *mac; + + mbx.nic_cfg.msg = NIC_MBOX_MSG_READY; + mbx.nic_cfg.vf_id = vf; + + mbx.nic_cfg.tns_mode = NIC_TNS_BYPASS_MODE; + + bgx_idx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); + + mac = bgx_get_lmac_mac(nic->node, bgx_idx, lmac); + if (mac) + ether_addr_copy((u8 *)&mbx.nic_cfg.mac_addr, mac); + + mbx.nic_cfg.node_id = nic->node; + nic_send_msg_to_vf(nic, vf, &mbx); +} + +/* ACKs VF's mailbox message + * @vf: VF to which ACK to be sent + */ +static void nic_mbx_send_ack(struct nicpf *nic, int vf) +{ + union nic_mbx mbx = {}; + + mbx.msg.msg = NIC_MBOX_MSG_ACK; + nic_send_msg_to_vf(nic, vf, &mbx); +} + +/* NACKs VF's mailbox message that PF is not able to + * complete the action + * @vf: VF to which ACK to be sent + */ +static void nic_mbx_send_nack(struct nicpf *nic, int vf) +{ + union nic_mbx mbx = {}; + + mbx.msg.msg = NIC_MBOX_MSG_NACK; + nic_send_msg_to_vf(nic, vf, &mbx); +} + +/* Flush all in flight receive packets to memory and + * bring down an active RQ + */ +static int nic_rcv_queue_sw_sync(struct nicpf *nic) +{ + u16 timeout = ~0x00; + + nic_reg_write(nic, NIC_PF_SW_SYNC_RX, 0x01); + /* Wait till sync cycle is finished */ + while (timeout) { + if (nic_reg_read(nic, NIC_PF_SW_SYNC_RX_DONE) & 0x1) + break; + timeout--; + } + nic_reg_write(nic, NIC_PF_SW_SYNC_RX, 0x00); + if (!timeout) { + dev_err(&nic->pdev->dev, "Receive queue software sync failed"); + return 1; + } + return 0; +} + +/* Get BGX Rx/Tx stats and respond to VF's request */ +static void nic_get_bgx_stats(struct nicpf *nic, struct bgx_stats_msg *bgx) +{ + int bgx_idx, lmac; + union nic_mbx mbx = {}; + + bgx_idx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[bgx->vf_id]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[bgx->vf_id]); + + mbx.bgx_stats.msg = NIC_MBOX_MSG_BGX_STATS; + mbx.bgx_stats.vf_id = bgx->vf_id; + mbx.bgx_stats.rx = bgx->rx; + mbx.bgx_stats.idx = bgx->idx; + if (bgx->rx) + mbx.bgx_stats.stats = bgx_get_rx_stats(nic->node, bgx_idx, + lmac, bgx->idx); + else + mbx.bgx_stats.stats = bgx_get_tx_stats(nic->node, bgx_idx, + lmac, bgx->idx); + nic_send_msg_to_vf(nic, bgx->vf_id, &mbx); +} + +/* Update hardware min/max frame size */ +static int nic_update_hw_frs(struct nicpf *nic, int new_frs, int vf) +{ + if ((new_frs > NIC_HW_MAX_FRS) || (new_frs < NIC_HW_MIN_FRS)) { + dev_err(&nic->pdev->dev, + "Invalid MTU setting from VF%d rejected, should be between %d and %d\n", + vf, NIC_HW_MIN_FRS, NIC_HW_MAX_FRS); + return 1; + } + new_frs += ETH_HLEN; + if (new_frs <= nic->pkind.maxlen) + return 0; + + nic->pkind.maxlen = new_frs; + nic_reg_write(nic, NIC_PF_PKIND_0_15_CFG, *(u64 *)&nic->pkind); + return 0; +} + +/* Set minimum transmit packet size */ +static void nic_set_tx_pkt_pad(struct nicpf *nic, int size) +{ + int lmac; + u64 lmac_cfg; + + /* Max value that can be set is 60 */ + if (size > 60) + size = 60; + + for (lmac = 0; lmac < (MAX_BGX_PER_CN88XX * MAX_LMAC_PER_BGX); lmac++) { + lmac_cfg = nic_reg_read(nic, NIC_PF_LMAC_0_7_CFG | (lmac << 3)); + lmac_cfg &= ~(0xF << 2); + lmac_cfg |= ((size / 4) << 2); + nic_reg_write(nic, NIC_PF_LMAC_0_7_CFG | (lmac << 3), lmac_cfg); + } +} + +/* Function to check number of LMACs present and set VF::LMAC mapping. + * Mapping will be used while initializing channels. + */ +static void nic_set_lmac_vf_mapping(struct nicpf *nic) +{ + unsigned bgx_map = bgx_get_map(nic->node); + int bgx, next_bgx_lmac = 0; + int lmac, lmac_cnt = 0; + u64 lmac_credit; + + nic->num_vf_en = 0; + + for (bgx = 0; bgx < NIC_MAX_BGX; bgx++) { + if (!(bgx_map & (1 << bgx))) + continue; + lmac_cnt = bgx_get_lmac_count(nic->node, bgx); + for (lmac = 0; lmac < lmac_cnt; lmac++) + nic->vf_lmac_map[next_bgx_lmac++] = + NIC_SET_VF_LMAC_MAP(bgx, lmac); + nic->num_vf_en += lmac_cnt; + + /* Program LMAC credits */ + lmac_credit = (1ull << 1); /* channel credit enable */ + lmac_credit |= (0x1ff << 2); /* Max outstanding pkt count */ + /* 48KB BGX Tx buffer size, each unit is of size 16bytes */ + lmac_credit |= (((((48 * 1024) / lmac_cnt) - + NIC_HW_MAX_FRS) / 16) << 12); + lmac = bgx * MAX_LMAC_PER_BGX; + for (; lmac < lmac_cnt + (bgx * MAX_LMAC_PER_BGX); lmac++) + nic_reg_write(nic, + NIC_PF_LMAC_0_7_CREDIT + (lmac * 8), + lmac_credit); + } +} + +#define BGX0_BLOCK 8 +#define BGX1_BLOCK 9 + +static void nic_init_hw(struct nicpf *nic) +{ + int i; + + /* Reset NIC, in case the driver is repeatedly inserted and removed */ + nic_reg_write(nic, NIC_PF_SOFT_RESET, 1); + + /* Enable NIC HW block */ + nic_reg_write(nic, NIC_PF_CFG, 0x3); + + /* Enable backpressure */ + nic_reg_write(nic, NIC_PF_BP_CFG, (1ULL << 6) | 0x03); + + /* Disable TNS mode on both interfaces */ + nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG, + (NIC_TNS_BYPASS_MODE << 7) | BGX0_BLOCK); + nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG | (1 << 8), + (NIC_TNS_BYPASS_MODE << 7) | BGX1_BLOCK); + nic_reg_write(nic, NIC_PF_INTF_0_1_BP_CFG, + (1ULL << 63) | BGX0_BLOCK); + nic_reg_write(nic, NIC_PF_INTF_0_1_BP_CFG + (1 << 8), + (1ULL << 63) | BGX1_BLOCK); + + /* PKIND configuration */ + nic->pkind.minlen = 0; + nic->pkind.maxlen = NIC_HW_MAX_FRS + ETH_HLEN; + nic->pkind.lenerr_en = 1; + nic->pkind.rx_hdr = 0; + nic->pkind.hdr_sl = 0; + + for (i = 0; i < NIC_MAX_PKIND; i++) + nic_reg_write(nic, NIC_PF_PKIND_0_15_CFG | (i << 3), + *(u64 *)&nic->pkind); + + nic_set_tx_pkt_pad(nic, NIC_HW_MIN_FRS); + + /* Timer config */ + nic_reg_write(nic, NIC_PF_INTR_TIMER_CFG, NICPF_CLK_PER_INT_TICK); +} + +/* Channel parse index configuration */ +static void nic_config_cpi(struct nicpf *nic, struct cpi_cfg_msg *cfg) +{ + u32 vnic, bgx, lmac, chan; + u32 padd, cpi_count = 0; + u64 cpi_base, cpi, rssi_base, rssi; + u8 qset, rq_idx = 0; + + vnic = cfg->vf_id; + bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); + + chan = (lmac * MAX_BGX_CHANS_PER_LMAC) + (bgx * NIC_CHANS_PER_INF); + cpi_base = (lmac * NIC_MAX_CPI_PER_LMAC) + (bgx * NIC_CPI_PER_BGX); + rssi_base = (lmac * nic->rss_ind_tbl_size) + (bgx * NIC_RSSI_PER_BGX); + + /* Rx channel configuration */ + nic_reg_write(nic, NIC_PF_CHAN_0_255_RX_BP_CFG | (chan << 3), + (1ull << 63) | (vnic << 0)); + nic_reg_write(nic, NIC_PF_CHAN_0_255_RX_CFG | (chan << 3), + ((u64)cfg->cpi_alg << 62) | (cpi_base << 48)); + + if (cfg->cpi_alg == CPI_ALG_NONE) + cpi_count = 1; + else if (cfg->cpi_alg == CPI_ALG_VLAN) /* 3 bits of PCP */ + cpi_count = 8; + else if (cfg->cpi_alg == CPI_ALG_VLAN16) /* 3 bits PCP + DEI */ + cpi_count = 16; + else if (cfg->cpi_alg == CPI_ALG_DIFF) /* 6bits DSCP */ + cpi_count = NIC_MAX_CPI_PER_LMAC; + + /* RSS Qset, Qidx mapping */ + qset = cfg->vf_id; + rssi = rssi_base; + for (; rssi < (rssi_base + cfg->rq_cnt); rssi++) { + nic_reg_write(nic, NIC_PF_RSSI_0_4097_RQ | (rssi << 3), + (qset << 3) | rq_idx); + rq_idx++; + } + + rssi = 0; + cpi = cpi_base; + for (; cpi < (cpi_base + cpi_count); cpi++) { + /* Determine port to channel adder */ + if (cfg->cpi_alg != CPI_ALG_DIFF) + padd = cpi % cpi_count; + else + padd = cpi % 8; /* 3 bits CS out of 6bits DSCP */ + + /* Leave RSS_SIZE as '0' to disable RSS */ + nic_reg_write(nic, NIC_PF_CPI_0_2047_CFG | (cpi << 3), + (vnic << 24) | (padd << 16) | (rssi_base + rssi)); + + if ((rssi + 1) >= cfg->rq_cnt) + continue; + + if (cfg->cpi_alg == CPI_ALG_VLAN) + rssi++; + else if (cfg->cpi_alg == CPI_ALG_VLAN16) + rssi = ((cpi - cpi_base) & 0xe) >> 1; + else if (cfg->cpi_alg == CPI_ALG_DIFF) + rssi = ((cpi - cpi_base) & 0x38) >> 3; + } + nic->cpi_base[cfg->vf_id] = cpi_base; +} + +/* Responsds to VF with its RSS indirection table size */ +static void nic_send_rss_size(struct nicpf *nic, int vf) +{ + union nic_mbx mbx = {}; + u64 *msg; + + msg = (u64 *)&mbx; + + mbx.rss_size.msg = NIC_MBOX_MSG_RSS_SIZE; + mbx.rss_size.ind_tbl_size = nic->rss_ind_tbl_size; + nic_send_msg_to_vf(nic, vf, &mbx); +} + +/* Receive side scaling configuration + * configure: + * - RSS index + * - indir table i.e hash::RQ mapping + * - no of hash bits to consider + */ +static void nic_config_rss(struct nicpf *nic, struct rss_cfg_msg *cfg) +{ + u8 qset, idx = 0; + u64 cpi_cfg, cpi_base, rssi_base, rssi; + + cpi_base = nic->cpi_base[cfg->vf_id]; + cpi_cfg = nic_reg_read(nic, NIC_PF_CPI_0_2047_CFG | (cpi_base << 3)); + rssi_base = (cpi_cfg & 0x0FFF) + cfg->tbl_offset; + + rssi = rssi_base; + qset = cfg->vf_id; + + for (; rssi < (rssi_base + cfg->tbl_len); rssi++) { + nic_reg_write(nic, NIC_PF_RSSI_0_4097_RQ | (rssi << 3), + (qset << 3) | (cfg->ind_tbl[idx] & 0x7)); + idx++; + } + + cpi_cfg &= ~(0xFULL << 20); + cpi_cfg |= (cfg->hash_bits << 20); + nic_reg_write(nic, NIC_PF_CPI_0_2047_CFG | (cpi_base << 3), cpi_cfg); +} + +/* 4 level transmit side scheduler configutation + * for TNS bypass mode + * + * Sample configuration for SQ0 + * VNIC0-SQ0 -> TL4(0) -> TL3[0] -> TL2[0] -> TL1[0] -> BGX0 + * VNIC1-SQ0 -> TL4(8) -> TL3[2] -> TL2[0] -> TL1[0] -> BGX0 + * VNIC2-SQ0 -> TL4(16) -> TL3[4] -> TL2[1] -> TL1[0] -> BGX0 + * VNIC3-SQ0 -> TL4(24) -> TL3[6] -> TL2[1] -> TL1[0] -> BGX0 + * VNIC4-SQ0 -> TL4(512) -> TL3[128] -> TL2[32] -> TL1[1] -> BGX1 + * VNIC5-SQ0 -> TL4(520) -> TL3[130] -> TL2[32] -> TL1[1] -> BGX1 + * VNIC6-SQ0 -> TL4(528) -> TL3[132] -> TL2[33] -> TL1[1] -> BGX1 + * VNIC7-SQ0 -> TL4(536) -> TL3[134] -> TL2[33] -> TL1[1] -> BGX1 + */ +static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, u8 sq_idx) +{ + u32 bgx, lmac, chan; + u32 tl2, tl3, tl4; + u32 rr_quantum; + + bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); + /* 24 bytes for FCS, IPG and preamble */ + rr_quantum = ((NIC_HW_MAX_FRS + 24) / 4); + + tl4 = (lmac * NIC_TL4_PER_LMAC) + (bgx * NIC_TL4_PER_BGX); + tl4 += sq_idx; + tl3 = tl4 / (NIC_MAX_TL4 / NIC_MAX_TL3); + nic_reg_write(nic, NIC_PF_QSET_0_127_SQ_0_7_CFG2 | + ((u64)vnic << NIC_QS_ID_SHIFT) | + ((u32)sq_idx << NIC_Q_NUM_SHIFT), tl4); + nic_reg_write(nic, NIC_PF_TL4_0_1023_CFG | (tl4 << 3), + ((u64)vnic << 27) | ((u32)sq_idx << 24) | rr_quantum); + + nic_reg_write(nic, NIC_PF_TL3_0_255_CFG | (tl3 << 3), rr_quantum); + chan = (lmac * MAX_BGX_CHANS_PER_LMAC) + (bgx * NIC_CHANS_PER_INF); + nic_reg_write(nic, NIC_PF_TL3_0_255_CHAN | (tl3 << 3), chan); + /* Enable backpressure on the channel */ + nic_reg_write(nic, NIC_PF_CHAN_0_255_TX_CFG | (chan << 3), 1); + + tl2 = tl3 >> 2; + nic_reg_write(nic, NIC_PF_TL3A_0_63_CFG | (tl2 << 3), tl2); + nic_reg_write(nic, NIC_PF_TL2_0_63_CFG | (tl2 << 3), rr_quantum); + /* No priorities as of now */ + nic_reg_write(nic, NIC_PF_TL2_0_63_PRI | (tl2 << 3), 0x00); +} + +/* Interrupt handler to handle mailbox messages from VFs */ +static void nic_handle_mbx_intr(struct nicpf *nic, int vf) +{ + union nic_mbx mbx = {}; + u64 *mbx_data; + u64 mbx_addr; + u64 reg_addr; + u64 mac_addr; + int bgx, lmac; + int i; + int ret = 0; + + nic->mbx_lock[vf] = true; + + mbx_addr = nic_get_mbx_addr(vf); + mbx_data = (u64 *)&mbx; + + for (i = 0; i < NIC_PF_VF_MAILBOX_SIZE; i++) { + *mbx_data = nic_reg_read(nic, mbx_addr); + mbx_data++; + mbx_addr += sizeof(u64); + } + + dev_dbg(&nic->pdev->dev, "%s: Mailbox msg %d from VF%d\n", + __func__, mbx.msg.msg, vf); + switch (mbx.msg.msg) { + case NIC_MBOX_MSG_READY: + nic_mbx_send_ready(nic, vf); + nic->link[vf] = 0; + nic->duplex[vf] = 0; + nic->speed[vf] = 0; + ret = 1; + break; + case NIC_MBOX_MSG_QS_CFG: + reg_addr = NIC_PF_QSET_0_127_CFG | + (mbx.qs.num << NIC_QS_ID_SHIFT); + nic_reg_write(nic, reg_addr, mbx.qs.cfg); + break; + case NIC_MBOX_MSG_RQ_CFG: + reg_addr = NIC_PF_QSET_0_127_RQ_0_7_CFG | + (mbx.rq.qs_num << NIC_QS_ID_SHIFT) | + (mbx.rq.rq_num << NIC_Q_NUM_SHIFT); + nic_reg_write(nic, reg_addr, mbx.rq.cfg); + break; + case NIC_MBOX_MSG_RQ_BP_CFG: + reg_addr = NIC_PF_QSET_0_127_RQ_0_7_BP_CFG | + (mbx.rq.qs_num << NIC_QS_ID_SHIFT) | + (mbx.rq.rq_num << NIC_Q_NUM_SHIFT); + nic_reg_write(nic, reg_addr, mbx.rq.cfg); + break; + case NIC_MBOX_MSG_RQ_SW_SYNC: + ret = nic_rcv_queue_sw_sync(nic); + break; + case NIC_MBOX_MSG_RQ_DROP_CFG: + reg_addr = NIC_PF_QSET_0_127_RQ_0_7_DROP_CFG | + (mbx.rq.qs_num << NIC_QS_ID_SHIFT) | + (mbx.rq.rq_num << NIC_Q_NUM_SHIFT); + nic_reg_write(nic, reg_addr, mbx.rq.cfg); + break; + case NIC_MBOX_MSG_SQ_CFG: + reg_addr = NIC_PF_QSET_0_127_SQ_0_7_CFG | + (mbx.sq.qs_num << NIC_QS_ID_SHIFT) | + (mbx.sq.sq_num << NIC_Q_NUM_SHIFT); + nic_reg_write(nic, reg_addr, mbx.sq.cfg); + nic_tx_channel_cfg(nic, mbx.qs.num, mbx.sq.sq_num); + break; + case NIC_MBOX_MSG_SET_MAC: + lmac = mbx.mac.vf_id; + bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[lmac]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[lmac]); +#ifdef __BIG_ENDIAN + mac_addr = cpu_to_be64(mbx.nic_cfg.mac_addr) << 16; +#else + mac_addr = cpu_to_be64(mbx.nic_cfg.mac_addr) >> 16; +#endif + bgx_set_lmac_mac(nic->node, bgx, lmac, (u8 *)&mac_addr); + break; + case NIC_MBOX_MSG_SET_MAX_FRS: + ret = nic_update_hw_frs(nic, mbx.frs.max_frs, + mbx.frs.vf_id); + break; + case NIC_MBOX_MSG_CPI_CFG: + nic_config_cpi(nic, &mbx.cpi_cfg); + break; + case NIC_MBOX_MSG_RSS_SIZE: + nic_send_rss_size(nic, vf); + goto unlock; + case NIC_MBOX_MSG_RSS_CFG: + case NIC_MBOX_MSG_RSS_CFG_CONT: + nic_config_rss(nic, &mbx.rss_cfg); + break; + case NIC_MBOX_MSG_CFG_DONE: + /* Last message of VF config msg sequence */ + nic->vf_enabled[vf] = true; + goto unlock; + case NIC_MBOX_MSG_SHUTDOWN: + /* First msg in VF teardown sequence */ + nic->vf_enabled[vf] = false; + break; + case NIC_MBOX_MSG_BGX_STATS: + nic_get_bgx_stats(nic, &mbx.bgx_stats); + goto unlock; + default: + dev_err(&nic->pdev->dev, + "Invalid msg from VF%d, msg 0x%x\n", vf, mbx.msg.msg); + break; + } + + if (!ret) + nic_mbx_send_ack(nic, vf); + else if (mbx.msg.msg != NIC_MBOX_MSG_READY) + nic_mbx_send_nack(nic, vf); +unlock: + nic->mbx_lock[vf] = false; +} + +static void nic_mbx_intr_handler (struct nicpf *nic, int mbx) +{ + u64 intr; + u8 vf, vf_per_mbx_reg = 64; + + intr = nic_reg_read(nic, NIC_PF_MAILBOX_INT + (mbx << 3)); + dev_dbg(&nic->pdev->dev, "PF interrupt Mbox%d 0x%llx\n", mbx, intr); + for (vf = 0; vf < vf_per_mbx_reg; vf++) { + if (intr & (1ULL << vf)) { + dev_dbg(&nic->pdev->dev, "Intr from VF %d\n", + vf + (mbx * vf_per_mbx_reg)); + if ((vf + (mbx * vf_per_mbx_reg)) > nic->num_vf_en) + break; + nic_handle_mbx_intr(nic, vf + (mbx * vf_per_mbx_reg)); + nic_clear_mbx_intr(nic, vf, mbx); + } + } +} + +static irqreturn_t nic_mbx0_intr_handler (int irq, void *nic_irq) +{ + struct nicpf *nic = (struct nicpf *)nic_irq; + + nic_mbx_intr_handler(nic, 0); + + return IRQ_HANDLED; +} + +static irqreturn_t nic_mbx1_intr_handler (int irq, void *nic_irq) +{ + struct nicpf *nic = (struct nicpf *)nic_irq; + + nic_mbx_intr_handler(nic, 1); + + return IRQ_HANDLED; +} + +static int nic_enable_msix(struct nicpf *nic) +{ + int i, ret; + + nic->num_vec = NIC_PF_MSIX_VECTORS; + + for (i = 0; i < nic->num_vec; i++) + nic->msix_entries[i].entry = i; + + ret = pci_enable_msix(nic->pdev, nic->msix_entries, nic->num_vec); + if (ret) { + dev_err(&nic->pdev->dev, + "Request for #%d msix vectors failed\n", + nic->num_vec); + return ret; + } + + nic->msix_enabled = 1; + return 0; +} + +static void nic_disable_msix(struct nicpf *nic) +{ + if (nic->msix_enabled) { + pci_disable_msix(nic->pdev); + nic->msix_enabled = 0; + nic->num_vec = 0; + } +} + +static void nic_free_all_interrupts(struct nicpf *nic) +{ + int irq; + + for (irq = 0; irq < nic->num_vec; irq++) { + if (nic->irq_allocated[irq]) + free_irq(nic->msix_entries[irq].vector, nic); + nic->irq_allocated[irq] = false; + } +} + +static int nic_register_interrupts(struct nicpf *nic) +{ + int ret; + + /* Enable MSI-X */ + ret = nic_enable_msix(nic); + if (ret) + return ret; + + /* Register mailbox interrupt handlers */ + ret = request_irq(nic->msix_entries[NIC_PF_INTR_ID_MBOX0].vector, + nic_mbx0_intr_handler, 0, "NIC Mbox0", nic); + if (ret) + goto fail; + + nic->irq_allocated[NIC_PF_INTR_ID_MBOX0] = true; + + ret = request_irq(nic->msix_entries[NIC_PF_INTR_ID_MBOX1].vector, + nic_mbx1_intr_handler, 0, "NIC Mbox1", nic); + if (ret) + goto fail; + + nic->irq_allocated[NIC_PF_INTR_ID_MBOX1] = true; + + /* Enable mailbox interrupt */ + nic_enable_mbx_intr(nic); + return 0; + +fail: + dev_err(&nic->pdev->dev, "Request irq failed\n"); + nic_free_all_interrupts(nic); + return ret; +} + +static void nic_unregister_interrupts(struct nicpf *nic) +{ + nic_free_all_interrupts(nic); + nic_disable_msix(nic); +} + +static int nic_sriov_init(struct pci_dev *pdev, struct nicpf *nic) +{ + int pos = 0; + int err; + u16 total_vf_cnt; + + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_SRIOV); + if (!pos) { + dev_err(&pdev->dev, "SRIOV capability is not found in PCIe config space\n"); + return -ENODEV; + } + + pci_read_config_word(pdev, (pos + PCI_SRIOV_TOTAL_VF), &total_vf_cnt); + if (total_vf_cnt < nic->num_vf_en) + nic->num_vf_en = total_vf_cnt; + + if (!total_vf_cnt) + return 0; + + err = pci_enable_sriov(pdev, nic->num_vf_en); + if (err) { + dev_err(&pdev->dev, "SRIOV enable failed, num VF is %d\n", + nic->num_vf_en); + nic->num_vf_en = 0; + return err; + } + + dev_info(&pdev->dev, "SRIOV enabled, number of VF available %d\n", + nic->num_vf_en); + + nic->flags |= NIC_SRIOV_ENABLED; + return 0; +} + +/* Poll for BGX LMAC link status and update corresponding VF + * if there is a change, valid only if internal L2 switch + * is not present otherwise VF link is always treated as up + */ +static void nic_poll_for_link(struct work_struct *work) +{ + union nic_mbx mbx = {}; + struct nicpf *nic; + struct bgx_link_status link; + u8 vf, bgx, lmac; + + nic = container_of(work, struct nicpf, dwork.work); + + mbx.link_status.msg = NIC_MBOX_MSG_BGX_LINK_CHANGE; + + for (vf = 0; vf < nic->num_vf_en; vf++) { + /* Poll only if VF is UP */ + if (!nic->vf_enabled[vf]) + continue; + + /* Get BGX, LMAC indices for the VF */ + bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); + lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); + /* Get interface link status */ + bgx_get_lmac_link_state(nic->node, bgx, lmac, &link); + + /* Inform VF only if link status changed */ + if (nic->link[vf] == link.link_up) + continue; + + if (!nic->mbx_lock[vf]) { + nic->link[vf] = link.link_up; + nic->duplex[vf] = link.duplex; + nic->speed[vf] = link.speed; + + /* Send a mbox message to VF with current link status */ + mbx.link_status.link_up = link.link_up; + mbx.link_status.duplex = link.duplex; + mbx.link_status.speed = link.speed; + nic_send_msg_to_vf(nic, vf, &mbx); + } + } + queue_delayed_work(nic->check_link, &nic->dwork, HZ * 2); +} + +static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + struct device *dev = &pdev->dev; + struct nicpf *nic; + int err; + + BUILD_BUG_ON(sizeof(union nic_mbx) > 16); + + nic = devm_kzalloc(dev, sizeof(*nic), GFP_KERNEL); + if (!nic) + return -ENOMEM; + + pci_set_drvdata(pdev, nic); + + nic->pdev = pdev; + + err = pci_enable_device(pdev); + if (err) { + dev_err(dev, "Failed to enable PCI device\n"); + pci_set_drvdata(pdev, NULL); + return err; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + dev_err(dev, "PCI request regions failed 0x%x\n", err); + goto err_disable_device; + } + + err = pci_set_dma_mask(pdev, DMA_BIT_MASK(48)); + if (err) { + dev_err(dev, "Unable to get usable DMA configuration\n"); + goto err_release_regions; + } + + err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(48)); + if (err) { + dev_err(dev, "Unable to get 48-bit DMA for consistent allocations\n"); + goto err_release_regions; + } + + /* MAP PF's configuration registers */ + nic->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); + if (!nic->reg_base) { + dev_err(dev, "Cannot map config register space, aborting\n"); + err = -ENOMEM; + goto err_release_regions; + } + + pci_read_config_byte(pdev, PCI_REVISION_ID, &nic->rev_id); + + nic->node = NIC_NODE_ID(pci_resource_start(pdev, PCI_CFG_REG_BAR_NUM)); + + nic_set_lmac_vf_mapping(nic); + + /* Initialize hardware */ + nic_init_hw(nic); + + /* Set RSS TBL size for each VF */ + nic->rss_ind_tbl_size = NIC_MAX_RSS_IDR_TBL_SIZE; + + /* Register interrupts */ + err = nic_register_interrupts(nic); + if (err) + goto err_release_regions; + + /* Configure SRIOV */ + err = nic_sriov_init(pdev, nic); + if (err) + goto err_unregister_interrupts; + + /* Register a physical link status poll fn() */ + nic->check_link = alloc_workqueue("check_link_status", + WQ_UNBOUND | WQ_MEM_RECLAIM, 1); + if (!nic->check_link) { + err = -ENOMEM; + goto err_disable_sriov; + } + + INIT_DELAYED_WORK(&nic->dwork, nic_poll_for_link); + queue_delayed_work(nic->check_link, &nic->dwork, 0); + + return 0; + +err_disable_sriov: + if (nic->flags & NIC_SRIOV_ENABLED) + pci_disable_sriov(pdev); +err_unregister_interrupts: + nic_unregister_interrupts(nic); +err_release_regions: + pci_release_regions(pdev); +err_disable_device: + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + return err; +} + +static void nic_remove(struct pci_dev *pdev) +{ + struct nicpf *nic = pci_get_drvdata(pdev); + + if (nic->flags & NIC_SRIOV_ENABLED) + pci_disable_sriov(pdev); + + if (nic->check_link) { + /* Destroy work Queue */ + cancel_delayed_work(&nic->dwork); + flush_workqueue(nic->check_link); + destroy_workqueue(nic->check_link); + } + + nic_unregister_interrupts(nic); + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); +} + +static struct pci_driver nic_driver = { + .name = DRV_NAME, + .id_table = nic_id_table, + .probe = nic_probe, + .remove = nic_remove, +}; + +static int __init nic_init_module(void) +{ + pr_info("%s, ver %s\n", DRV_NAME, DRV_VERSION); + + return pci_register_driver(&nic_driver); +} + +static void __exit nic_cleanup_module(void) +{ + pci_unregister_driver(&nic_driver); +} + +module_init(nic_init_module); +module_exit(nic_cleanup_module); diff --git a/drivers/net/ethernet/cavium/thunder/nic_reg.h b/drivers/net/ethernet/cavium/thunder/nic_reg.h new file mode 100644 index 000000000000..58197bb2f805 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nic_reg.h @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#ifndef NIC_REG_H +#define NIC_REG_H + +#define NIC_PF_REG_COUNT 29573 +#define NIC_VF_REG_COUNT 249 + +/* Physical function register offsets */ +#define NIC_PF_CFG (0x0000) +#define NIC_PF_STATUS (0x0010) +#define NIC_PF_INTR_TIMER_CFG (0x0030) +#define NIC_PF_BIST_STATUS (0x0040) +#define NIC_PF_SOFT_RESET (0x0050) +#define NIC_PF_TCP_TIMER (0x0060) +#define NIC_PF_BP_CFG (0x0080) +#define NIC_PF_RRM_CFG (0x0088) +#define NIC_PF_CQM_CF (0x00A0) +#define NIC_PF_CNM_CF (0x00A8) +#define NIC_PF_CNM_STATUS (0x00B0) +#define NIC_PF_CQ_AVG_CFG (0x00C0) +#define NIC_PF_RRM_AVG_CFG (0x00C8) +#define NIC_PF_INTF_0_1_SEND_CFG (0x0200) +#define NIC_PF_INTF_0_1_BP_CFG (0x0208) +#define NIC_PF_INTF_0_1_BP_DIS_0_1 (0x0210) +#define NIC_PF_INTF_0_1_BP_SW_0_1 (0x0220) +#define NIC_PF_RBDR_BP_STATE_0_3 (0x0240) +#define NIC_PF_MAILBOX_INT (0x0410) +#define NIC_PF_MAILBOX_INT_W1S (0x0430) +#define NIC_PF_MAILBOX_ENA_W1C (0x0450) +#define NIC_PF_MAILBOX_ENA_W1S (0x0470) +#define NIC_PF_RX_ETYPE_0_7 (0x0500) +#define NIC_PF_PKIND_0_15_CFG (0x0600) +#define NIC_PF_ECC0_FLIP0 (0x1000) +#define NIC_PF_ECC1_FLIP0 (0x1008) +#define NIC_PF_ECC2_FLIP0 (0x1010) +#define NIC_PF_ECC3_FLIP0 (0x1018) +#define NIC_PF_ECC0_FLIP1 (0x1080) +#define NIC_PF_ECC1_FLIP1 (0x1088) +#define NIC_PF_ECC2_FLIP1 (0x1090) +#define NIC_PF_ECC3_FLIP1 (0x1098) +#define NIC_PF_ECC0_CDIS (0x1100) +#define NIC_PF_ECC1_CDIS (0x1108) +#define NIC_PF_ECC2_CDIS (0x1110) +#define NIC_PF_ECC3_CDIS (0x1118) +#define NIC_PF_BIST0_STATUS (0x1280) +#define NIC_PF_BIST1_STATUS (0x1288) +#define NIC_PF_BIST2_STATUS (0x1290) +#define NIC_PF_BIST3_STATUS (0x1298) +#define NIC_PF_ECC0_SBE_INT (0x2000) +#define NIC_PF_ECC0_SBE_INT_W1S (0x2008) +#define NIC_PF_ECC0_SBE_ENA_W1C (0x2010) +#define NIC_PF_ECC0_SBE_ENA_W1S (0x2018) +#define NIC_PF_ECC0_DBE_INT (0x2100) +#define NIC_PF_ECC0_DBE_INT_W1S (0x2108) +#define NIC_PF_ECC0_DBE_ENA_W1C (0x2110) +#define NIC_PF_ECC0_DBE_ENA_W1S (0x2118) +#define NIC_PF_ECC1_SBE_INT (0x2200) +#define NIC_PF_ECC1_SBE_INT_W1S (0x2208) +#define NIC_PF_ECC1_SBE_ENA_W1C (0x2210) +#define NIC_PF_ECC1_SBE_ENA_W1S (0x2218) +#define NIC_PF_ECC1_DBE_INT (0x2300) +#define NIC_PF_ECC1_DBE_INT_W1S (0x2308) +#define NIC_PF_ECC1_DBE_ENA_W1C (0x2310) +#define NIC_PF_ECC1_DBE_ENA_W1S (0x2318) +#define NIC_PF_ECC2_SBE_INT (0x2400) +#define NIC_PF_ECC2_SBE_INT_W1S (0x2408) +#define NIC_PF_ECC2_SBE_ENA_W1C (0x2410) +#define NIC_PF_ECC2_SBE_ENA_W1S (0x2418) +#define NIC_PF_ECC2_DBE_INT (0x2500) +#define NIC_PF_ECC2_DBE_INT_W1S (0x2508) +#define NIC_PF_ECC2_DBE_ENA_W1C (0x2510) +#define NIC_PF_ECC2_DBE_ENA_W1S (0x2518) +#define NIC_PF_ECC3_SBE_INT (0x2600) +#define NIC_PF_ECC3_SBE_INT_W1S (0x2608) +#define NIC_PF_ECC3_SBE_ENA_W1C (0x2610) +#define NIC_PF_ECC3_SBE_ENA_W1S (0x2618) +#define NIC_PF_ECC3_DBE_INT (0x2700) +#define NIC_PF_ECC3_DBE_INT_W1S (0x2708) +#define NIC_PF_ECC3_DBE_ENA_W1C (0x2710) +#define NIC_PF_ECC3_DBE_ENA_W1S (0x2718) +#define NIC_PF_CPI_0_2047_CFG (0x200000) +#define NIC_PF_RSSI_0_4097_RQ (0x220000) +#define NIC_PF_LMAC_0_7_CFG (0x240000) +#define NIC_PF_LMAC_0_7_SW_XOFF (0x242000) +#define NIC_PF_LMAC_0_7_CREDIT (0x244000) +#define NIC_PF_CHAN_0_255_TX_CFG (0x400000) +#define NIC_PF_CHAN_0_255_RX_CFG (0x420000) +#define NIC_PF_CHAN_0_255_SW_XOFF (0x440000) +#define NIC_PF_CHAN_0_255_CREDIT (0x460000) +#define NIC_PF_CHAN_0_255_RX_BP_CFG (0x480000) +#define NIC_PF_SW_SYNC_RX (0x490000) +#define NIC_PF_SW_SYNC_RX_DONE (0x490008) +#define NIC_PF_TL2_0_63_CFG (0x500000) +#define NIC_PF_TL2_0_63_PRI (0x520000) +#define NIC_PF_TL2_0_63_SH_STATUS (0x580000) +#define NIC_PF_TL3A_0_63_CFG (0x5F0000) +#define NIC_PF_TL3_0_255_CFG (0x600000) +#define NIC_PF_TL3_0_255_CHAN (0x620000) +#define NIC_PF_TL3_0_255_PIR (0x640000) +#define NIC_PF_TL3_0_255_SW_XOFF (0x660000) +#define NIC_PF_TL3_0_255_CNM_RATE (0x680000) +#define NIC_PF_TL3_0_255_SH_STATUS (0x6A0000) +#define NIC_PF_TL4A_0_255_CFG (0x6F0000) +#define NIC_PF_TL4_0_1023_CFG (0x800000) +#define NIC_PF_TL4_0_1023_SW_XOFF (0x820000) +#define NIC_PF_TL4_0_1023_SH_STATUS (0x840000) +#define NIC_PF_TL4A_0_1023_CNM_RATE (0x880000) +#define NIC_PF_TL4A_0_1023_CNM_STATUS (0x8A0000) +#define NIC_PF_VF_0_127_MAILBOX_0_1 (0x20002030) +#define NIC_PF_VNIC_0_127_TX_STAT_0_4 (0x20004000) +#define NIC_PF_VNIC_0_127_RX_STAT_0_13 (0x20004100) +#define NIC_PF_QSET_0_127_LOCK_0_15 (0x20006000) +#define NIC_PF_QSET_0_127_CFG (0x20010000) +#define NIC_PF_QSET_0_127_RQ_0_7_CFG (0x20010400) +#define NIC_PF_QSET_0_127_RQ_0_7_DROP_CFG (0x20010420) +#define NIC_PF_QSET_0_127_RQ_0_7_BP_CFG (0x20010500) +#define NIC_PF_QSET_0_127_RQ_0_7_STAT_0_1 (0x20010600) +#define NIC_PF_QSET_0_127_SQ_0_7_CFG (0x20010C00) +#define NIC_PF_QSET_0_127_SQ_0_7_CFG2 (0x20010C08) +#define NIC_PF_QSET_0_127_SQ_0_7_STAT_0_1 (0x20010D00) + +#define NIC_PF_MSIX_VEC_0_18_ADDR (0x000000) +#define NIC_PF_MSIX_VEC_0_CTL (0x000008) +#define NIC_PF_MSIX_PBA_0 (0x0F0000) + +/* Virtual function register offsets */ +#define NIC_VNIC_CFG (0x000020) +#define NIC_VF_PF_MAILBOX_0_1 (0x000130) +#define NIC_VF_INT (0x000200) +#define NIC_VF_INT_W1S (0x000220) +#define NIC_VF_ENA_W1C (0x000240) +#define NIC_VF_ENA_W1S (0x000260) + +#define NIC_VNIC_RSS_CFG (0x0020E0) +#define NIC_VNIC_RSS_KEY_0_4 (0x002200) +#define NIC_VNIC_TX_STAT_0_4 (0x004000) +#define NIC_VNIC_RX_STAT_0_13 (0x004100) +#define NIC_QSET_RQ_GEN_CFG (0x010010) + +#define NIC_QSET_CQ_0_7_CFG (0x010400) +#define NIC_QSET_CQ_0_7_CFG2 (0x010408) +#define NIC_QSET_CQ_0_7_THRESH (0x010410) +#define NIC_QSET_CQ_0_7_BASE (0x010420) +#define NIC_QSET_CQ_0_7_HEAD (0x010428) +#define NIC_QSET_CQ_0_7_TAIL (0x010430) +#define NIC_QSET_CQ_0_7_DOOR (0x010438) +#define NIC_QSET_CQ_0_7_STATUS (0x010440) +#define NIC_QSET_CQ_0_7_STATUS2 (0x010448) +#define NIC_QSET_CQ_0_7_DEBUG (0x010450) + +#define NIC_QSET_RQ_0_7_CFG (0x010600) +#define NIC_QSET_RQ_0_7_STAT_0_1 (0x010700) + +#define NIC_QSET_SQ_0_7_CFG (0x010800) +#define NIC_QSET_SQ_0_7_THRESH (0x010810) +#define NIC_QSET_SQ_0_7_BASE (0x010820) +#define NIC_QSET_SQ_0_7_HEAD (0x010828) +#define NIC_QSET_SQ_0_7_TAIL (0x010830) +#define NIC_QSET_SQ_0_7_DOOR (0x010838) +#define NIC_QSET_SQ_0_7_STATUS (0x010840) +#define NIC_QSET_SQ_0_7_DEBUG (0x010848) +#define NIC_QSET_SQ_0_7_CNM_CHG (0x010860) +#define NIC_QSET_SQ_0_7_STAT_0_1 (0x010900) + +#define NIC_QSET_RBDR_0_1_CFG (0x010C00) +#define NIC_QSET_RBDR_0_1_THRESH (0x010C10) +#define NIC_QSET_RBDR_0_1_BASE (0x010C20) +#define NIC_QSET_RBDR_0_1_HEAD (0x010C28) +#define NIC_QSET_RBDR_0_1_TAIL (0x010C30) +#define NIC_QSET_RBDR_0_1_DOOR (0x010C38) +#define NIC_QSET_RBDR_0_1_STATUS0 (0x010C40) +#define NIC_QSET_RBDR_0_1_STATUS1 (0x010C48) +#define NIC_QSET_RBDR_0_1_PREFETCH_STATUS (0x010C50) + +#define NIC_VF_MSIX_VECTOR_0_19_ADDR (0x000000) +#define NIC_VF_MSIX_VECTOR_0_19_CTL (0x000008) +#define NIC_VF_MSIX_PBA (0x0F0000) + +/* Offsets within registers */ +#define NIC_MSIX_VEC_SHIFT 4 +#define NIC_Q_NUM_SHIFT 18 +#define NIC_QS_ID_SHIFT 21 +#define NIC_VF_NUM_SHIFT 21 + +/* Port kind configuration register */ +struct pkind_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_42_63:22; + u64 hdr_sl:5; /* Header skip length */ + u64 rx_hdr:3; /* TNS Receive header present */ + u64 lenerr_en:1;/* L2 length error check enable */ + u64 reserved_32_32:1; + u64 maxlen:16; /* Max frame size */ + u64 minlen:16; /* Min frame size */ +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 minlen:16; + u64 maxlen:16; + u64 reserved_32_32:1; + u64 lenerr_en:1; + u64 rx_hdr:3; + u64 hdr_sl:5; + u64 reserved_42_63:22; +#endif +}; + +#endif /* NIC_REG_H */ diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c new file mode 100644 index 000000000000..0fc4a536afc9 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c @@ -0,0 +1,601 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +/* ETHTOOL Support for VNIC_VF Device*/ + +#include <linux/pci.h> + +#include "nic_reg.h" +#include "nic.h" +#include "nicvf_queues.h" +#include "q_struct.h" +#include "thunder_bgx.h" + +#define DRV_NAME "thunder-nicvf" +#define DRV_VERSION "1.0" + +struct nicvf_stat { + char name[ETH_GSTRING_LEN]; + unsigned int index; +}; + +#define NICVF_HW_STAT(stat) { \ + .name = #stat, \ + .index = offsetof(struct nicvf_hw_stats, stat) / sizeof(u64), \ +} + +#define NICVF_DRV_STAT(stat) { \ + .name = #stat, \ + .index = offsetof(struct nicvf_drv_stats, stat) / sizeof(u64), \ +} + +static const struct nicvf_stat nicvf_hw_stats[] = { + NICVF_HW_STAT(rx_bytes_ok), + NICVF_HW_STAT(rx_ucast_frames_ok), + NICVF_HW_STAT(rx_bcast_frames_ok), + NICVF_HW_STAT(rx_mcast_frames_ok), + NICVF_HW_STAT(rx_fcs_errors), + NICVF_HW_STAT(rx_l2_errors), + NICVF_HW_STAT(rx_drop_red), + NICVF_HW_STAT(rx_drop_red_bytes), + NICVF_HW_STAT(rx_drop_overrun), + NICVF_HW_STAT(rx_drop_overrun_bytes), + NICVF_HW_STAT(rx_drop_bcast), + NICVF_HW_STAT(rx_drop_mcast), + NICVF_HW_STAT(rx_drop_l3_bcast), + NICVF_HW_STAT(rx_drop_l3_mcast), + NICVF_HW_STAT(tx_bytes_ok), + NICVF_HW_STAT(tx_ucast_frames_ok), + NICVF_HW_STAT(tx_bcast_frames_ok), + NICVF_HW_STAT(tx_mcast_frames_ok), +}; + +static const struct nicvf_stat nicvf_drv_stats[] = { + NICVF_DRV_STAT(rx_frames_ok), + NICVF_DRV_STAT(rx_frames_64), + NICVF_DRV_STAT(rx_frames_127), + NICVF_DRV_STAT(rx_frames_255), + NICVF_DRV_STAT(rx_frames_511), + NICVF_DRV_STAT(rx_frames_1023), + NICVF_DRV_STAT(rx_frames_1518), + NICVF_DRV_STAT(rx_frames_jumbo), + NICVF_DRV_STAT(rx_drops), + NICVF_DRV_STAT(tx_frames_ok), + NICVF_DRV_STAT(tx_busy), + NICVF_DRV_STAT(tx_tso), + NICVF_DRV_STAT(tx_drops), +}; + +static const struct nicvf_stat nicvf_queue_stats[] = { + { "bytes", 0 }, + { "frames", 1 }, +}; + +static const unsigned int nicvf_n_hw_stats = ARRAY_SIZE(nicvf_hw_stats); +static const unsigned int nicvf_n_drv_stats = ARRAY_SIZE(nicvf_drv_stats); +static const unsigned int nicvf_n_queue_stats = ARRAY_SIZE(nicvf_queue_stats); + +static int nicvf_get_settings(struct net_device *netdev, + struct ethtool_cmd *cmd) +{ + struct nicvf *nic = netdev_priv(netdev); + + cmd->supported = 0; + cmd->transceiver = XCVR_EXTERNAL; + if (nic->speed <= 1000) { + cmd->port = PORT_MII; + cmd->autoneg = AUTONEG_ENABLE; + } else { + cmd->port = PORT_FIBRE; + cmd->autoneg = AUTONEG_DISABLE; + } + cmd->duplex = nic->duplex; + ethtool_cmd_speed_set(cmd, nic->speed); + + return 0; +} + +static void nicvf_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *info) +{ + struct nicvf *nic = netdev_priv(netdev); + + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(nic->pdev), sizeof(info->bus_info)); +} + +static u32 nicvf_get_msglevel(struct net_device *netdev) +{ + struct nicvf *nic = netdev_priv(netdev); + + return nic->msg_enable; +} + +static void nicvf_set_msglevel(struct net_device *netdev, u32 lvl) +{ + struct nicvf *nic = netdev_priv(netdev); + + nic->msg_enable = lvl; +} + +static void nicvf_get_strings(struct net_device *netdev, u32 sset, u8 *data) +{ + int stats, qidx; + + if (sset != ETH_SS_STATS) + return; + + for (stats = 0; stats < nicvf_n_hw_stats; stats++) { + memcpy(data, nicvf_hw_stats[stats].name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + + for (stats = 0; stats < nicvf_n_drv_stats; stats++) { + memcpy(data, nicvf_drv_stats[stats].name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + + for (qidx = 0; qidx < MAX_RCV_QUEUES_PER_QS; qidx++) { + for (stats = 0; stats < nicvf_n_queue_stats; stats++) { + sprintf(data, "rxq%d: %s", qidx, + nicvf_queue_stats[stats].name); + data += ETH_GSTRING_LEN; + } + } + + for (qidx = 0; qidx < MAX_SND_QUEUES_PER_QS; qidx++) { + for (stats = 0; stats < nicvf_n_queue_stats; stats++) { + sprintf(data, "txq%d: %s", qidx, + nicvf_queue_stats[stats].name); + data += ETH_GSTRING_LEN; + } + } + + for (stats = 0; stats < BGX_RX_STATS_COUNT; stats++) { + sprintf(data, "bgx_rxstat%d: ", stats); + data += ETH_GSTRING_LEN; + } + + for (stats = 0; stats < BGX_TX_STATS_COUNT; stats++) { + sprintf(data, "bgx_txstat%d: ", stats); + data += ETH_GSTRING_LEN; + } +} + +static int nicvf_get_sset_count(struct net_device *netdev, int sset) +{ + if (sset != ETH_SS_STATS) + return -EINVAL; + + return nicvf_n_hw_stats + nicvf_n_drv_stats + + (nicvf_n_queue_stats * + (MAX_RCV_QUEUES_PER_QS + MAX_SND_QUEUES_PER_QS)) + + BGX_RX_STATS_COUNT + BGX_TX_STATS_COUNT; +} + +static void nicvf_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, u64 *data) +{ + struct nicvf *nic = netdev_priv(netdev); + int stat, qidx; + + nicvf_update_stats(nic); + + /* Update LMAC stats */ + nicvf_update_lmac_stats(nic); + + for (stat = 0; stat < nicvf_n_hw_stats; stat++) + *(data++) = ((u64 *)&nic->stats) + [nicvf_hw_stats[stat].index]; + for (stat = 0; stat < nicvf_n_drv_stats; stat++) + *(data++) = ((u64 *)&nic->drv_stats) + [nicvf_drv_stats[stat].index]; + + for (qidx = 0; qidx < MAX_RCV_QUEUES_PER_QS; qidx++) { + for (stat = 0; stat < nicvf_n_queue_stats; stat++) + *(data++) = ((u64 *)&nic->qs->rq[qidx].stats) + [nicvf_queue_stats[stat].index]; + } + + for (qidx = 0; qidx < MAX_SND_QUEUES_PER_QS; qidx++) { + for (stat = 0; stat < nicvf_n_queue_stats; stat++) + *(data++) = ((u64 *)&nic->qs->sq[qidx].stats) + [nicvf_queue_stats[stat].index]; + } + + for (stat = 0; stat < BGX_RX_STATS_COUNT; stat++) + *(data++) = nic->bgx_stats.rx_stats[stat]; + for (stat = 0; stat < BGX_TX_STATS_COUNT; stat++) + *(data++) = nic->bgx_stats.tx_stats[stat]; +} + +static int nicvf_get_regs_len(struct net_device *dev) +{ + return sizeof(u64) * NIC_VF_REG_COUNT; +} + +static void nicvf_get_regs(struct net_device *dev, + struct ethtool_regs *regs, void *reg) +{ + struct nicvf *nic = netdev_priv(dev); + u64 *p = (u64 *)reg; + u64 reg_offset; + int mbox, key, stat, q; + int i = 0; + + regs->version = 0; + memset(p, 0, NIC_VF_REG_COUNT); + + p[i++] = nicvf_reg_read(nic, NIC_VNIC_CFG); + /* Mailbox registers */ + for (mbox = 0; mbox < NIC_PF_VF_MAILBOX_SIZE; mbox++) + p[i++] = nicvf_reg_read(nic, + NIC_VF_PF_MAILBOX_0_1 | (mbox << 3)); + + p[i++] = nicvf_reg_read(nic, NIC_VF_INT); + p[i++] = nicvf_reg_read(nic, NIC_VF_INT_W1S); + p[i++] = nicvf_reg_read(nic, NIC_VF_ENA_W1C); + p[i++] = nicvf_reg_read(nic, NIC_VF_ENA_W1S); + p[i++] = nicvf_reg_read(nic, NIC_VNIC_RSS_CFG); + + for (key = 0; key < RSS_HASH_KEY_SIZE; key++) + p[i++] = nicvf_reg_read(nic, NIC_VNIC_RSS_KEY_0_4 | (key << 3)); + + /* Tx/Rx statistics */ + for (stat = 0; stat < TX_STATS_ENUM_LAST; stat++) + p[i++] = nicvf_reg_read(nic, + NIC_VNIC_TX_STAT_0_4 | (stat << 3)); + + for (i = 0; i < RX_STATS_ENUM_LAST; i++) + p[i++] = nicvf_reg_read(nic, + NIC_VNIC_RX_STAT_0_13 | (stat << 3)); + + p[i++] = nicvf_reg_read(nic, NIC_QSET_RQ_GEN_CFG); + + /* All completion queue's registers */ + for (q = 0; q < MAX_CMP_QUEUES_PER_QS; q++) { + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_CFG, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_CFG2, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_THRESH, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_BASE, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_HEAD, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_TAIL, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_DOOR, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_STATUS, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_STATUS2, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_DEBUG, q); + } + + /* All receive queue's registers */ + for (q = 0; q < MAX_RCV_QUEUES_PER_QS; q++) { + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RQ_0_7_CFG, q); + p[i++] = nicvf_queue_reg_read(nic, + NIC_QSET_RQ_0_7_STAT_0_1, q); + reg_offset = NIC_QSET_RQ_0_7_STAT_0_1 | (1 << 3); + p[i++] = nicvf_queue_reg_read(nic, reg_offset, q); + } + + for (q = 0; q < MAX_SND_QUEUES_PER_QS; q++) { + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_CFG, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_THRESH, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_BASE, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_HEAD, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_TAIL, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_DOOR, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_STATUS, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_DEBUG, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_CNM_CHG, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_STAT_0_1, q); + reg_offset = NIC_QSET_SQ_0_7_STAT_0_1 | (1 << 3); + p[i++] = nicvf_queue_reg_read(nic, reg_offset, q); + } + + for (q = 0; q < MAX_RCV_BUF_DESC_RINGS_PER_QS; q++) { + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_CFG, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_THRESH, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_BASE, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_HEAD, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_TAIL, q); + p[i++] = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_DOOR, q); + p[i++] = nicvf_queue_reg_read(nic, + NIC_QSET_RBDR_0_1_STATUS0, q); + p[i++] = nicvf_queue_reg_read(nic, + NIC_QSET_RBDR_0_1_STATUS1, q); + reg_offset = NIC_QSET_RBDR_0_1_PREFETCH_STATUS; + p[i++] = nicvf_queue_reg_read(nic, reg_offset, q); + } +} + +static int nicvf_get_coalesce(struct net_device *netdev, + struct ethtool_coalesce *cmd) +{ + struct nicvf *nic = netdev_priv(netdev); + + cmd->rx_coalesce_usecs = nic->cq_coalesce_usecs; + return 0; +} + +static void nicvf_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct nicvf *nic = netdev_priv(netdev); + struct queue_set *qs = nic->qs; + + ring->rx_max_pending = MAX_RCV_BUF_COUNT; + ring->rx_pending = qs->rbdr_len; + ring->tx_max_pending = MAX_SND_QUEUE_LEN; + ring->tx_pending = qs->sq_len; +} + +static int nicvf_get_rss_hash_opts(struct nicvf *nic, + struct ethtool_rxnfc *info) +{ + info->data = 0; + + switch (info->flow_type) { + case TCP_V4_FLOW: + case TCP_V6_FLOW: + case UDP_V4_FLOW: + case UDP_V6_FLOW: + case SCTP_V4_FLOW: + case SCTP_V6_FLOW: + info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3; + case IPV4_FLOW: + case IPV6_FLOW: + info->data |= RXH_IP_SRC | RXH_IP_DST; + break; + default: + return -EINVAL; + } + + return 0; +} + +static int nicvf_get_rxnfc(struct net_device *dev, + struct ethtool_rxnfc *info, u32 *rules) +{ + struct nicvf *nic = netdev_priv(dev); + int ret = -EOPNOTSUPP; + + switch (info->cmd) { + case ETHTOOL_GRXRINGS: + info->data = nic->qs->rq_cnt; + ret = 0; + break; + case ETHTOOL_GRXFH: + return nicvf_get_rss_hash_opts(nic, info); + default: + break; + } + return ret; +} + +static int nicvf_set_rss_hash_opts(struct nicvf *nic, + struct ethtool_rxnfc *info) +{ + struct nicvf_rss_info *rss = &nic->rss_info; + u64 rss_cfg = nicvf_reg_read(nic, NIC_VNIC_RSS_CFG); + + if (!rss->enable) + netdev_err(nic->netdev, + "RSS is disabled, hash cannot be set\n"); + + netdev_info(nic->netdev, "Set RSS flow type = %d, data = %lld\n", + info->flow_type, info->data); + + if (!(info->data & RXH_IP_SRC) || !(info->data & RXH_IP_DST)) + return -EINVAL; + + switch (info->flow_type) { + case TCP_V4_FLOW: + case TCP_V6_FLOW: + switch (info->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) { + case 0: + rss_cfg &= ~(1ULL << RSS_HASH_TCP); + break; + case (RXH_L4_B_0_1 | RXH_L4_B_2_3): + rss_cfg |= (1ULL << RSS_HASH_TCP); + break; + default: + return -EINVAL; + } + break; + case UDP_V4_FLOW: + case UDP_V6_FLOW: + switch (info->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) { + case 0: + rss_cfg &= ~(1ULL << RSS_HASH_UDP); + break; + case (RXH_L4_B_0_1 | RXH_L4_B_2_3): + rss_cfg |= (1ULL << RSS_HASH_UDP); + break; + default: + return -EINVAL; + } + break; + case SCTP_V4_FLOW: + case SCTP_V6_FLOW: + switch (info->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) { + case 0: + rss_cfg &= ~(1ULL << RSS_HASH_L4ETC); + break; + case (RXH_L4_B_0_1 | RXH_L4_B_2_3): + rss_cfg |= (1ULL << RSS_HASH_L4ETC); + break; + default: + return -EINVAL; + } + break; + case IPV4_FLOW: + case IPV6_FLOW: + rss_cfg = RSS_HASH_IP; + break; + default: + return -EINVAL; + } + + nicvf_reg_write(nic, NIC_VNIC_RSS_CFG, rss_cfg); + return 0; +} + +static int nicvf_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info) +{ + struct nicvf *nic = netdev_priv(dev); + + switch (info->cmd) { + case ETHTOOL_SRXFH: + return nicvf_set_rss_hash_opts(nic, info); + default: + break; + } + return -EOPNOTSUPP; +} + +static u32 nicvf_get_rxfh_key_size(struct net_device *netdev) +{ + return RSS_HASH_KEY_SIZE * sizeof(u64); +} + +static u32 nicvf_get_rxfh_indir_size(struct net_device *dev) +{ + struct nicvf *nic = netdev_priv(dev); + + return nic->rss_info.rss_size; +} + +static int nicvf_get_rxfh(struct net_device *dev, u32 *indir, u8 *hkey, + u8 *hfunc) +{ + struct nicvf *nic = netdev_priv(dev); + struct nicvf_rss_info *rss = &nic->rss_info; + int idx; + + if (indir) { + for (idx = 0; idx < rss->rss_size; idx++) + indir[idx] = rss->ind_tbl[idx]; + } + + if (hkey) + memcpy(hkey, rss->key, RSS_HASH_KEY_SIZE * sizeof(u64)); + + if (hfunc) + *hfunc = ETH_RSS_HASH_TOP; + + return 0; +} + +static int nicvf_set_rxfh(struct net_device *dev, const u32 *indir, + const u8 *hkey, u8 hfunc) +{ + struct nicvf *nic = netdev_priv(dev); + struct nicvf_rss_info *rss = &nic->rss_info; + int idx; + + if ((nic->qs->rq_cnt <= 1) || (nic->cpi_alg != CPI_ALG_NONE)) { + rss->enable = false; + rss->hash_bits = 0; + return -EIO; + } + + /* We do not allow change in unsupported parameters */ + if (hkey || + (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)) + return -EOPNOTSUPP; + + rss->enable = true; + if (indir) { + for (idx = 0; idx < rss->rss_size; idx++) + rss->ind_tbl[idx] = indir[idx]; + } + + if (hkey) { + memcpy(rss->key, hkey, RSS_HASH_KEY_SIZE * sizeof(u64)); + nicvf_set_rss_key(nic); + } + + nicvf_config_rss(nic); + return 0; +} + +/* Get no of queues device supports and current queue count */ +static void nicvf_get_channels(struct net_device *dev, + struct ethtool_channels *channel) +{ + struct nicvf *nic = netdev_priv(dev); + + memset(channel, 0, sizeof(*channel)); + + channel->max_rx = MAX_RCV_QUEUES_PER_QS; + channel->max_tx = MAX_SND_QUEUES_PER_QS; + + channel->rx_count = nic->qs->rq_cnt; + channel->tx_count = nic->qs->sq_cnt; +} + +/* Set no of Tx, Rx queues to be used */ +static int nicvf_set_channels(struct net_device *dev, + struct ethtool_channels *channel) +{ + struct nicvf *nic = netdev_priv(dev); + int err = 0; + + if (!channel->rx_count || !channel->tx_count) + return -EINVAL; + if (channel->rx_count > MAX_RCV_QUEUES_PER_QS) + return -EINVAL; + if (channel->tx_count > MAX_SND_QUEUES_PER_QS) + return -EINVAL; + + nic->qs->rq_cnt = channel->rx_count; + nic->qs->sq_cnt = channel->tx_count; + nic->qs->cq_cnt = max(nic->qs->rq_cnt, nic->qs->sq_cnt); + + err = nicvf_set_real_num_queues(dev, nic->qs->sq_cnt, nic->qs->rq_cnt); + if (err) + return err; + + if (!netif_running(dev)) + return err; + + nicvf_stop(dev); + nicvf_open(dev); + netdev_info(dev, "Setting num Tx rings to %d, Rx rings to %d success\n", + nic->qs->sq_cnt, nic->qs->rq_cnt); + + return err; +} + +static const struct ethtool_ops nicvf_ethtool_ops = { + .get_settings = nicvf_get_settings, + .get_link = ethtool_op_get_link, + .get_drvinfo = nicvf_get_drvinfo, + .get_msglevel = nicvf_get_msglevel, + .set_msglevel = nicvf_set_msglevel, + .get_strings = nicvf_get_strings, + .get_sset_count = nicvf_get_sset_count, + .get_ethtool_stats = nicvf_get_ethtool_stats, + .get_regs_len = nicvf_get_regs_len, + .get_regs = nicvf_get_regs, + .get_coalesce = nicvf_get_coalesce, + .get_ringparam = nicvf_get_ringparam, + .get_rxnfc = nicvf_get_rxnfc, + .set_rxnfc = nicvf_set_rxnfc, + .get_rxfh_key_size = nicvf_get_rxfh_key_size, + .get_rxfh_indir_size = nicvf_get_rxfh_indir_size, + .get_rxfh = nicvf_get_rxfh, + .set_rxfh = nicvf_set_rxfh, + .get_channels = nicvf_get_channels, + .set_channels = nicvf_set_channels, + .get_ts_info = ethtool_op_get_ts_info, +}; + +void nicvf_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &nicvf_ethtool_ops; +} diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c new file mode 100644 index 000000000000..abd446e6155b --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -0,0 +1,1332 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/ethtool.h> +#include <linux/log2.h> +#include <linux/prefetch.h> +#include <linux/irq.h> + +#include "nic_reg.h" +#include "nic.h" +#include "nicvf_queues.h" +#include "thunder_bgx.h" + +#define DRV_NAME "thunder-nicvf" +#define DRV_VERSION "1.0" + +/* Supported devices */ +static const struct pci_device_id nicvf_id_table[] = { + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, + PCI_DEVICE_ID_THUNDER_NIC_VF, + PCI_VENDOR_ID_CAVIUM, 0xA11E) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, + PCI_DEVICE_ID_THUNDER_PASS1_NIC_VF, + PCI_VENDOR_ID_CAVIUM, 0xA11E) }, + { 0, } /* end of table */ +}; + +MODULE_AUTHOR("Sunil Goutham"); +MODULE_DESCRIPTION("Cavium Thunder NIC Virtual Function Driver"); +MODULE_LICENSE("GPL v2"); +MODULE_VERSION(DRV_VERSION); +MODULE_DEVICE_TABLE(pci, nicvf_id_table); + +static int debug = 0x00; +module_param(debug, int, 0644); +MODULE_PARM_DESC(debug, "Debug message level bitmap"); + +static int cpi_alg = CPI_ALG_NONE; +module_param(cpi_alg, int, S_IRUGO); +MODULE_PARM_DESC(cpi_alg, + "PFC algorithm (0=none, 1=VLAN, 2=VLAN16, 3=IP Diffserv)"); + +static int nicvf_enable_msix(struct nicvf *nic); +static netdev_tx_t nicvf_xmit(struct sk_buff *skb, struct net_device *netdev); +static void nicvf_read_bgx_stats(struct nicvf *nic, struct bgx_stats_msg *bgx); + +static inline void nicvf_set_rx_frame_cnt(struct nicvf *nic, + struct sk_buff *skb) +{ + if (skb->len <= 64) + nic->drv_stats.rx_frames_64++; + else if (skb->len <= 127) + nic->drv_stats.rx_frames_127++; + else if (skb->len <= 255) + nic->drv_stats.rx_frames_255++; + else if (skb->len <= 511) + nic->drv_stats.rx_frames_511++; + else if (skb->len <= 1023) + nic->drv_stats.rx_frames_1023++; + else if (skb->len <= 1518) + nic->drv_stats.rx_frames_1518++; + else + nic->drv_stats.rx_frames_jumbo++; +} + +/* The Cavium ThunderX network controller can *only* be found in SoCs + * containing the ThunderX ARM64 CPU implementation. All accesses to the device + * registers on this platform are implicitly strongly ordered with respect + * to memory accesses. So writeq_relaxed() and readq_relaxed() are safe to use + * with no memory barriers in this driver. The readq()/writeq() functions add + * explicit ordering operation which in this case are redundant, and only + * add overhead. + */ + +/* Register read/write APIs */ +void nicvf_reg_write(struct nicvf *nic, u64 offset, u64 val) +{ + writeq_relaxed(val, nic->reg_base + offset); +} + +u64 nicvf_reg_read(struct nicvf *nic, u64 offset) +{ + return readq_relaxed(nic->reg_base + offset); +} + +void nicvf_queue_reg_write(struct nicvf *nic, u64 offset, + u64 qidx, u64 val) +{ + void __iomem *addr = nic->reg_base + offset; + + writeq_relaxed(val, addr + (qidx << NIC_Q_NUM_SHIFT)); +} + +u64 nicvf_queue_reg_read(struct nicvf *nic, u64 offset, u64 qidx) +{ + void __iomem *addr = nic->reg_base + offset; + + return readq_relaxed(addr + (qidx << NIC_Q_NUM_SHIFT)); +} + +/* VF -> PF mailbox communication */ + +int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx) +{ + int timeout = NIC_MBOX_MSG_TIMEOUT; + int sleep = 10; + u64 *msg = (u64 *)mbx; + + nic->pf_acked = false; + nic->pf_nacked = false; + + nicvf_reg_write(nic, NIC_VF_PF_MAILBOX_0_1 + 0, msg[0]); + nicvf_reg_write(nic, NIC_VF_PF_MAILBOX_0_1 + 8, msg[1]); + + /* Wait for previous message to be acked, timeout 2sec */ + while (!nic->pf_acked) { + if (nic->pf_nacked) + return -EINVAL; + msleep(sleep); + if (nic->pf_acked) + break; + timeout -= sleep; + if (!timeout) { + netdev_err(nic->netdev, + "PF didn't ack to mbox msg %d from VF%d\n", + (mbx->msg.msg & 0xFF), nic->vf_id); + return -EBUSY; + } + } + return 0; +} + +/* Checks if VF is able to comminicate with PF +* and also gets the VNIC number this VF is associated to. +*/ +static int nicvf_check_pf_ready(struct nicvf *nic) +{ + int timeout = 5000, sleep = 20; + + nic->pf_ready_to_rcv_msg = false; + + nicvf_reg_write(nic, NIC_VF_PF_MAILBOX_0_1 + 0, + le64_to_cpu(NIC_MBOX_MSG_READY)); + nicvf_reg_write(nic, NIC_VF_PF_MAILBOX_0_1 + 8, 1ULL); + + while (!nic->pf_ready_to_rcv_msg) { + msleep(sleep); + if (nic->pf_ready_to_rcv_msg) + break; + timeout -= sleep; + if (!timeout) { + netdev_err(nic->netdev, + "PF didn't respond to READY msg\n"); + return 0; + } + } + return 1; +} + +static void nicvf_handle_mbx_intr(struct nicvf *nic) +{ + union nic_mbx mbx = {}; + u64 *mbx_data; + u64 mbx_addr; + int i; + + mbx_addr = NIC_VF_PF_MAILBOX_0_1; + mbx_data = (u64 *)&mbx; + + for (i = 0; i < NIC_PF_VF_MAILBOX_SIZE; i++) { + *mbx_data = nicvf_reg_read(nic, mbx_addr); + mbx_data++; + mbx_addr += sizeof(u64); + } + + netdev_dbg(nic->netdev, "Mbox message: msg: 0x%x\n", mbx.msg.msg); + switch (mbx.msg.msg) { + case NIC_MBOX_MSG_READY: + nic->pf_ready_to_rcv_msg = true; + nic->vf_id = mbx.nic_cfg.vf_id & 0x7F; + nic->tns_mode = mbx.nic_cfg.tns_mode & 0x7F; + nic->node = mbx.nic_cfg.node_id; + ether_addr_copy(nic->netdev->dev_addr, + (u8 *)&mbx.nic_cfg.mac_addr); + nic->link_up = false; + nic->duplex = 0; + nic->speed = 0; + break; + case NIC_MBOX_MSG_ACK: + nic->pf_acked = true; + break; + case NIC_MBOX_MSG_NACK: + nic->pf_nacked = true; + break; + case NIC_MBOX_MSG_RSS_SIZE: + nic->rss_info.rss_size = mbx.rss_size.ind_tbl_size; + nic->pf_acked = true; + break; + case NIC_MBOX_MSG_BGX_STATS: + nicvf_read_bgx_stats(nic, &mbx.bgx_stats); + nic->pf_acked = true; + nic->bgx_stats_acked = true; + break; + case NIC_MBOX_MSG_BGX_LINK_CHANGE: + nic->pf_acked = true; + nic->link_up = mbx.link_status.link_up; + nic->duplex = mbx.link_status.duplex; + nic->speed = mbx.link_status.speed; + if (nic->link_up) { + netdev_info(nic->netdev, "%s: Link is Up %d Mbps %s\n", + nic->netdev->name, nic->speed, + nic->duplex == DUPLEX_FULL ? + "Full duplex" : "Half duplex"); + netif_carrier_on(nic->netdev); + netif_tx_wake_all_queues(nic->netdev); + } else { + netdev_info(nic->netdev, "%s: Link is Down\n", + nic->netdev->name); + netif_carrier_off(nic->netdev); + netif_tx_stop_all_queues(nic->netdev); + } + break; + default: + netdev_err(nic->netdev, + "Invalid message from PF, msg 0x%x\n", mbx.msg.msg); + break; + } + nicvf_clear_intr(nic, NICVF_INTR_MBOX, 0); +} + +static int nicvf_hw_set_mac_addr(struct nicvf *nic, struct net_device *netdev) +{ + union nic_mbx mbx = {}; + int i; + + mbx.mac.msg = NIC_MBOX_MSG_SET_MAC; + mbx.mac.vf_id = nic->vf_id; + for (i = 0; i < ETH_ALEN; i++) + mbx.mac.addr = (mbx.mac.addr << 8) | + netdev->dev_addr[i]; + + return nicvf_send_msg_to_pf(nic, &mbx); +} + +void nicvf_config_cpi(struct nicvf *nic) +{ + union nic_mbx mbx = {}; + + mbx.cpi_cfg.msg = NIC_MBOX_MSG_CPI_CFG; + mbx.cpi_cfg.vf_id = nic->vf_id; + mbx.cpi_cfg.cpi_alg = nic->cpi_alg; + mbx.cpi_cfg.rq_cnt = nic->qs->rq_cnt; + + nicvf_send_msg_to_pf(nic, &mbx); +} + +void nicvf_get_rss_size(struct nicvf *nic) +{ + union nic_mbx mbx = {}; + + mbx.rss_size.msg = NIC_MBOX_MSG_RSS_SIZE; + mbx.rss_size.vf_id = nic->vf_id; + nicvf_send_msg_to_pf(nic, &mbx); +} + +void nicvf_config_rss(struct nicvf *nic) +{ + union nic_mbx mbx = {}; + struct nicvf_rss_info *rss = &nic->rss_info; + int ind_tbl_len = rss->rss_size; + int i, nextq = 0; + + mbx.rss_cfg.vf_id = nic->vf_id; + mbx.rss_cfg.hash_bits = rss->hash_bits; + while (ind_tbl_len) { + mbx.rss_cfg.tbl_offset = nextq; + mbx.rss_cfg.tbl_len = min(ind_tbl_len, + RSS_IND_TBL_LEN_PER_MBX_MSG); + mbx.rss_cfg.msg = mbx.rss_cfg.tbl_offset ? + NIC_MBOX_MSG_RSS_CFG_CONT : NIC_MBOX_MSG_RSS_CFG; + + for (i = 0; i < mbx.rss_cfg.tbl_len; i++) + mbx.rss_cfg.ind_tbl[i] = rss->ind_tbl[nextq++]; + + nicvf_send_msg_to_pf(nic, &mbx); + + ind_tbl_len -= mbx.rss_cfg.tbl_len; + } +} + +void nicvf_set_rss_key(struct nicvf *nic) +{ + struct nicvf_rss_info *rss = &nic->rss_info; + u64 key_addr = NIC_VNIC_RSS_KEY_0_4; + int idx; + + for (idx = 0; idx < RSS_HASH_KEY_SIZE; idx++) { + nicvf_reg_write(nic, key_addr, rss->key[idx]); + key_addr += sizeof(u64); + } +} + +static int nicvf_rss_init(struct nicvf *nic) +{ + struct nicvf_rss_info *rss = &nic->rss_info; + int idx; + + nicvf_get_rss_size(nic); + + if ((nic->qs->rq_cnt <= 1) || (cpi_alg != CPI_ALG_NONE)) { + rss->enable = false; + rss->hash_bits = 0; + return 0; + } + + rss->enable = true; + + /* Using the HW reset value for now */ + rss->key[0] = 0xFEED0BADFEED0BAD; + rss->key[1] = 0xFEED0BADFEED0BAD; + rss->key[2] = 0xFEED0BADFEED0BAD; + rss->key[3] = 0xFEED0BADFEED0BAD; + rss->key[4] = 0xFEED0BADFEED0BAD; + + nicvf_set_rss_key(nic); + + rss->cfg = RSS_IP_HASH_ENA | RSS_TCP_HASH_ENA | RSS_UDP_HASH_ENA; + nicvf_reg_write(nic, NIC_VNIC_RSS_CFG, rss->cfg); + + rss->hash_bits = ilog2(rounddown_pow_of_two(rss->rss_size)); + + for (idx = 0; idx < rss->rss_size; idx++) + rss->ind_tbl[idx] = ethtool_rxfh_indir_default(idx, + nic->qs->rq_cnt); + nicvf_config_rss(nic); + return 1; +} + +int nicvf_set_real_num_queues(struct net_device *netdev, + int tx_queues, int rx_queues) +{ + int err = 0; + + err = netif_set_real_num_tx_queues(netdev, tx_queues); + if (err) { + netdev_err(netdev, + "Failed to set no of Tx queues: %d\n", tx_queues); + return err; + } + + err = netif_set_real_num_rx_queues(netdev, rx_queues); + if (err) + netdev_err(netdev, + "Failed to set no of Rx queues: %d\n", rx_queues); + return err; +} + +static int nicvf_init_resources(struct nicvf *nic) +{ + int err; + u64 mbx_addr = NIC_VF_PF_MAILBOX_0_1; + + /* Enable Qset */ + nicvf_qset_config(nic, true); + + /* Initialize queues and HW for data transfer */ + err = nicvf_config_data_transfer(nic, true); + if (err) { + netdev_err(nic->netdev, + "Failed to alloc/config VF's QSet resources\n"); + return err; + } + + /* Send VF config done msg to PF */ + nicvf_reg_write(nic, mbx_addr, le64_to_cpu(NIC_MBOX_MSG_CFG_DONE)); + mbx_addr += (NIC_PF_VF_MAILBOX_SIZE - 1) * 8; + nicvf_reg_write(nic, mbx_addr, 1ULL); + + return 0; +} + +static void nicvf_snd_pkt_handler(struct net_device *netdev, + struct cmp_queue *cq, + struct cqe_send_t *cqe_tx, int cqe_type) +{ + struct sk_buff *skb = NULL; + struct nicvf *nic = netdev_priv(netdev); + struct snd_queue *sq; + struct sq_hdr_subdesc *hdr; + + sq = &nic->qs->sq[cqe_tx->sq_idx]; + + hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, cqe_tx->sqe_ptr); + if (hdr->subdesc_type != SQ_DESC_TYPE_HEADER) + return; + + netdev_dbg(nic->netdev, + "%s Qset #%d SQ #%d SQ ptr #%d subdesc count %d\n", + __func__, cqe_tx->sq_qs, cqe_tx->sq_idx, + cqe_tx->sqe_ptr, hdr->subdesc_cnt); + + nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); + nicvf_check_cqe_tx_errs(nic, cq, cqe_tx); + skb = (struct sk_buff *)sq->skbuff[cqe_tx->sqe_ptr]; + /* For TSO offloaded packets only one head SKB needs to be freed */ + if (skb) { + prefetch(skb); + dev_consume_skb_any(skb); + } +} + +static void nicvf_rcv_pkt_handler(struct net_device *netdev, + struct napi_struct *napi, + struct cmp_queue *cq, + struct cqe_rx_t *cqe_rx, int cqe_type) +{ + struct sk_buff *skb; + struct nicvf *nic = netdev_priv(netdev); + int err = 0; + + /* Check for errors */ + err = nicvf_check_cqe_rx_errs(nic, cq, cqe_rx); + if (err && !cqe_rx->rb_cnt) + return; + + skb = nicvf_get_rcv_skb(nic, cqe_rx); + if (!skb) { + netdev_dbg(nic->netdev, "Packet not received\n"); + return; + } + + if (netif_msg_pktdata(nic)) { + netdev_info(nic->netdev, "%s: skb 0x%p, len=%d\n", netdev->name, + skb, skb->len); + print_hex_dump(KERN_INFO, "", DUMP_PREFIX_OFFSET, 16, 1, + skb->data, skb->len, true); + } + + nicvf_set_rx_frame_cnt(nic, skb); + + skb_record_rx_queue(skb, cqe_rx->rq_idx); + if (netdev->hw_features & NETIF_F_RXCSUM) { + /* HW by default verifies TCP/UDP/SCTP checksums */ + skb->ip_summed = CHECKSUM_UNNECESSARY; + } else { + skb_checksum_none_assert(skb); + } + + skb->protocol = eth_type_trans(skb, netdev); + + if (napi && (netdev->features & NETIF_F_GRO)) + napi_gro_receive(napi, skb); + else + netif_receive_skb(skb); +} + +static int nicvf_cq_intr_handler(struct net_device *netdev, u8 cq_idx, + struct napi_struct *napi, int budget) +{ + int processed_cqe, work_done = 0; + int cqe_count, cqe_head; + struct nicvf *nic = netdev_priv(netdev); + struct queue_set *qs = nic->qs; + struct cmp_queue *cq = &qs->cq[cq_idx]; + struct cqe_rx_t *cq_desc; + + spin_lock_bh(&cq->lock); +loop: + processed_cqe = 0; + /* Get no of valid CQ entries to process */ + cqe_count = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_STATUS, cq_idx); + cqe_count &= CQ_CQE_COUNT; + if (!cqe_count) + goto done; + + /* Get head of the valid CQ entries */ + cqe_head = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_HEAD, cq_idx) >> 9; + cqe_head &= 0xFFFF; + + netdev_dbg(nic->netdev, "%s cqe_count %d cqe_head %d\n", + __func__, cqe_count, cqe_head); + while (processed_cqe < cqe_count) { + /* Get the CQ descriptor */ + cq_desc = (struct cqe_rx_t *)GET_CQ_DESC(cq, cqe_head); + cqe_head++; + cqe_head &= (cq->dmem.q_len - 1); + /* Initiate prefetch for next descriptor */ + prefetch((struct cqe_rx_t *)GET_CQ_DESC(cq, cqe_head)); + + if ((work_done >= budget) && napi && + (cq_desc->cqe_type != CQE_TYPE_SEND)) { + break; + } + + netdev_dbg(nic->netdev, "cq_desc->cqe_type %d\n", + cq_desc->cqe_type); + switch (cq_desc->cqe_type) { + case CQE_TYPE_RX: + nicvf_rcv_pkt_handler(netdev, napi, cq, + cq_desc, CQE_TYPE_RX); + work_done++; + break; + case CQE_TYPE_SEND: + nicvf_snd_pkt_handler(netdev, cq, + (void *)cq_desc, CQE_TYPE_SEND); + break; + case CQE_TYPE_INVALID: + case CQE_TYPE_RX_SPLIT: + case CQE_TYPE_RX_TCP: + case CQE_TYPE_SEND_PTP: + /* Ignore for now */ + break; + } + processed_cqe++; + } + netdev_dbg(nic->netdev, "%s processed_cqe %d work_done %d budget %d\n", + __func__, processed_cqe, work_done, budget); + + /* Ring doorbell to inform H/W to reuse processed CQEs */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_DOOR, + cq_idx, processed_cqe); + + if ((work_done < budget) && napi) + goto loop; + +done: + spin_unlock_bh(&cq->lock); + return work_done; +} + +static int nicvf_poll(struct napi_struct *napi, int budget) +{ + u64 cq_head; + int work_done = 0; + struct net_device *netdev = napi->dev; + struct nicvf *nic = netdev_priv(netdev); + struct nicvf_cq_poll *cq; + struct netdev_queue *txq; + + cq = container_of(napi, struct nicvf_cq_poll, napi); + work_done = nicvf_cq_intr_handler(netdev, cq->cq_idx, napi, budget); + + txq = netdev_get_tx_queue(netdev, cq->cq_idx); + if (netif_tx_queue_stopped(txq)) + netif_tx_wake_queue(txq); + + if (work_done < budget) { + /* Slow packet rate, exit polling */ + napi_complete(napi); + /* Re-enable interrupts */ + cq_head = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_HEAD, + cq->cq_idx); + nicvf_clear_intr(nic, NICVF_INTR_CQ, cq->cq_idx); + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_HEAD, + cq->cq_idx, cq_head); + nicvf_enable_intr(nic, NICVF_INTR_CQ, cq->cq_idx); + } + return work_done; +} + +/* Qset error interrupt handler + * + * As of now only CQ errors are handled + */ +void nicvf_handle_qs_err(unsigned long data) +{ + struct nicvf *nic = (struct nicvf *)data; + struct queue_set *qs = nic->qs; + int qidx; + u64 status; + + netif_tx_disable(nic->netdev); + + /* Check if it is CQ err */ + for (qidx = 0; qidx < qs->cq_cnt; qidx++) { + status = nicvf_queue_reg_read(nic, NIC_QSET_CQ_0_7_STATUS, + qidx); + if (!(status & CQ_ERR_MASK)) + continue; + /* Process already queued CQEs and reconfig CQ */ + nicvf_disable_intr(nic, NICVF_INTR_CQ, qidx); + nicvf_sq_disable(nic, qidx); + nicvf_cq_intr_handler(nic->netdev, qidx, NULL, 0); + nicvf_cmp_queue_config(nic, qs, qidx, true); + nicvf_sq_free_used_descs(nic->netdev, &qs->sq[qidx], qidx); + nicvf_sq_enable(nic, &qs->sq[qidx], qidx); + + nicvf_enable_intr(nic, NICVF_INTR_CQ, qidx); + } + + netif_tx_start_all_queues(nic->netdev); + /* Re-enable Qset error interrupt */ + nicvf_enable_intr(nic, NICVF_INTR_QS_ERR, 0); +} + +static irqreturn_t nicvf_misc_intr_handler(int irq, void *nicvf_irq) +{ + struct nicvf *nic = (struct nicvf *)nicvf_irq; + u64 intr; + + intr = nicvf_reg_read(nic, NIC_VF_INT); + /* Check for spurious interrupt */ + if (!(intr & NICVF_INTR_MBOX_MASK)) + return IRQ_HANDLED; + + nicvf_handle_mbx_intr(nic); + + return IRQ_HANDLED; +} + +static irqreturn_t nicvf_intr_handler(int irq, void *nicvf_irq) +{ + u64 qidx, intr, clear_intr = 0; + u64 cq_intr, rbdr_intr, qs_err_intr; + struct nicvf *nic = (struct nicvf *)nicvf_irq; + struct queue_set *qs = nic->qs; + struct nicvf_cq_poll *cq_poll = NULL; + + intr = nicvf_reg_read(nic, NIC_VF_INT); + if (netif_msg_intr(nic)) + netdev_info(nic->netdev, "%s: interrupt status 0x%llx\n", + nic->netdev->name, intr); + + qs_err_intr = intr & NICVF_INTR_QS_ERR_MASK; + if (qs_err_intr) { + /* Disable Qset err interrupt and schedule softirq */ + nicvf_disable_intr(nic, NICVF_INTR_QS_ERR, 0); + tasklet_hi_schedule(&nic->qs_err_task); + clear_intr |= qs_err_intr; + } + + /* Disable interrupts and start polling */ + cq_intr = (intr & NICVF_INTR_CQ_MASK) >> NICVF_INTR_CQ_SHIFT; + for (qidx = 0; qidx < qs->cq_cnt; qidx++) { + if (!(cq_intr & (1 << qidx))) + continue; + if (!nicvf_is_intr_enabled(nic, NICVF_INTR_CQ, qidx)) + continue; + + nicvf_disable_intr(nic, NICVF_INTR_CQ, qidx); + clear_intr |= ((1 << qidx) << NICVF_INTR_CQ_SHIFT); + + cq_poll = nic->napi[qidx]; + /* Schedule NAPI */ + if (cq_poll) + napi_schedule(&cq_poll->napi); + } + + /* Handle RBDR interrupts */ + rbdr_intr = (intr & NICVF_INTR_RBDR_MASK) >> NICVF_INTR_RBDR_SHIFT; + if (rbdr_intr) { + /* Disable RBDR interrupt and schedule softirq */ + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) { + if (!nicvf_is_intr_enabled(nic, NICVF_INTR_RBDR, qidx)) + continue; + nicvf_disable_intr(nic, NICVF_INTR_RBDR, qidx); + tasklet_hi_schedule(&nic->rbdr_task); + clear_intr |= ((1 << qidx) << NICVF_INTR_RBDR_SHIFT); + } + } + + /* Clear interrupts */ + nicvf_reg_write(nic, NIC_VF_INT, clear_intr); + return IRQ_HANDLED; +} + +static int nicvf_enable_msix(struct nicvf *nic) +{ + int ret, vec; + + nic->num_vec = NIC_VF_MSIX_VECTORS; + + for (vec = 0; vec < nic->num_vec; vec++) + nic->msix_entries[vec].entry = vec; + + ret = pci_enable_msix(nic->pdev, nic->msix_entries, nic->num_vec); + if (ret) { + netdev_err(nic->netdev, + "Req for #%d msix vectors failed\n", nic->num_vec); + return 0; + } + nic->msix_enabled = 1; + return 1; +} + +static void nicvf_disable_msix(struct nicvf *nic) +{ + if (nic->msix_enabled) { + pci_disable_msix(nic->pdev); + nic->msix_enabled = 0; + nic->num_vec = 0; + } +} + +static int nicvf_register_interrupts(struct nicvf *nic) +{ + int irq, free, ret = 0; + int vector; + + for_each_cq_irq(irq) + sprintf(nic->irq_name[irq], "NICVF%d CQ%d", + nic->vf_id, irq); + + for_each_sq_irq(irq) + sprintf(nic->irq_name[irq], "NICVF%d SQ%d", + nic->vf_id, irq - NICVF_INTR_ID_SQ); + + for_each_rbdr_irq(irq) + sprintf(nic->irq_name[irq], "NICVF%d RBDR%d", + nic->vf_id, irq - NICVF_INTR_ID_RBDR); + + /* Register all interrupts except mailbox */ + for (irq = 0; irq < NICVF_INTR_ID_SQ; irq++) { + vector = nic->msix_entries[irq].vector; + ret = request_irq(vector, nicvf_intr_handler, + 0, nic->irq_name[irq], nic); + if (ret) + break; + nic->irq_allocated[irq] = true; + } + + for (irq = NICVF_INTR_ID_SQ; irq < NICVF_INTR_ID_MISC; irq++) { + vector = nic->msix_entries[irq].vector; + ret = request_irq(vector, nicvf_intr_handler, + 0, nic->irq_name[irq], nic); + if (ret) + break; + nic->irq_allocated[irq] = true; + } + + sprintf(nic->irq_name[NICVF_INTR_ID_QS_ERR], + "NICVF%d Qset error", nic->vf_id); + if (!ret) { + vector = nic->msix_entries[NICVF_INTR_ID_QS_ERR].vector; + irq = NICVF_INTR_ID_QS_ERR; + ret = request_irq(vector, nicvf_intr_handler, + 0, nic->irq_name[irq], nic); + if (!ret) + nic->irq_allocated[irq] = true; + } + + if (ret) { + netdev_err(nic->netdev, "Request irq failed\n"); + for (free = 0; free < irq; free++) + free_irq(nic->msix_entries[free].vector, nic); + return ret; + } + + return 0; +} + +static void nicvf_unregister_interrupts(struct nicvf *nic) +{ + int irq; + + /* Free registered interrupts */ + for (irq = 0; irq < nic->num_vec; irq++) { + if (nic->irq_allocated[irq]) + free_irq(nic->msix_entries[irq].vector, nic); + nic->irq_allocated[irq] = false; + } + + /* Disable MSI-X */ + nicvf_disable_msix(nic); +} + +/* Initialize MSIX vectors and register MISC interrupt. + * Send READY message to PF to check if its alive + */ +static int nicvf_register_misc_interrupt(struct nicvf *nic) +{ + int ret = 0; + int irq = NICVF_INTR_ID_MISC; + + /* Return if mailbox interrupt is already registered */ + if (nic->msix_enabled) + return 0; + + /* Enable MSI-X */ + if (!nicvf_enable_msix(nic)) + return 1; + + sprintf(nic->irq_name[irq], "%s Mbox", "NICVF"); + /* Register Misc interrupt */ + ret = request_irq(nic->msix_entries[irq].vector, + nicvf_misc_intr_handler, 0, nic->irq_name[irq], nic); + + if (ret) + return ret; + nic->irq_allocated[irq] = true; + + /* Enable mailbox interrupt */ + nicvf_enable_intr(nic, NICVF_INTR_MBOX, 0); + + /* Check if VF is able to communicate with PF */ + if (!nicvf_check_pf_ready(nic)) { + nicvf_disable_intr(nic, NICVF_INTR_MBOX, 0); + nicvf_unregister_interrupts(nic); + return 1; + } + + return 0; +} + +static netdev_tx_t nicvf_xmit(struct sk_buff *skb, struct net_device *netdev) +{ + struct nicvf *nic = netdev_priv(netdev); + int qid = skb_get_queue_mapping(skb); + struct netdev_queue *txq = netdev_get_tx_queue(netdev, qid); + + /* Check for minimum packet length */ + if (skb->len <= ETH_HLEN) { + dev_kfree_skb(skb); + return NETDEV_TX_OK; + } + + if (!nicvf_sq_append_skb(nic, skb) && !netif_tx_queue_stopped(txq)) { + netif_tx_stop_queue(txq); + nic->drv_stats.tx_busy++; + if (netif_msg_tx_err(nic)) + netdev_warn(netdev, + "%s: Transmit ring full, stopping SQ%d\n", + netdev->name, qid); + + return NETDEV_TX_BUSY; + } + + return NETDEV_TX_OK; +} + +int nicvf_stop(struct net_device *netdev) +{ + int irq, qidx; + struct nicvf *nic = netdev_priv(netdev); + struct queue_set *qs = nic->qs; + struct nicvf_cq_poll *cq_poll = NULL; + union nic_mbx mbx = {}; + + mbx.msg.msg = NIC_MBOX_MSG_SHUTDOWN; + nicvf_send_msg_to_pf(nic, &mbx); + + netif_carrier_off(netdev); + netif_tx_disable(netdev); + + /* Disable RBDR & QS error interrupts */ + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) { + nicvf_disable_intr(nic, NICVF_INTR_RBDR, qidx); + nicvf_clear_intr(nic, NICVF_INTR_RBDR, qidx); + } + nicvf_disable_intr(nic, NICVF_INTR_QS_ERR, 0); + nicvf_clear_intr(nic, NICVF_INTR_QS_ERR, 0); + + /* Wait for pending IRQ handlers to finish */ + for (irq = 0; irq < nic->num_vec; irq++) + synchronize_irq(nic->msix_entries[irq].vector); + + tasklet_kill(&nic->rbdr_task); + tasklet_kill(&nic->qs_err_task); + if (nic->rb_work_scheduled) + cancel_delayed_work_sync(&nic->rbdr_work); + + for (qidx = 0; qidx < nic->qs->cq_cnt; qidx++) { + cq_poll = nic->napi[qidx]; + if (!cq_poll) + continue; + nic->napi[qidx] = NULL; + napi_synchronize(&cq_poll->napi); + /* CQ intr is enabled while napi_complete, + * so disable it now + */ + nicvf_disable_intr(nic, NICVF_INTR_CQ, qidx); + nicvf_clear_intr(nic, NICVF_INTR_CQ, qidx); + napi_disable(&cq_poll->napi); + netif_napi_del(&cq_poll->napi); + kfree(cq_poll); + } + + /* Free resources */ + nicvf_config_data_transfer(nic, false); + + /* Disable HW Qset */ + nicvf_qset_config(nic, false); + + /* disable mailbox interrupt */ + nicvf_disable_intr(nic, NICVF_INTR_MBOX, 0); + + nicvf_unregister_interrupts(nic); + + return 0; +} + +int nicvf_open(struct net_device *netdev) +{ + int err, qidx; + struct nicvf *nic = netdev_priv(netdev); + struct queue_set *qs = nic->qs; + struct nicvf_cq_poll *cq_poll = NULL; + + nic->mtu = netdev->mtu; + + netif_carrier_off(netdev); + + err = nicvf_register_misc_interrupt(nic); + if (err) + return err; + + /* Register NAPI handler for processing CQEs */ + for (qidx = 0; qidx < qs->cq_cnt; qidx++) { + cq_poll = kzalloc(sizeof(*cq_poll), GFP_KERNEL); + if (!cq_poll) { + err = -ENOMEM; + goto napi_del; + } + cq_poll->cq_idx = qidx; + netif_napi_add(netdev, &cq_poll->napi, nicvf_poll, + NAPI_POLL_WEIGHT); + napi_enable(&cq_poll->napi); + nic->napi[qidx] = cq_poll; + } + + /* Check if we got MAC address from PF or else generate a radom MAC */ + if (is_zero_ether_addr(netdev->dev_addr)) { + eth_hw_addr_random(netdev); + nicvf_hw_set_mac_addr(nic, netdev); + } + + /* Init tasklet for handling Qset err interrupt */ + tasklet_init(&nic->qs_err_task, nicvf_handle_qs_err, + (unsigned long)nic); + + /* Init RBDR tasklet which will refill RBDR */ + tasklet_init(&nic->rbdr_task, nicvf_rbdr_task, + (unsigned long)nic); + INIT_DELAYED_WORK(&nic->rbdr_work, nicvf_rbdr_work); + + /* Configure CPI alorithm */ + nic->cpi_alg = cpi_alg; + nicvf_config_cpi(nic); + + /* Configure receive side scaling */ + nicvf_rss_init(nic); + + err = nicvf_register_interrupts(nic); + if (err) + goto cleanup; + + /* Initialize the queues */ + err = nicvf_init_resources(nic); + if (err) + goto cleanup; + + /* Make sure queue initialization is written */ + wmb(); + + nicvf_reg_write(nic, NIC_VF_INT, -1); + /* Enable Qset err interrupt */ + nicvf_enable_intr(nic, NICVF_INTR_QS_ERR, 0); + + /* Enable completion queue interrupt */ + for (qidx = 0; qidx < qs->cq_cnt; qidx++) + nicvf_enable_intr(nic, NICVF_INTR_CQ, qidx); + + /* Enable RBDR threshold interrupt */ + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) + nicvf_enable_intr(nic, NICVF_INTR_RBDR, qidx); + + netif_carrier_on(netdev); + netif_tx_start_all_queues(netdev); + + return 0; +cleanup: + nicvf_disable_intr(nic, NICVF_INTR_MBOX, 0); + nicvf_unregister_interrupts(nic); +napi_del: + for (qidx = 0; qidx < qs->cq_cnt; qidx++) { + cq_poll = nic->napi[qidx]; + if (!cq_poll) + continue; + napi_disable(&cq_poll->napi); + netif_napi_del(&cq_poll->napi); + kfree(cq_poll); + nic->napi[qidx] = NULL; + } + return err; +} + +static int nicvf_update_hw_max_frs(struct nicvf *nic, int mtu) +{ + union nic_mbx mbx = {}; + + mbx.frs.msg = NIC_MBOX_MSG_SET_MAX_FRS; + mbx.frs.max_frs = mtu; + mbx.frs.vf_id = nic->vf_id; + + return nicvf_send_msg_to_pf(nic, &mbx); +} + +static int nicvf_change_mtu(struct net_device *netdev, int new_mtu) +{ + struct nicvf *nic = netdev_priv(netdev); + + if (new_mtu > NIC_HW_MAX_FRS) + return -EINVAL; + + if (new_mtu < NIC_HW_MIN_FRS) + return -EINVAL; + + if (nicvf_update_hw_max_frs(nic, new_mtu)) + return -EINVAL; + netdev->mtu = new_mtu; + nic->mtu = new_mtu; + + return 0; +} + +static int nicvf_set_mac_address(struct net_device *netdev, void *p) +{ + struct sockaddr *addr = p; + struct nicvf *nic = netdev_priv(netdev); + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); + + if (nic->msix_enabled) + if (nicvf_hw_set_mac_addr(nic, netdev)) + return -EBUSY; + + return 0; +} + +static void nicvf_read_bgx_stats(struct nicvf *nic, struct bgx_stats_msg *bgx) +{ + if (bgx->rx) + nic->bgx_stats.rx_stats[bgx->idx] = bgx->stats; + else + nic->bgx_stats.tx_stats[bgx->idx] = bgx->stats; +} + +void nicvf_update_lmac_stats(struct nicvf *nic) +{ + int stat = 0; + union nic_mbx mbx = {}; + int timeout; + + if (!netif_running(nic->netdev)) + return; + + mbx.bgx_stats.msg = NIC_MBOX_MSG_BGX_STATS; + mbx.bgx_stats.vf_id = nic->vf_id; + /* Rx stats */ + mbx.bgx_stats.rx = 1; + while (stat < BGX_RX_STATS_COUNT) { + nic->bgx_stats_acked = 0; + mbx.bgx_stats.idx = stat; + nicvf_send_msg_to_pf(nic, &mbx); + timeout = 0; + while ((!nic->bgx_stats_acked) && (timeout < 10)) { + msleep(2); + timeout++; + } + stat++; + } + + stat = 0; + + /* Tx stats */ + mbx.bgx_stats.rx = 0; + while (stat < BGX_TX_STATS_COUNT) { + nic->bgx_stats_acked = 0; + mbx.bgx_stats.idx = stat; + nicvf_send_msg_to_pf(nic, &mbx); + timeout = 0; + while ((!nic->bgx_stats_acked) && (timeout < 10)) { + msleep(2); + timeout++; + } + stat++; + } +} + +void nicvf_update_stats(struct nicvf *nic) +{ + int qidx; + struct nicvf_hw_stats *stats = &nic->stats; + struct nicvf_drv_stats *drv_stats = &nic->drv_stats; + struct queue_set *qs = nic->qs; + +#define GET_RX_STATS(reg) \ + nicvf_reg_read(nic, NIC_VNIC_RX_STAT_0_13 | (reg << 3)) +#define GET_TX_STATS(reg) \ + nicvf_reg_read(nic, NIC_VNIC_TX_STAT_0_4 | (reg << 3)) + + stats->rx_bytes_ok = GET_RX_STATS(RX_OCTS); + stats->rx_ucast_frames_ok = GET_RX_STATS(RX_UCAST); + stats->rx_bcast_frames_ok = GET_RX_STATS(RX_BCAST); + stats->rx_mcast_frames_ok = GET_RX_STATS(RX_MCAST); + stats->rx_fcs_errors = GET_RX_STATS(RX_FCS); + stats->rx_l2_errors = GET_RX_STATS(RX_L2ERR); + stats->rx_drop_red = GET_RX_STATS(RX_RED); + stats->rx_drop_overrun = GET_RX_STATS(RX_ORUN); + stats->rx_drop_bcast = GET_RX_STATS(RX_DRP_BCAST); + stats->rx_drop_mcast = GET_RX_STATS(RX_DRP_MCAST); + stats->rx_drop_l3_bcast = GET_RX_STATS(RX_DRP_L3BCAST); + stats->rx_drop_l3_mcast = GET_RX_STATS(RX_DRP_L3MCAST); + + stats->tx_bytes_ok = GET_TX_STATS(TX_OCTS); + stats->tx_ucast_frames_ok = GET_TX_STATS(TX_UCAST); + stats->tx_bcast_frames_ok = GET_TX_STATS(TX_BCAST); + stats->tx_mcast_frames_ok = GET_TX_STATS(TX_MCAST); + stats->tx_drops = GET_TX_STATS(TX_DROP); + + drv_stats->rx_frames_ok = stats->rx_ucast_frames_ok + + stats->rx_bcast_frames_ok + + stats->rx_mcast_frames_ok; + drv_stats->tx_frames_ok = stats->tx_ucast_frames_ok + + stats->tx_bcast_frames_ok + + stats->tx_mcast_frames_ok; + drv_stats->rx_drops = stats->rx_drop_red + + stats->rx_drop_overrun; + drv_stats->tx_drops = stats->tx_drops; + + /* Update RQ and SQ stats */ + for (qidx = 0; qidx < qs->rq_cnt; qidx++) + nicvf_update_rq_stats(nic, qidx); + for (qidx = 0; qidx < qs->sq_cnt; qidx++) + nicvf_update_sq_stats(nic, qidx); +} + +struct rtnl_link_stats64 *nicvf_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct nicvf *nic = netdev_priv(netdev); + struct nicvf_hw_stats *hw_stats = &nic->stats; + struct nicvf_drv_stats *drv_stats = &nic->drv_stats; + + nicvf_update_stats(nic); + + stats->rx_bytes = hw_stats->rx_bytes_ok; + stats->rx_packets = drv_stats->rx_frames_ok; + stats->rx_dropped = drv_stats->rx_drops; + + stats->tx_bytes = hw_stats->tx_bytes_ok; + stats->tx_packets = drv_stats->tx_frames_ok; + stats->tx_dropped = drv_stats->tx_drops; + + return stats; +} + +static void nicvf_tx_timeout(struct net_device *dev) +{ + struct nicvf *nic = netdev_priv(dev); + + if (netif_msg_tx_err(nic)) + netdev_warn(dev, "%s: Transmit timed out, resetting\n", + dev->name); + + schedule_work(&nic->reset_task); +} + +static void nicvf_reset_task(struct work_struct *work) +{ + struct nicvf *nic; + + nic = container_of(work, struct nicvf, reset_task); + + if (!netif_running(nic->netdev)) + return; + + nicvf_stop(nic->netdev); + nicvf_open(nic->netdev); + nic->netdev->trans_start = jiffies; +} + +static const struct net_device_ops nicvf_netdev_ops = { + .ndo_open = nicvf_open, + .ndo_stop = nicvf_stop, + .ndo_start_xmit = nicvf_xmit, + .ndo_change_mtu = nicvf_change_mtu, + .ndo_set_mac_address = nicvf_set_mac_address, + .ndo_get_stats64 = nicvf_get_stats64, + .ndo_tx_timeout = nicvf_tx_timeout, +}; + +static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + struct device *dev = &pdev->dev; + struct net_device *netdev; + struct nicvf *nic; + struct queue_set *qs; + int err; + + err = pci_enable_device(pdev); + if (err) { + dev_err(dev, "Failed to enable PCI device\n"); + return err; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + dev_err(dev, "PCI request regions failed 0x%x\n", err); + goto err_disable_device; + } + + err = pci_set_dma_mask(pdev, DMA_BIT_MASK(48)); + if (err) { + dev_err(dev, "Unable to get usable DMA configuration\n"); + goto err_release_regions; + } + + err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(48)); + if (err) { + dev_err(dev, "unable to get 48-bit DMA for consistent allocations\n"); + goto err_release_regions; + } + + netdev = alloc_etherdev_mqs(sizeof(struct nicvf), + MAX_RCV_QUEUES_PER_QS, + MAX_SND_QUEUES_PER_QS); + if (!netdev) { + err = -ENOMEM; + goto err_release_regions; + } + + pci_set_drvdata(pdev, netdev); + + SET_NETDEV_DEV(netdev, &pdev->dev); + + nic = netdev_priv(netdev); + nic->netdev = netdev; + nic->pdev = pdev; + + /* MAP VF's configuration registers */ + nic->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); + if (!nic->reg_base) { + dev_err(dev, "Cannot map config register space, aborting\n"); + err = -ENOMEM; + goto err_free_netdev; + } + + err = nicvf_set_qset_resources(nic); + if (err) + goto err_free_netdev; + + qs = nic->qs; + + err = nicvf_set_real_num_queues(netdev, qs->sq_cnt, qs->rq_cnt); + if (err) + goto err_free_netdev; + + /* Check if PF is alive and get MAC address for this VF */ + err = nicvf_register_misc_interrupt(nic); + if (err) + goto err_free_netdev; + + netdev->features |= (NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_SG | + NETIF_F_TSO | NETIF_F_GRO); + netdev->hw_features = netdev->features; + + netdev->netdev_ops = &nicvf_netdev_ops; + + INIT_WORK(&nic->reset_task, nicvf_reset_task); + + err = register_netdev(netdev); + if (err) { + dev_err(dev, "Failed to register netdevice\n"); + goto err_unregister_interrupts; + } + + nic->msg_enable = debug; + + nicvf_set_ethtool_ops(netdev); + + return 0; + +err_unregister_interrupts: + nicvf_unregister_interrupts(nic); +err_free_netdev: + pci_set_drvdata(pdev, NULL); + free_netdev(netdev); +err_release_regions: + pci_release_regions(pdev); +err_disable_device: + pci_disable_device(pdev); + return err; +} + +static void nicvf_remove(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct nicvf *nic = netdev_priv(netdev); + + unregister_netdev(netdev); + nicvf_unregister_interrupts(nic); + pci_set_drvdata(pdev, NULL); + free_netdev(netdev); + pci_release_regions(pdev); + pci_disable_device(pdev); +} + +static struct pci_driver nicvf_driver = { + .name = DRV_NAME, + .id_table = nicvf_id_table, + .probe = nicvf_probe, + .remove = nicvf_remove, +}; + +static int __init nicvf_init_module(void) +{ + pr_info("%s, ver %s\n", DRV_NAME, DRV_VERSION); + + return pci_register_driver(&nicvf_driver); +} + +static void __exit nicvf_cleanup_module(void) +{ + pci_unregister_driver(&nicvf_driver); +} + +module_init(nicvf_init_module); +module_exit(nicvf_cleanup_module); diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c new file mode 100644 index 000000000000..196246665444 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -0,0 +1,1544 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/ip.h> +#include <linux/etherdevice.h> +#include <net/ip.h> +#include <net/tso.h> + +#include "nic_reg.h" +#include "nic.h" +#include "q_struct.h" +#include "nicvf_queues.h" + +struct rbuf_info { + struct page *page; + void *data; + u64 offset; +}; + +#define GET_RBUF_INFO(x) ((struct rbuf_info *)(x - NICVF_RCV_BUF_ALIGN_BYTES)) + +/* Poll a register for a specific value */ +static int nicvf_poll_reg(struct nicvf *nic, int qidx, + u64 reg, int bit_pos, int bits, int val) +{ + u64 bit_mask; + u64 reg_val; + int timeout = 10; + + bit_mask = (1ULL << bits) - 1; + bit_mask = (bit_mask << bit_pos); + + while (timeout) { + reg_val = nicvf_queue_reg_read(nic, reg, qidx); + if (((reg_val & bit_mask) >> bit_pos) == val) + return 0; + usleep_range(1000, 2000); + timeout--; + } + netdev_err(nic->netdev, "Poll on reg 0x%llx failed\n", reg); + return 1; +} + +/* Allocate memory for a queue's descriptors */ +static int nicvf_alloc_q_desc_mem(struct nicvf *nic, struct q_desc_mem *dmem, + int q_len, int desc_size, int align_bytes) +{ + dmem->q_len = q_len; + dmem->size = (desc_size * q_len) + align_bytes; + /* Save address, need it while freeing */ + dmem->unalign_base = dma_zalloc_coherent(&nic->pdev->dev, dmem->size, + &dmem->dma, GFP_KERNEL); + if (!dmem->unalign_base) + return -ENOMEM; + + /* Align memory address for 'align_bytes' */ + dmem->phys_base = NICVF_ALIGNED_ADDR((u64)dmem->dma, align_bytes); + dmem->base = (void *)((u8 *)dmem->unalign_base + + (dmem->phys_base - dmem->dma)); + return 0; +} + +/* Free queue's descriptor memory */ +static void nicvf_free_q_desc_mem(struct nicvf *nic, struct q_desc_mem *dmem) +{ + if (!dmem) + return; + + dma_free_coherent(&nic->pdev->dev, dmem->size, + dmem->unalign_base, dmem->dma); + dmem->unalign_base = NULL; + dmem->base = NULL; +} + +/* Allocate buffer for packet reception + * HW returns memory address where packet is DMA'ed but not a pointer + * into RBDR ring, so save buffer address at the start of fragment and + * align the start address to a cache aligned address + */ +static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp, + u32 buf_len, u64 **rbuf) +{ + u64 data; + struct rbuf_info *rinfo; + int order = get_order(buf_len); + + /* Check if request can be accomodated in previous allocated page */ + if (nic->rb_page) { + if ((nic->rb_page_offset + buf_len + buf_len) > + (PAGE_SIZE << order)) { + nic->rb_page = NULL; + } else { + nic->rb_page_offset += buf_len; + get_page(nic->rb_page); + } + } + + /* Allocate a new page */ + if (!nic->rb_page) { + nic->rb_page = alloc_pages(gfp | __GFP_COMP, order); + if (!nic->rb_page) { + netdev_err(nic->netdev, "Failed to allocate new rcv buffer\n"); + return -ENOMEM; + } + nic->rb_page_offset = 0; + } + + data = (u64)page_address(nic->rb_page) + nic->rb_page_offset; + + /* Align buffer addr to cache line i.e 128 bytes */ + rinfo = (struct rbuf_info *)(data + NICVF_RCV_BUF_ALIGN_LEN(data)); + /* Save page address for reference updation */ + rinfo->page = nic->rb_page; + /* Store start address for later retrieval */ + rinfo->data = (void *)data; + /* Store alignment offset */ + rinfo->offset = NICVF_RCV_BUF_ALIGN_LEN(data); + + data += rinfo->offset; + + /* Give next aligned address to hw for DMA */ + *rbuf = (u64 *)(data + NICVF_RCV_BUF_ALIGN_BYTES); + return 0; +} + +/* Retrieve actual buffer start address and build skb for received packet */ +static struct sk_buff *nicvf_rb_ptr_to_skb(struct nicvf *nic, + u64 rb_ptr, int len) +{ + struct sk_buff *skb; + struct rbuf_info *rinfo; + + rb_ptr = (u64)phys_to_virt(rb_ptr); + /* Get buffer start address and alignment offset */ + rinfo = GET_RBUF_INFO(rb_ptr); + + /* Now build an skb to give to stack */ + skb = build_skb(rinfo->data, RCV_FRAG_LEN); + if (!skb) { + put_page(rinfo->page); + return NULL; + } + + /* Set correct skb->data */ + skb_reserve(skb, rinfo->offset + NICVF_RCV_BUF_ALIGN_BYTES); + + prefetch((void *)rb_ptr); + return skb; +} + +/* Allocate RBDR ring and populate receive buffers */ +static int nicvf_init_rbdr(struct nicvf *nic, struct rbdr *rbdr, + int ring_len, int buf_size) +{ + int idx; + u64 *rbuf; + struct rbdr_entry_t *desc; + int err; + + err = nicvf_alloc_q_desc_mem(nic, &rbdr->dmem, ring_len, + sizeof(struct rbdr_entry_t), + NICVF_RCV_BUF_ALIGN_BYTES); + if (err) + return err; + + rbdr->desc = rbdr->dmem.base; + /* Buffer size has to be in multiples of 128 bytes */ + rbdr->dma_size = buf_size; + rbdr->enable = true; + rbdr->thresh = RBDR_THRESH; + + nic->rb_page = NULL; + for (idx = 0; idx < ring_len; idx++) { + err = nicvf_alloc_rcv_buffer(nic, GFP_KERNEL, RCV_FRAG_LEN, + &rbuf); + if (err) + return err; + + desc = GET_RBDR_DESC(rbdr, idx); + desc->buf_addr = virt_to_phys(rbuf) >> NICVF_RCV_BUF_ALIGN; + } + return 0; +} + +/* Free RBDR ring and its receive buffers */ +static void nicvf_free_rbdr(struct nicvf *nic, struct rbdr *rbdr) +{ + int head, tail; + u64 buf_addr; + struct rbdr_entry_t *desc; + struct rbuf_info *rinfo; + + if (!rbdr) + return; + + rbdr->enable = false; + if (!rbdr->dmem.base) + return; + + head = rbdr->head; + tail = rbdr->tail; + + /* Free SKBs */ + while (head != tail) { + desc = GET_RBDR_DESC(rbdr, head); + buf_addr = desc->buf_addr << NICVF_RCV_BUF_ALIGN; + rinfo = GET_RBUF_INFO((u64)phys_to_virt(buf_addr)); + put_page(rinfo->page); + head++; + head &= (rbdr->dmem.q_len - 1); + } + /* Free SKB of tail desc */ + desc = GET_RBDR_DESC(rbdr, tail); + buf_addr = desc->buf_addr << NICVF_RCV_BUF_ALIGN; + rinfo = GET_RBUF_INFO((u64)phys_to_virt(buf_addr)); + put_page(rinfo->page); + + /* Free RBDR ring */ + nicvf_free_q_desc_mem(nic, &rbdr->dmem); +} + +/* Refill receive buffer descriptors with new buffers. + */ +void nicvf_refill_rbdr(struct nicvf *nic, gfp_t gfp) +{ + struct queue_set *qs = nic->qs; + int rbdr_idx = qs->rbdr_cnt; + int tail, qcount; + int refill_rb_cnt; + struct rbdr *rbdr; + struct rbdr_entry_t *desc; + u64 *rbuf; + int new_rb = 0; + +refill: + if (!rbdr_idx) + return; + rbdr_idx--; + rbdr = &qs->rbdr[rbdr_idx]; + /* Check if it's enabled */ + if (!rbdr->enable) + goto next_rbdr; + + /* Get no of desc's to be refilled */ + qcount = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_STATUS0, rbdr_idx); + qcount &= 0x7FFFF; + /* Doorbell can be ringed with a max of ring size minus 1 */ + if (qcount >= (qs->rbdr_len - 1)) + goto next_rbdr; + else + refill_rb_cnt = qs->rbdr_len - qcount - 1; + + /* Start filling descs from tail */ + tail = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_TAIL, rbdr_idx) >> 3; + while (refill_rb_cnt) { + tail++; + tail &= (rbdr->dmem.q_len - 1); + + if (nicvf_alloc_rcv_buffer(nic, gfp, RCV_FRAG_LEN, &rbuf)) + break; + + desc = GET_RBDR_DESC(rbdr, tail); + desc->buf_addr = virt_to_phys(rbuf) >> NICVF_RCV_BUF_ALIGN; + refill_rb_cnt--; + new_rb++; + } + + /* make sure all memory stores are done before ringing doorbell */ + smp_wmb(); + + /* Check if buffer allocation failed */ + if (refill_rb_cnt) + nic->rb_alloc_fail = true; + else + nic->rb_alloc_fail = false; + + /* Notify HW */ + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_DOOR, + rbdr_idx, new_rb); +next_rbdr: + /* Re-enable RBDR interrupts only if buffer allocation is success */ + if (!nic->rb_alloc_fail && rbdr->enable) + nicvf_enable_intr(nic, NICVF_INTR_RBDR, rbdr_idx); + + if (rbdr_idx) + goto refill; +} + +/* Alloc rcv buffers in non-atomic mode for better success */ +void nicvf_rbdr_work(struct work_struct *work) +{ + struct nicvf *nic = container_of(work, struct nicvf, rbdr_work.work); + + nicvf_refill_rbdr(nic, GFP_KERNEL); + if (nic->rb_alloc_fail) + schedule_delayed_work(&nic->rbdr_work, msecs_to_jiffies(10)); + else + nic->rb_work_scheduled = false; +} + +/* In Softirq context, alloc rcv buffers in atomic mode */ +void nicvf_rbdr_task(unsigned long data) +{ + struct nicvf *nic = (struct nicvf *)data; + + nicvf_refill_rbdr(nic, GFP_ATOMIC); + if (nic->rb_alloc_fail) { + nic->rb_work_scheduled = true; + schedule_delayed_work(&nic->rbdr_work, msecs_to_jiffies(10)); + } +} + +/* Initialize completion queue */ +static int nicvf_init_cmp_queue(struct nicvf *nic, + struct cmp_queue *cq, int q_len) +{ + int err; + + err = nicvf_alloc_q_desc_mem(nic, &cq->dmem, q_len, CMP_QUEUE_DESC_SIZE, + NICVF_CQ_BASE_ALIGN_BYTES); + if (err) + return err; + + cq->desc = cq->dmem.base; + cq->thresh = CMP_QUEUE_CQE_THRESH; + nic->cq_coalesce_usecs = (CMP_QUEUE_TIMER_THRESH * 0.05) - 1; + + return 0; +} + +static void nicvf_free_cmp_queue(struct nicvf *nic, struct cmp_queue *cq) +{ + if (!cq) + return; + if (!cq->dmem.base) + return; + + nicvf_free_q_desc_mem(nic, &cq->dmem); +} + +/* Initialize transmit queue */ +static int nicvf_init_snd_queue(struct nicvf *nic, + struct snd_queue *sq, int q_len) +{ + int err; + + err = nicvf_alloc_q_desc_mem(nic, &sq->dmem, q_len, SND_QUEUE_DESC_SIZE, + NICVF_SQ_BASE_ALIGN_BYTES); + if (err) + return err; + + sq->desc = sq->dmem.base; + sq->skbuff = kcalloc(q_len, sizeof(u64), GFP_ATOMIC); + sq->head = 0; + sq->tail = 0; + atomic_set(&sq->free_cnt, q_len - 1); + sq->thresh = SND_QUEUE_THRESH; + + /* Preallocate memory for TSO segment's header */ + sq->tso_hdrs = dma_alloc_coherent(&nic->pdev->dev, + q_len * TSO_HEADER_SIZE, + &sq->tso_hdrs_phys, GFP_KERNEL); + if (!sq->tso_hdrs) + return -ENOMEM; + + return 0; +} + +static void nicvf_free_snd_queue(struct nicvf *nic, struct snd_queue *sq) +{ + if (!sq) + return; + if (!sq->dmem.base) + return; + + if (sq->tso_hdrs) + dma_free_coherent(&nic->pdev->dev, sq->dmem.q_len, + sq->tso_hdrs, sq->tso_hdrs_phys); + + kfree(sq->skbuff); + nicvf_free_q_desc_mem(nic, &sq->dmem); +} + +static void nicvf_reclaim_snd_queue(struct nicvf *nic, + struct queue_set *qs, int qidx) +{ + /* Disable send queue */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, 0); + /* Check if SQ is stopped */ + if (nicvf_poll_reg(nic, qidx, NIC_QSET_SQ_0_7_STATUS, 21, 1, 0x01)) + return; + /* Reset send queue */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, NICVF_SQ_RESET); +} + +static void nicvf_reclaim_rcv_queue(struct nicvf *nic, + struct queue_set *qs, int qidx) +{ + union nic_mbx mbx = {}; + + /* Make sure all packets in the pipeline are written back into mem */ + mbx.msg.msg = NIC_MBOX_MSG_RQ_SW_SYNC; + nicvf_send_msg_to_pf(nic, &mbx); +} + +static void nicvf_reclaim_cmp_queue(struct nicvf *nic, + struct queue_set *qs, int qidx) +{ + /* Disable timer threshold (doesn't get reset upon CQ reset */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG2, qidx, 0); + /* Disable completion queue */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG, qidx, 0); + /* Reset completion queue */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG, qidx, NICVF_CQ_RESET); +} + +static void nicvf_reclaim_rbdr(struct nicvf *nic, + struct rbdr *rbdr, int qidx) +{ + u64 tmp, fifo_state; + int timeout = 10; + + /* Save head and tail pointers for feeing up buffers */ + rbdr->head = nicvf_queue_reg_read(nic, + NIC_QSET_RBDR_0_1_HEAD, + qidx) >> 3; + rbdr->tail = nicvf_queue_reg_read(nic, + NIC_QSET_RBDR_0_1_TAIL, + qidx) >> 3; + + /* If RBDR FIFO is in 'FAIL' state then do a reset first + * before relaiming. + */ + fifo_state = nicvf_queue_reg_read(nic, NIC_QSET_RBDR_0_1_STATUS0, qidx); + if (((fifo_state >> 62) & 0x03) == 0x3) + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_CFG, + qidx, NICVF_RBDR_RESET); + + /* Disable RBDR */ + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_CFG, qidx, 0); + if (nicvf_poll_reg(nic, qidx, NIC_QSET_RBDR_0_1_STATUS0, 62, 2, 0x00)) + return; + while (1) { + tmp = nicvf_queue_reg_read(nic, + NIC_QSET_RBDR_0_1_PREFETCH_STATUS, + qidx); + if ((tmp & 0xFFFFFFFF) == ((tmp >> 32) & 0xFFFFFFFF)) + break; + usleep_range(1000, 2000); + timeout--; + if (!timeout) { + netdev_err(nic->netdev, + "Failed polling on prefetch status\n"); + return; + } + } + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_CFG, + qidx, NICVF_RBDR_RESET); + + if (nicvf_poll_reg(nic, qidx, NIC_QSET_RBDR_0_1_STATUS0, 62, 2, 0x02)) + return; + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_CFG, qidx, 0x00); + if (nicvf_poll_reg(nic, qidx, NIC_QSET_RBDR_0_1_STATUS0, 62, 2, 0x00)) + return; +} + +/* Configures receive queue */ +static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs, + int qidx, bool enable) +{ + union nic_mbx mbx = {}; + struct rcv_queue *rq; + struct rq_cfg rq_cfg; + + rq = &qs->rq[qidx]; + rq->enable = enable; + + /* Disable receive queue */ + nicvf_queue_reg_write(nic, NIC_QSET_RQ_0_7_CFG, qidx, 0); + + if (!rq->enable) { + nicvf_reclaim_rcv_queue(nic, qs, qidx); + return; + } + + rq->cq_qs = qs->vnic_id; + rq->cq_idx = qidx; + rq->start_rbdr_qs = qs->vnic_id; + rq->start_qs_rbdr_idx = qs->rbdr_cnt - 1; + rq->cont_rbdr_qs = qs->vnic_id; + rq->cont_qs_rbdr_idx = qs->rbdr_cnt - 1; + /* all writes of RBDR data to be loaded into L2 Cache as well*/ + rq->caching = 1; + + /* Send a mailbox msg to PF to config RQ */ + mbx.rq.msg = NIC_MBOX_MSG_RQ_CFG; + mbx.rq.qs_num = qs->vnic_id; + mbx.rq.rq_num = qidx; + mbx.rq.cfg = (rq->caching << 26) | (rq->cq_qs << 19) | + (rq->cq_idx << 16) | (rq->cont_rbdr_qs << 9) | + (rq->cont_qs_rbdr_idx << 8) | + (rq->start_rbdr_qs << 1) | (rq->start_qs_rbdr_idx); + nicvf_send_msg_to_pf(nic, &mbx); + + mbx.rq.msg = NIC_MBOX_MSG_RQ_BP_CFG; + mbx.rq.cfg = (1ULL << 63) | (1ULL << 62) | (qs->vnic_id << 0); + nicvf_send_msg_to_pf(nic, &mbx); + + /* RQ drop config + * Enable CQ drop to reserve sufficient CQEs for all tx packets + */ + mbx.rq.msg = NIC_MBOX_MSG_RQ_DROP_CFG; + mbx.rq.cfg = (1ULL << 62) | (RQ_CQ_DROP << 8); + nicvf_send_msg_to_pf(nic, &mbx); + + nicvf_queue_reg_write(nic, NIC_QSET_RQ_GEN_CFG, qidx, 0x00); + + /* Enable Receive queue */ + rq_cfg.ena = 1; + rq_cfg.tcp_ena = 0; + nicvf_queue_reg_write(nic, NIC_QSET_RQ_0_7_CFG, qidx, *(u64 *)&rq_cfg); +} + +/* Configures completion queue */ +void nicvf_cmp_queue_config(struct nicvf *nic, struct queue_set *qs, + int qidx, bool enable) +{ + struct cmp_queue *cq; + struct cq_cfg cq_cfg; + + cq = &qs->cq[qidx]; + cq->enable = enable; + + if (!cq->enable) { + nicvf_reclaim_cmp_queue(nic, qs, qidx); + return; + } + + /* Reset completion queue */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG, qidx, NICVF_CQ_RESET); + + if (!cq->enable) + return; + + spin_lock_init(&cq->lock); + /* Set completion queue base address */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_BASE, + qidx, (u64)(cq->dmem.phys_base)); + + /* Enable Completion queue */ + cq_cfg.ena = 1; + cq_cfg.reset = 0; + cq_cfg.caching = 0; + cq_cfg.qsize = CMP_QSIZE; + cq_cfg.avg_con = 0; + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG, qidx, *(u64 *)&cq_cfg); + + /* Set threshold value for interrupt generation */ + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_THRESH, qidx, cq->thresh); + nicvf_queue_reg_write(nic, NIC_QSET_CQ_0_7_CFG2, + qidx, nic->cq_coalesce_usecs); +} + +/* Configures transmit queue */ +static void nicvf_snd_queue_config(struct nicvf *nic, struct queue_set *qs, + int qidx, bool enable) +{ + union nic_mbx mbx = {}; + struct snd_queue *sq; + struct sq_cfg sq_cfg; + + sq = &qs->sq[qidx]; + sq->enable = enable; + + if (!sq->enable) { + nicvf_reclaim_snd_queue(nic, qs, qidx); + return; + } + + /* Reset send queue */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, NICVF_SQ_RESET); + + sq->cq_qs = qs->vnic_id; + sq->cq_idx = qidx; + + /* Send a mailbox msg to PF to config SQ */ + mbx.sq.msg = NIC_MBOX_MSG_SQ_CFG; + mbx.sq.qs_num = qs->vnic_id; + mbx.sq.sq_num = qidx; + mbx.sq.cfg = (sq->cq_qs << 3) | sq->cq_idx; + nicvf_send_msg_to_pf(nic, &mbx); + + /* Set queue base address */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_BASE, + qidx, (u64)(sq->dmem.phys_base)); + + /* Enable send queue & set queue size */ + sq_cfg.ena = 1; + sq_cfg.reset = 0; + sq_cfg.ldwb = 0; + sq_cfg.qsize = SND_QSIZE; + sq_cfg.tstmp_bgx_intf = 0; + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, *(u64 *)&sq_cfg); + + /* Set threshold value for interrupt generation */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_THRESH, qidx, sq->thresh); + + /* Set queue:cpu affinity for better load distribution */ + if (cpu_online(qidx)) { + cpumask_set_cpu(qidx, &sq->affinity_mask); + netif_set_xps_queue(nic->netdev, + &sq->affinity_mask, qidx); + } +} + +/* Configures receive buffer descriptor ring */ +static void nicvf_rbdr_config(struct nicvf *nic, struct queue_set *qs, + int qidx, bool enable) +{ + struct rbdr *rbdr; + struct rbdr_cfg rbdr_cfg; + + rbdr = &qs->rbdr[qidx]; + nicvf_reclaim_rbdr(nic, rbdr, qidx); + if (!enable) + return; + + /* Set descriptor base address */ + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_BASE, + qidx, (u64)(rbdr->dmem.phys_base)); + + /* Enable RBDR & set queue size */ + /* Buffer size should be in multiples of 128 bytes */ + rbdr_cfg.ena = 1; + rbdr_cfg.reset = 0; + rbdr_cfg.ldwb = 0; + rbdr_cfg.qsize = RBDR_SIZE; + rbdr_cfg.avg_con = 0; + rbdr_cfg.lines = rbdr->dma_size / 128; + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_CFG, + qidx, *(u64 *)&rbdr_cfg); + + /* Notify HW */ + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_DOOR, + qidx, qs->rbdr_len - 1); + + /* Set threshold value for interrupt generation */ + nicvf_queue_reg_write(nic, NIC_QSET_RBDR_0_1_THRESH, + qidx, rbdr->thresh - 1); +} + +/* Requests PF to assign and enable Qset */ +void nicvf_qset_config(struct nicvf *nic, bool enable) +{ + union nic_mbx mbx = {}; + struct queue_set *qs = nic->qs; + struct qs_cfg *qs_cfg; + + if (!qs) { + netdev_warn(nic->netdev, + "Qset is still not allocated, don't init queues\n"); + return; + } + + qs->enable = enable; + qs->vnic_id = nic->vf_id; + + /* Send a mailbox msg to PF to config Qset */ + mbx.qs.msg = NIC_MBOX_MSG_QS_CFG; + mbx.qs.num = qs->vnic_id; + + mbx.qs.cfg = 0; + qs_cfg = (struct qs_cfg *)&mbx.qs.cfg; + if (qs->enable) { + qs_cfg->ena = 1; +#ifdef __BIG_ENDIAN + qs_cfg->be = 1; +#endif + qs_cfg->vnic = qs->vnic_id; + } + nicvf_send_msg_to_pf(nic, &mbx); +} + +static void nicvf_free_resources(struct nicvf *nic) +{ + int qidx; + struct queue_set *qs = nic->qs; + + /* Free receive buffer descriptor ring */ + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) + nicvf_free_rbdr(nic, &qs->rbdr[qidx]); + + /* Free completion queue */ + for (qidx = 0; qidx < qs->cq_cnt; qidx++) + nicvf_free_cmp_queue(nic, &qs->cq[qidx]); + + /* Free send queue */ + for (qidx = 0; qidx < qs->sq_cnt; qidx++) + nicvf_free_snd_queue(nic, &qs->sq[qidx]); +} + +static int nicvf_alloc_resources(struct nicvf *nic) +{ + int qidx; + struct queue_set *qs = nic->qs; + + /* Alloc receive buffer descriptor ring */ + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) { + if (nicvf_init_rbdr(nic, &qs->rbdr[qidx], qs->rbdr_len, + DMA_BUFFER_LEN)) + goto alloc_fail; + } + + /* Alloc send queue */ + for (qidx = 0; qidx < qs->sq_cnt; qidx++) { + if (nicvf_init_snd_queue(nic, &qs->sq[qidx], qs->sq_len)) + goto alloc_fail; + } + + /* Alloc completion queue */ + for (qidx = 0; qidx < qs->cq_cnt; qidx++) { + if (nicvf_init_cmp_queue(nic, &qs->cq[qidx], qs->cq_len)) + goto alloc_fail; + } + + return 0; +alloc_fail: + nicvf_free_resources(nic); + return -ENOMEM; +} + +int nicvf_set_qset_resources(struct nicvf *nic) +{ + struct queue_set *qs; + + qs = devm_kzalloc(&nic->pdev->dev, sizeof(*qs), GFP_KERNEL); + if (!qs) + return -ENOMEM; + nic->qs = qs; + + /* Set count of each queue */ + qs->rbdr_cnt = RBDR_CNT; + qs->rq_cnt = RCV_QUEUE_CNT; + qs->sq_cnt = SND_QUEUE_CNT; + qs->cq_cnt = CMP_QUEUE_CNT; + + /* Set queue lengths */ + qs->rbdr_len = RCV_BUF_COUNT; + qs->sq_len = SND_QUEUE_LEN; + qs->cq_len = CMP_QUEUE_LEN; + return 0; +} + +int nicvf_config_data_transfer(struct nicvf *nic, bool enable) +{ + bool disable = false; + struct queue_set *qs = nic->qs; + int qidx; + + if (!qs) + return 0; + + if (enable) { + if (nicvf_alloc_resources(nic)) + return -ENOMEM; + + for (qidx = 0; qidx < qs->sq_cnt; qidx++) + nicvf_snd_queue_config(nic, qs, qidx, enable); + for (qidx = 0; qidx < qs->cq_cnt; qidx++) + nicvf_cmp_queue_config(nic, qs, qidx, enable); + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) + nicvf_rbdr_config(nic, qs, qidx, enable); + for (qidx = 0; qidx < qs->rq_cnt; qidx++) + nicvf_rcv_queue_config(nic, qs, qidx, enable); + } else { + for (qidx = 0; qidx < qs->rq_cnt; qidx++) + nicvf_rcv_queue_config(nic, qs, qidx, disable); + for (qidx = 0; qidx < qs->rbdr_cnt; qidx++) + nicvf_rbdr_config(nic, qs, qidx, disable); + for (qidx = 0; qidx < qs->sq_cnt; qidx++) + nicvf_snd_queue_config(nic, qs, qidx, disable); + for (qidx = 0; qidx < qs->cq_cnt; qidx++) + nicvf_cmp_queue_config(nic, qs, qidx, disable); + + nicvf_free_resources(nic); + } + + return 0; +} + +/* Get a free desc from SQ + * returns descriptor ponter & descriptor number + */ +static inline int nicvf_get_sq_desc(struct snd_queue *sq, int desc_cnt) +{ + int qentry; + + qentry = sq->tail; + atomic_sub(desc_cnt, &sq->free_cnt); + sq->tail += desc_cnt; + sq->tail &= (sq->dmem.q_len - 1); + + return qentry; +} + +/* Free descriptor back to SQ for future use */ +void nicvf_put_sq_desc(struct snd_queue *sq, int desc_cnt) +{ + atomic_add(desc_cnt, &sq->free_cnt); + sq->head += desc_cnt; + sq->head &= (sq->dmem.q_len - 1); +} + +static inline int nicvf_get_nxt_sqentry(struct snd_queue *sq, int qentry) +{ + qentry++; + qentry &= (sq->dmem.q_len - 1); + return qentry; +} + +void nicvf_sq_enable(struct nicvf *nic, struct snd_queue *sq, int qidx) +{ + u64 sq_cfg; + + sq_cfg = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_CFG, qidx); + sq_cfg |= NICVF_SQ_EN; + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, sq_cfg); + /* Ring doorbell so that H/W restarts processing SQEs */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_DOOR, qidx, 0); +} + +void nicvf_sq_disable(struct nicvf *nic, int qidx) +{ + u64 sq_cfg; + + sq_cfg = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_CFG, qidx); + sq_cfg &= ~NICVF_SQ_EN; + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_CFG, qidx, sq_cfg); +} + +void nicvf_sq_free_used_descs(struct net_device *netdev, struct snd_queue *sq, + int qidx) +{ + u64 head, tail; + struct sk_buff *skb; + struct nicvf *nic = netdev_priv(netdev); + struct sq_hdr_subdesc *hdr; + + head = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_HEAD, qidx) >> 4; + tail = nicvf_queue_reg_read(nic, NIC_QSET_SQ_0_7_TAIL, qidx) >> 4; + while (sq->head != head) { + hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, sq->head); + if (hdr->subdesc_type != SQ_DESC_TYPE_HEADER) { + nicvf_put_sq_desc(sq, 1); + continue; + } + skb = (struct sk_buff *)sq->skbuff[sq->head]; + atomic64_add(1, (atomic64_t *)&netdev->stats.tx_packets); + atomic64_add(hdr->tot_len, + (atomic64_t *)&netdev->stats.tx_bytes); + dev_kfree_skb_any(skb); + nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); + } +} + +/* Calculate no of SQ subdescriptors needed to transmit all + * segments of this TSO packet. + * Taken from 'Tilera network driver' with a minor modification. + */ +static int nicvf_tso_count_subdescs(struct sk_buff *skb) +{ + struct skb_shared_info *sh = skb_shinfo(skb); + unsigned int sh_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + unsigned int data_len = skb->len - sh_len; + unsigned int p_len = sh->gso_size; + long f_id = -1; /* id of the current fragment */ + long f_size = skb_headlen(skb) - sh_len; /* current fragment size */ + long f_used = 0; /* bytes used from the current fragment */ + long n; /* size of the current piece of payload */ + int num_edescs = 0; + int segment; + + for (segment = 0; segment < sh->gso_segs; segment++) { + unsigned int p_used = 0; + + /* One edesc for header and for each piece of the payload. */ + for (num_edescs++; p_used < p_len; num_edescs++) { + /* Advance as needed. */ + while (f_used >= f_size) { + f_id++; + f_size = skb_frag_size(&sh->frags[f_id]); + f_used = 0; + } + + /* Use bytes from the current fragment. */ + n = p_len - p_used; + if (n > f_size - f_used) + n = f_size - f_used; + f_used += n; + p_used += n; + } + + /* The last segment may be less than gso_size. */ + data_len -= p_len; + if (data_len < p_len) + p_len = data_len; + } + + /* '+ gso_segs' for SQ_HDR_SUDESCs for each segment */ + return num_edescs + sh->gso_segs; +} + +/* Get the number of SQ descriptors needed to xmit this skb */ +static int nicvf_sq_subdesc_required(struct nicvf *nic, struct sk_buff *skb) +{ + int subdesc_cnt = MIN_SQ_DESC_PER_PKT_XMIT; + + if (skb_shinfo(skb)->gso_size) { + subdesc_cnt = nicvf_tso_count_subdescs(skb); + return subdesc_cnt; + } + + if (skb_shinfo(skb)->nr_frags) + subdesc_cnt += skb_shinfo(skb)->nr_frags; + + return subdesc_cnt; +} + +/* Add SQ HEADER subdescriptor. + * First subdescriptor for every send descriptor. + */ +static inline void +nicvf_sq_add_hdr_subdesc(struct snd_queue *sq, int qentry, + int subdesc_cnt, struct sk_buff *skb, int len) +{ + int proto; + struct sq_hdr_subdesc *hdr; + + hdr = (struct sq_hdr_subdesc *)GET_SQ_DESC(sq, qentry); + sq->skbuff[qentry] = (u64)skb; + + memset(hdr, 0, SND_QUEUE_DESC_SIZE); + hdr->subdesc_type = SQ_DESC_TYPE_HEADER; + /* Enable notification via CQE after processing SQE */ + hdr->post_cqe = 1; + /* No of subdescriptors following this */ + hdr->subdesc_cnt = subdesc_cnt; + hdr->tot_len = len; + + /* Offload checksum calculation to HW */ + if (skb->ip_summed == CHECKSUM_PARTIAL) { + if (skb->protocol != htons(ETH_P_IP)) + return; + + hdr->csum_l3 = 1; /* Enable IP csum calculation */ + hdr->l3_offset = skb_network_offset(skb); + hdr->l4_offset = skb_transport_offset(skb); + + proto = ip_hdr(skb)->protocol; + switch (proto) { + case IPPROTO_TCP: + hdr->csum_l4 = SEND_L4_CSUM_TCP; + break; + case IPPROTO_UDP: + hdr->csum_l4 = SEND_L4_CSUM_UDP; + break; + case IPPROTO_SCTP: + hdr->csum_l4 = SEND_L4_CSUM_SCTP; + break; + } + } +} + +/* SQ GATHER subdescriptor + * Must follow HDR descriptor + */ +static inline void nicvf_sq_add_gather_subdesc(struct snd_queue *sq, int qentry, + int size, u64 data) +{ + struct sq_gather_subdesc *gather; + + qentry &= (sq->dmem.q_len - 1); + gather = (struct sq_gather_subdesc *)GET_SQ_DESC(sq, qentry); + + memset(gather, 0, SND_QUEUE_DESC_SIZE); + gather->subdesc_type = SQ_DESC_TYPE_GATHER; + gather->ld_type = NIC_SEND_LD_TYPE_E_LDWB; + gather->size = size; + gather->addr = data; +} + +/* Segment a TSO packet into 'gso_size' segments and append + * them to SQ for transfer + */ +static int nicvf_sq_append_tso(struct nicvf *nic, struct snd_queue *sq, + int qentry, struct sk_buff *skb) +{ + struct tso_t tso; + int seg_subdescs = 0, desc_cnt = 0; + int seg_len, total_len, data_left; + int hdr_qentry = qentry; + int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + + tso_start(skb, &tso); + total_len = skb->len - hdr_len; + while (total_len > 0) { + char *hdr; + + /* Save Qentry for adding HDR_SUBDESC at the end */ + hdr_qentry = qentry; + + data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len); + total_len -= data_left; + + /* Add segment's header */ + qentry = nicvf_get_nxt_sqentry(sq, qentry); + hdr = sq->tso_hdrs + qentry * TSO_HEADER_SIZE; + tso_build_hdr(skb, hdr, &tso, data_left, total_len == 0); + nicvf_sq_add_gather_subdesc(sq, qentry, hdr_len, + sq->tso_hdrs_phys + + qentry * TSO_HEADER_SIZE); + /* HDR_SUDESC + GATHER */ + seg_subdescs = 2; + seg_len = hdr_len; + + /* Add segment's payload fragments */ + while (data_left > 0) { + int size; + + size = min_t(int, tso.size, data_left); + + qentry = nicvf_get_nxt_sqentry(sq, qentry); + nicvf_sq_add_gather_subdesc(sq, qentry, size, + virt_to_phys(tso.data)); + seg_subdescs++; + seg_len += size; + + data_left -= size; + tso_build_data(skb, &tso, size); + } + nicvf_sq_add_hdr_subdesc(sq, hdr_qentry, + seg_subdescs - 1, skb, seg_len); + sq->skbuff[hdr_qentry] = 0; + qentry = nicvf_get_nxt_sqentry(sq, qentry); + + desc_cnt += seg_subdescs; + } + /* Save SKB in the last segment for freeing */ + sq->skbuff[hdr_qentry] = (u64)skb; + + /* make sure all memory stores are done before ringing doorbell */ + smp_wmb(); + + /* Inform HW to xmit all TSO segments */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_DOOR, + skb_get_queue_mapping(skb), desc_cnt); + return 1; +} + +/* Append an skb to a SQ for packet transfer. */ +int nicvf_sq_append_skb(struct nicvf *nic, struct sk_buff *skb) +{ + int i, size; + int subdesc_cnt; + int sq_num, qentry; + struct queue_set *qs = nic->qs; + struct snd_queue *sq; + + sq_num = skb_get_queue_mapping(skb); + sq = &qs->sq[sq_num]; + + subdesc_cnt = nicvf_sq_subdesc_required(nic, skb); + if (subdesc_cnt > atomic_read(&sq->free_cnt)) + goto append_fail; + + qentry = nicvf_get_sq_desc(sq, subdesc_cnt); + + /* Check if its a TSO packet */ + if (skb_shinfo(skb)->gso_size) + return nicvf_sq_append_tso(nic, sq, qentry, skb); + + /* Add SQ header subdesc */ + nicvf_sq_add_hdr_subdesc(sq, qentry, subdesc_cnt - 1, skb, skb->len); + + /* Add SQ gather subdescs */ + qentry = nicvf_get_nxt_sqentry(sq, qentry); + size = skb_is_nonlinear(skb) ? skb_headlen(skb) : skb->len; + nicvf_sq_add_gather_subdesc(sq, qentry, size, virt_to_phys(skb->data)); + + /* Check for scattered buffer */ + if (!skb_is_nonlinear(skb)) + goto doorbell; + + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + const struct skb_frag_struct *frag; + + frag = &skb_shinfo(skb)->frags[i]; + + qentry = nicvf_get_nxt_sqentry(sq, qentry); + size = skb_frag_size(frag); + nicvf_sq_add_gather_subdesc(sq, qentry, size, + virt_to_phys( + skb_frag_address(frag))); + } + +doorbell: + /* make sure all memory stores are done before ringing doorbell */ + smp_wmb(); + + /* Inform HW to xmit new packet */ + nicvf_queue_reg_write(nic, NIC_QSET_SQ_0_7_DOOR, + sq_num, subdesc_cnt); + return 1; + +append_fail: + netdev_dbg(nic->netdev, "Not enough SQ descriptors to xmit pkt\n"); + return 0; +} + +static inline unsigned frag_num(unsigned i) +{ +#ifdef __BIG_ENDIAN + return (i & ~3) + 3 - (i & 3); +#else + return i; +#endif +} + +/* Returns SKB for a received packet */ +struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic, struct cqe_rx_t *cqe_rx) +{ + int frag; + int payload_len = 0; + struct sk_buff *skb = NULL; + struct sk_buff *skb_frag = NULL; + struct sk_buff *prev_frag = NULL; + u16 *rb_lens = NULL; + u64 *rb_ptrs = NULL; + + rb_lens = (void *)cqe_rx + (3 * sizeof(u64)); + rb_ptrs = (void *)cqe_rx + (6 * sizeof(u64)); + + netdev_dbg(nic->netdev, "%s rb_cnt %d rb0_ptr %llx rb0_sz %d\n", + __func__, cqe_rx->rb_cnt, cqe_rx->rb0_ptr, cqe_rx->rb0_sz); + + for (frag = 0; frag < cqe_rx->rb_cnt; frag++) { + payload_len = rb_lens[frag_num(frag)]; + if (!frag) { + /* First fragment */ + skb = nicvf_rb_ptr_to_skb(nic, + *rb_ptrs - cqe_rx->align_pad, + payload_len); + if (!skb) + return NULL; + skb_reserve(skb, cqe_rx->align_pad); + skb_put(skb, payload_len); + } else { + /* Add fragments */ + skb_frag = nicvf_rb_ptr_to_skb(nic, *rb_ptrs, + payload_len); + if (!skb_frag) { + dev_kfree_skb(skb); + return NULL; + } + + if (!skb_shinfo(skb)->frag_list) + skb_shinfo(skb)->frag_list = skb_frag; + else + prev_frag->next = skb_frag; + + prev_frag = skb_frag; + skb->len += payload_len; + skb->data_len += payload_len; + skb_frag->len = payload_len; + } + /* Next buffer pointer */ + rb_ptrs++; + } + return skb; +} + +/* Enable interrupt */ +void nicvf_enable_intr(struct nicvf *nic, int int_type, int q_idx) +{ + u64 reg_val; + + reg_val = nicvf_reg_read(nic, NIC_VF_ENA_W1S); + + switch (int_type) { + case NICVF_INTR_CQ: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_CQ_SHIFT); + break; + case NICVF_INTR_SQ: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_SQ_SHIFT); + break; + case NICVF_INTR_RBDR: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_RBDR_SHIFT); + break; + case NICVF_INTR_PKT_DROP: + reg_val |= (1ULL << NICVF_INTR_PKT_DROP_SHIFT); + break; + case NICVF_INTR_TCP_TIMER: + reg_val |= (1ULL << NICVF_INTR_TCP_TIMER_SHIFT); + break; + case NICVF_INTR_MBOX: + reg_val |= (1ULL << NICVF_INTR_MBOX_SHIFT); + break; + case NICVF_INTR_QS_ERR: + reg_val |= (1ULL << NICVF_INTR_QS_ERR_SHIFT); + break; + default: + netdev_err(nic->netdev, + "Failed to enable interrupt: unknown type\n"); + break; + } + + nicvf_reg_write(nic, NIC_VF_ENA_W1S, reg_val); +} + +/* Disable interrupt */ +void nicvf_disable_intr(struct nicvf *nic, int int_type, int q_idx) +{ + u64 reg_val = 0; + + switch (int_type) { + case NICVF_INTR_CQ: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_CQ_SHIFT); + break; + case NICVF_INTR_SQ: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_SQ_SHIFT); + break; + case NICVF_INTR_RBDR: + reg_val |= ((1ULL << q_idx) << NICVF_INTR_RBDR_SHIFT); + break; + case NICVF_INTR_PKT_DROP: + reg_val |= (1ULL << NICVF_INTR_PKT_DROP_SHIFT); + break; + case NICVF_INTR_TCP_TIMER: + reg_val |= (1ULL << NICVF_INTR_TCP_TIMER_SHIFT); + break; + case NICVF_INTR_MBOX: + reg_val |= (1ULL << NICVF_INTR_MBOX_SHIFT); + break; + case NICVF_INTR_QS_ERR: + reg_val |= (1ULL << NICVF_INTR_QS_ERR_SHIFT); + break; + default: + netdev_err(nic->netdev, + "Failed to disable interrupt: unknown type\n"); + break; + } + + nicvf_reg_write(nic, NIC_VF_ENA_W1C, reg_val); +} + +/* Clear interrupt */ +void nicvf_clear_intr(struct nicvf *nic, int int_type, int q_idx) +{ + u64 reg_val = 0; + + switch (int_type) { + case NICVF_INTR_CQ: + reg_val = ((1ULL << q_idx) << NICVF_INTR_CQ_SHIFT); + break; + case NICVF_INTR_SQ: + reg_val = ((1ULL << q_idx) << NICVF_INTR_SQ_SHIFT); + break; + case NICVF_INTR_RBDR: + reg_val = ((1ULL << q_idx) << NICVF_INTR_RBDR_SHIFT); + break; + case NICVF_INTR_PKT_DROP: + reg_val = (1ULL << NICVF_INTR_PKT_DROP_SHIFT); + break; + case NICVF_INTR_TCP_TIMER: + reg_val = (1ULL << NICVF_INTR_TCP_TIMER_SHIFT); + break; + case NICVF_INTR_MBOX: + reg_val = (1ULL << NICVF_INTR_MBOX_SHIFT); + break; + case NICVF_INTR_QS_ERR: + reg_val |= (1ULL << NICVF_INTR_QS_ERR_SHIFT); + break; + default: + netdev_err(nic->netdev, + "Failed to clear interrupt: unknown type\n"); + break; + } + + nicvf_reg_write(nic, NIC_VF_INT, reg_val); +} + +/* Check if interrupt is enabled */ +int nicvf_is_intr_enabled(struct nicvf *nic, int int_type, int q_idx) +{ + u64 reg_val; + u64 mask = 0xff; + + reg_val = nicvf_reg_read(nic, NIC_VF_ENA_W1S); + + switch (int_type) { + case NICVF_INTR_CQ: + mask = ((1ULL << q_idx) << NICVF_INTR_CQ_SHIFT); + break; + case NICVF_INTR_SQ: + mask = ((1ULL << q_idx) << NICVF_INTR_SQ_SHIFT); + break; + case NICVF_INTR_RBDR: + mask = ((1ULL << q_idx) << NICVF_INTR_RBDR_SHIFT); + break; + case NICVF_INTR_PKT_DROP: + mask = NICVF_INTR_PKT_DROP_MASK; + break; + case NICVF_INTR_TCP_TIMER: + mask = NICVF_INTR_TCP_TIMER_MASK; + break; + case NICVF_INTR_MBOX: + mask = NICVF_INTR_MBOX_MASK; + break; + case NICVF_INTR_QS_ERR: + mask = NICVF_INTR_QS_ERR_MASK; + break; + default: + netdev_err(nic->netdev, + "Failed to check interrupt enable: unknown type\n"); + break; + } + + return (reg_val & mask); +} + +void nicvf_update_rq_stats(struct nicvf *nic, int rq_idx) +{ + struct rcv_queue *rq; + +#define GET_RQ_STATS(reg) \ + nicvf_reg_read(nic, NIC_QSET_RQ_0_7_STAT_0_1 |\ + (rq_idx << NIC_Q_NUM_SHIFT) | (reg << 3)) + + rq = &nic->qs->rq[rq_idx]; + rq->stats.bytes = GET_RQ_STATS(RQ_SQ_STATS_OCTS); + rq->stats.pkts = GET_RQ_STATS(RQ_SQ_STATS_PKTS); +} + +void nicvf_update_sq_stats(struct nicvf *nic, int sq_idx) +{ + struct snd_queue *sq; + +#define GET_SQ_STATS(reg) \ + nicvf_reg_read(nic, NIC_QSET_SQ_0_7_STAT_0_1 |\ + (sq_idx << NIC_Q_NUM_SHIFT) | (reg << 3)) + + sq = &nic->qs->sq[sq_idx]; + sq->stats.bytes = GET_SQ_STATS(RQ_SQ_STATS_OCTS); + sq->stats.pkts = GET_SQ_STATS(RQ_SQ_STATS_PKTS); +} + +/* Check for errors in the receive cmp.queue entry */ +int nicvf_check_cqe_rx_errs(struct nicvf *nic, + struct cmp_queue *cq, struct cqe_rx_t *cqe_rx) +{ + struct cmp_queue_stats *stats = &cq->stats; + + if (!cqe_rx->err_level && !cqe_rx->err_opcode) { + stats->rx.errop.good++; + return 0; + } + + if (netif_msg_rx_err(nic)) + netdev_err(nic->netdev, + "%s: RX error CQE err_level 0x%x err_opcode 0x%x\n", + nic->netdev->name, + cqe_rx->err_level, cqe_rx->err_opcode); + + switch (cqe_rx->err_level) { + case CQ_ERRLVL_MAC: + stats->rx.errlvl.mac_errs++; + break; + case CQ_ERRLVL_L2: + stats->rx.errlvl.l2_errs++; + break; + case CQ_ERRLVL_L3: + stats->rx.errlvl.l3_errs++; + break; + case CQ_ERRLVL_L4: + stats->rx.errlvl.l4_errs++; + break; + } + + switch (cqe_rx->err_opcode) { + case CQ_RX_ERROP_RE_PARTIAL: + stats->rx.errop.partial_pkts++; + break; + case CQ_RX_ERROP_RE_JABBER: + stats->rx.errop.jabber_errs++; + break; + case CQ_RX_ERROP_RE_FCS: + stats->rx.errop.fcs_errs++; + break; + case CQ_RX_ERROP_RE_TERMINATE: + stats->rx.errop.terminate_errs++; + break; + case CQ_RX_ERROP_RE_RX_CTL: + stats->rx.errop.bgx_rx_errs++; + break; + case CQ_RX_ERROP_PREL2_ERR: + stats->rx.errop.prel2_errs++; + break; + case CQ_RX_ERROP_L2_FRAGMENT: + stats->rx.errop.l2_frags++; + break; + case CQ_RX_ERROP_L2_OVERRUN: + stats->rx.errop.l2_overruns++; + break; + case CQ_RX_ERROP_L2_PFCS: + stats->rx.errop.l2_pfcs++; + break; + case CQ_RX_ERROP_L2_PUNY: + stats->rx.errop.l2_puny++; + break; + case CQ_RX_ERROP_L2_MAL: + stats->rx.errop.l2_hdr_malformed++; + break; + case CQ_RX_ERROP_L2_OVERSIZE: + stats->rx.errop.l2_oversize++; + break; + case CQ_RX_ERROP_L2_UNDERSIZE: + stats->rx.errop.l2_undersize++; + break; + case CQ_RX_ERROP_L2_LENMISM: + stats->rx.errop.l2_len_mismatch++; + break; + case CQ_RX_ERROP_L2_PCLP: + stats->rx.errop.l2_pclp++; + break; + case CQ_RX_ERROP_IP_NOT: + stats->rx.errop.non_ip++; + break; + case CQ_RX_ERROP_IP_CSUM_ERR: + stats->rx.errop.ip_csum_err++; + break; + case CQ_RX_ERROP_IP_MAL: + stats->rx.errop.ip_hdr_malformed++; + break; + case CQ_RX_ERROP_IP_MALD: + stats->rx.errop.ip_payload_malformed++; + break; + case CQ_RX_ERROP_IP_HOP: + stats->rx.errop.ip_hop_errs++; + break; + case CQ_RX_ERROP_L3_ICRC: + stats->rx.errop.l3_icrc_errs++; + break; + case CQ_RX_ERROP_L3_PCLP: + stats->rx.errop.l3_pclp++; + break; + case CQ_RX_ERROP_L4_MAL: + stats->rx.errop.l4_malformed++; + break; + case CQ_RX_ERROP_L4_CHK: + stats->rx.errop.l4_csum_errs++; + break; + case CQ_RX_ERROP_UDP_LEN: + stats->rx.errop.udp_len_err++; + break; + case CQ_RX_ERROP_L4_PORT: + stats->rx.errop.bad_l4_port++; + break; + case CQ_RX_ERROP_TCP_FLAG: + stats->rx.errop.bad_tcp_flag++; + break; + case CQ_RX_ERROP_TCP_OFFSET: + stats->rx.errop.tcp_offset_errs++; + break; + case CQ_RX_ERROP_L4_PCLP: + stats->rx.errop.l4_pclp++; + break; + case CQ_RX_ERROP_RBDR_TRUNC: + stats->rx.errop.pkt_truncated++; + break; + } + + return 1; +} + +/* Check for errors in the send cmp.queue entry */ +int nicvf_check_cqe_tx_errs(struct nicvf *nic, + struct cmp_queue *cq, struct cqe_send_t *cqe_tx) +{ + struct cmp_queue_stats *stats = &cq->stats; + + switch (cqe_tx->send_status) { + case CQ_TX_ERROP_GOOD: + stats->tx.good++; + return 0; + case CQ_TX_ERROP_DESC_FAULT: + stats->tx.desc_fault++; + break; + case CQ_TX_ERROP_HDR_CONS_ERR: + stats->tx.hdr_cons_err++; + break; + case CQ_TX_ERROP_SUBDC_ERR: + stats->tx.subdesc_err++; + break; + case CQ_TX_ERROP_IMM_SIZE_OFLOW: + stats->tx.imm_size_oflow++; + break; + case CQ_TX_ERROP_DATA_SEQUENCE_ERR: + stats->tx.data_seq_err++; + break; + case CQ_TX_ERROP_MEM_SEQUENCE_ERR: + stats->tx.mem_seq_err++; + break; + case CQ_TX_ERROP_LOCK_VIOL: + stats->tx.lock_viol++; + break; + case CQ_TX_ERROP_DATA_FAULT: + stats->tx.data_fault++; + break; + case CQ_TX_ERROP_TSTMP_CONFLICT: + stats->tx.tstmp_conflict++; + break; + case CQ_TX_ERROP_TSTMP_TIMEOUT: + stats->tx.tstmp_timeout++; + break; + case CQ_TX_ERROP_MEM_FAULT: + stats->tx.mem_fault++; + break; + case CQ_TX_ERROP_CK_OVERLAP: + stats->tx.csum_overlap++; + break; + case CQ_TX_ERROP_CK_OFLOW: + stats->tx.csum_overflow++; + break; + } + + return 1; +} diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h new file mode 100644 index 000000000000..8341bdf755d1 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h @@ -0,0 +1,381 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#ifndef NICVF_QUEUES_H +#define NICVF_QUEUES_H + +#include <linux/netdevice.h> +#include "q_struct.h" + +#define MAX_QUEUE_SET 128 +#define MAX_RCV_QUEUES_PER_QS 8 +#define MAX_RCV_BUF_DESC_RINGS_PER_QS 2 +#define MAX_SND_QUEUES_PER_QS 8 +#define MAX_CMP_QUEUES_PER_QS 8 + +/* VF's queue interrupt ranges */ +#define NICVF_INTR_ID_CQ 0 +#define NICVF_INTR_ID_SQ 8 +#define NICVF_INTR_ID_RBDR 16 +#define NICVF_INTR_ID_MISC 18 +#define NICVF_INTR_ID_QS_ERR 19 + +#define for_each_cq_irq(irq) \ + for (irq = NICVF_INTR_ID_CQ; irq < NICVF_INTR_ID_SQ; irq++) +#define for_each_sq_irq(irq) \ + for (irq = NICVF_INTR_ID_SQ; irq < NICVF_INTR_ID_RBDR; irq++) +#define for_each_rbdr_irq(irq) \ + for (irq = NICVF_INTR_ID_RBDR; irq < NICVF_INTR_ID_MISC; irq++) + +#define RBDR_SIZE0 0ULL /* 8K entries */ +#define RBDR_SIZE1 1ULL /* 16K entries */ +#define RBDR_SIZE2 2ULL /* 32K entries */ +#define RBDR_SIZE3 3ULL /* 64K entries */ +#define RBDR_SIZE4 4ULL /* 126K entries */ +#define RBDR_SIZE5 5ULL /* 256K entries */ +#define RBDR_SIZE6 6ULL /* 512K entries */ + +#define SND_QUEUE_SIZE0 0ULL /* 1K entries */ +#define SND_QUEUE_SIZE1 1ULL /* 2K entries */ +#define SND_QUEUE_SIZE2 2ULL /* 4K entries */ +#define SND_QUEUE_SIZE3 3ULL /* 8K entries */ +#define SND_QUEUE_SIZE4 4ULL /* 16K entries */ +#define SND_QUEUE_SIZE5 5ULL /* 32K entries */ +#define SND_QUEUE_SIZE6 6ULL /* 64K entries */ + +#define CMP_QUEUE_SIZE0 0ULL /* 1K entries */ +#define CMP_QUEUE_SIZE1 1ULL /* 2K entries */ +#define CMP_QUEUE_SIZE2 2ULL /* 4K entries */ +#define CMP_QUEUE_SIZE3 3ULL /* 8K entries */ +#define CMP_QUEUE_SIZE4 4ULL /* 16K entries */ +#define CMP_QUEUE_SIZE5 5ULL /* 32K entries */ +#define CMP_QUEUE_SIZE6 6ULL /* 64K entries */ + +/* Default queue count per QS, its lengths and threshold values */ +#define RBDR_CNT 1 +#define RCV_QUEUE_CNT 8 +#define SND_QUEUE_CNT 8 +#define CMP_QUEUE_CNT 8 /* Max of RCV and SND qcount */ + +#define SND_QSIZE SND_QUEUE_SIZE4 +#define SND_QUEUE_LEN (1ULL << (SND_QSIZE + 10)) +#define MAX_SND_QUEUE_LEN (1ULL << (SND_QUEUE_SIZE6 + 10)) +#define SND_QUEUE_THRESH 2ULL +#define MIN_SQ_DESC_PER_PKT_XMIT 2 +/* Since timestamp not enabled, otherwise 2 */ +#define MAX_CQE_PER_PKT_XMIT 1 + +#define CMP_QSIZE CMP_QUEUE_SIZE4 +#define CMP_QUEUE_LEN (1ULL << (CMP_QSIZE + 10)) +#define CMP_QUEUE_CQE_THRESH 0 +#define CMP_QUEUE_TIMER_THRESH 220 /* 10usec */ + +#define RBDR_SIZE RBDR_SIZE0 +#define RCV_BUF_COUNT (1ULL << (RBDR_SIZE + 13)) +#define MAX_RCV_BUF_COUNT (1ULL << (RBDR_SIZE6 + 13)) +#define RBDR_THRESH (RCV_BUF_COUNT / 2) +#define DMA_BUFFER_LEN 2048 /* In multiples of 128bytes */ +#define RCV_FRAG_LEN (SKB_DATA_ALIGN(DMA_BUFFER_LEN + NET_SKB_PAD) + \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) + \ + (NICVF_RCV_BUF_ALIGN_BYTES * 2)) +#define RCV_DATA_OFFSET NICVF_RCV_BUF_ALIGN_BYTES + +#define MAX_CQES_FOR_TX ((SND_QUEUE_LEN / MIN_SQ_DESC_PER_PKT_XMIT) * \ + MAX_CQE_PER_PKT_XMIT) +#define RQ_CQ_DROP ((CMP_QUEUE_LEN - MAX_CQES_FOR_TX) / 256) + +/* Descriptor size in bytes */ +#define SND_QUEUE_DESC_SIZE 16 +#define CMP_QUEUE_DESC_SIZE 512 + +/* Buffer / descriptor alignments */ +#define NICVF_RCV_BUF_ALIGN 7 +#define NICVF_RCV_BUF_ALIGN_BYTES (1ULL << NICVF_RCV_BUF_ALIGN) +#define NICVF_CQ_BASE_ALIGN_BYTES 512 /* 9 bits */ +#define NICVF_SQ_BASE_ALIGN_BYTES 128 /* 7 bits */ + +#define NICVF_ALIGNED_ADDR(ADDR, ALIGN_BYTES) ALIGN(ADDR, ALIGN_BYTES) +#define NICVF_ADDR_ALIGN_LEN(ADDR, BYTES)\ + (NICVF_ALIGNED_ADDR(ADDR, BYTES) - BYTES) +#define NICVF_RCV_BUF_ALIGN_LEN(X)\ + (NICVF_ALIGNED_ADDR(X, NICVF_RCV_BUF_ALIGN_BYTES) - X) + +/* Queue enable/disable */ +#define NICVF_SQ_EN BIT_ULL(19) + +/* Queue reset */ +#define NICVF_CQ_RESET BIT_ULL(41) +#define NICVF_SQ_RESET BIT_ULL(17) +#define NICVF_RBDR_RESET BIT_ULL(43) + +enum CQ_RX_ERRLVL_E { + CQ_ERRLVL_MAC, + CQ_ERRLVL_L2, + CQ_ERRLVL_L3, + CQ_ERRLVL_L4, +}; + +enum CQ_RX_ERROP_E { + CQ_RX_ERROP_RE_NONE = 0x0, + CQ_RX_ERROP_RE_PARTIAL = 0x1, + CQ_RX_ERROP_RE_JABBER = 0x2, + CQ_RX_ERROP_RE_FCS = 0x7, + CQ_RX_ERROP_RE_TERMINATE = 0x9, + CQ_RX_ERROP_RE_RX_CTL = 0xb, + CQ_RX_ERROP_PREL2_ERR = 0x1f, + CQ_RX_ERROP_L2_FRAGMENT = 0x20, + CQ_RX_ERROP_L2_OVERRUN = 0x21, + CQ_RX_ERROP_L2_PFCS = 0x22, + CQ_RX_ERROP_L2_PUNY = 0x23, + CQ_RX_ERROP_L2_MAL = 0x24, + CQ_RX_ERROP_L2_OVERSIZE = 0x25, + CQ_RX_ERROP_L2_UNDERSIZE = 0x26, + CQ_RX_ERROP_L2_LENMISM = 0x27, + CQ_RX_ERROP_L2_PCLP = 0x28, + CQ_RX_ERROP_IP_NOT = 0x41, + CQ_RX_ERROP_IP_CSUM_ERR = 0x42, + CQ_RX_ERROP_IP_MAL = 0x43, + CQ_RX_ERROP_IP_MALD = 0x44, + CQ_RX_ERROP_IP_HOP = 0x45, + CQ_RX_ERROP_L3_ICRC = 0x46, + CQ_RX_ERROP_L3_PCLP = 0x47, + CQ_RX_ERROP_L4_MAL = 0x61, + CQ_RX_ERROP_L4_CHK = 0x62, + CQ_RX_ERROP_UDP_LEN = 0x63, + CQ_RX_ERROP_L4_PORT = 0x64, + CQ_RX_ERROP_TCP_FLAG = 0x65, + CQ_RX_ERROP_TCP_OFFSET = 0x66, + CQ_RX_ERROP_L4_PCLP = 0x67, + CQ_RX_ERROP_RBDR_TRUNC = 0x70, +}; + +enum CQ_TX_ERROP_E { + CQ_TX_ERROP_GOOD = 0x0, + CQ_TX_ERROP_DESC_FAULT = 0x10, + CQ_TX_ERROP_HDR_CONS_ERR = 0x11, + CQ_TX_ERROP_SUBDC_ERR = 0x12, + CQ_TX_ERROP_IMM_SIZE_OFLOW = 0x80, + CQ_TX_ERROP_DATA_SEQUENCE_ERR = 0x81, + CQ_TX_ERROP_MEM_SEQUENCE_ERR = 0x82, + CQ_TX_ERROP_LOCK_VIOL = 0x83, + CQ_TX_ERROP_DATA_FAULT = 0x84, + CQ_TX_ERROP_TSTMP_CONFLICT = 0x85, + CQ_TX_ERROP_TSTMP_TIMEOUT = 0x86, + CQ_TX_ERROP_MEM_FAULT = 0x87, + CQ_TX_ERROP_CK_OVERLAP = 0x88, + CQ_TX_ERROP_CK_OFLOW = 0x89, + CQ_TX_ERROP_ENUM_LAST = 0x8a, +}; + +struct cmp_queue_stats { + struct rx_stats { + struct { + u64 mac_errs; + u64 l2_errs; + u64 l3_errs; + u64 l4_errs; + } errlvl; + struct { + u64 good; + u64 partial_pkts; + u64 jabber_errs; + u64 fcs_errs; + u64 terminate_errs; + u64 bgx_rx_errs; + u64 prel2_errs; + u64 l2_frags; + u64 l2_overruns; + u64 l2_pfcs; + u64 l2_puny; + u64 l2_hdr_malformed; + u64 l2_oversize; + u64 l2_undersize; + u64 l2_len_mismatch; + u64 l2_pclp; + u64 non_ip; + u64 ip_csum_err; + u64 ip_hdr_malformed; + u64 ip_payload_malformed; + u64 ip_hop_errs; + u64 l3_icrc_errs; + u64 l3_pclp; + u64 l4_malformed; + u64 l4_csum_errs; + u64 udp_len_err; + u64 bad_l4_port; + u64 bad_tcp_flag; + u64 tcp_offset_errs; + u64 l4_pclp; + u64 pkt_truncated; + } errop; + } rx; + struct tx_stats { + u64 good; + u64 desc_fault; + u64 hdr_cons_err; + u64 subdesc_err; + u64 imm_size_oflow; + u64 data_seq_err; + u64 mem_seq_err; + u64 lock_viol; + u64 data_fault; + u64 tstmp_conflict; + u64 tstmp_timeout; + u64 mem_fault; + u64 csum_overlap; + u64 csum_overflow; + } tx; +} ____cacheline_aligned_in_smp; + +enum RQ_SQ_STATS { + RQ_SQ_STATS_OCTS, + RQ_SQ_STATS_PKTS, +}; + +struct rx_tx_queue_stats { + u64 bytes; + u64 pkts; +} ____cacheline_aligned_in_smp; + +struct q_desc_mem { + dma_addr_t dma; + u64 size; + u16 q_len; + dma_addr_t phys_base; + void *base; + void *unalign_base; +}; + +struct rbdr { + bool enable; + u32 dma_size; + u32 frag_len; + u32 thresh; /* Threshold level for interrupt */ + void *desc; + u32 head; + u32 tail; + struct q_desc_mem dmem; +} ____cacheline_aligned_in_smp; + +struct rcv_queue { + bool enable; + struct rbdr *rbdr_start; + struct rbdr *rbdr_cont; + bool en_tcp_reassembly; + u8 cq_qs; /* CQ's QS to which this RQ is assigned */ + u8 cq_idx; /* CQ index (0 to 7) in the QS */ + u8 cont_rbdr_qs; /* Continue buffer ptrs - QS num */ + u8 cont_qs_rbdr_idx; /* RBDR idx in the cont QS */ + u8 start_rbdr_qs; /* First buffer ptrs - QS num */ + u8 start_qs_rbdr_idx; /* RBDR idx in the above QS */ + u8 caching; + struct rx_tx_queue_stats stats; +} ____cacheline_aligned_in_smp; + +struct cmp_queue { + bool enable; + u16 thresh; + spinlock_t lock; /* lock to serialize processing CQEs */ + void *desc; + struct q_desc_mem dmem; + struct cmp_queue_stats stats; +} ____cacheline_aligned_in_smp; + +struct snd_queue { + bool enable; + u8 cq_qs; /* CQ's QS to which this SQ is pointing */ + u8 cq_idx; /* CQ index (0 to 7) in the above QS */ + u16 thresh; + atomic_t free_cnt; + u32 head; + u32 tail; + u64 *skbuff; + void *desc; + +#define TSO_HEADER_SIZE 128 + /* For TSO segment's header */ + char *tso_hdrs; + dma_addr_t tso_hdrs_phys; + + cpumask_t affinity_mask; + struct q_desc_mem dmem; + struct rx_tx_queue_stats stats; +} ____cacheline_aligned_in_smp; + +struct queue_set { + bool enable; + bool be_en; + u8 vnic_id; + u8 rq_cnt; + u8 cq_cnt; + u64 cq_len; + u8 sq_cnt; + u64 sq_len; + u8 rbdr_cnt; + u64 rbdr_len; + struct rcv_queue rq[MAX_RCV_QUEUES_PER_QS]; + struct cmp_queue cq[MAX_CMP_QUEUES_PER_QS]; + struct snd_queue sq[MAX_SND_QUEUES_PER_QS]; + struct rbdr rbdr[MAX_RCV_BUF_DESC_RINGS_PER_QS]; +} ____cacheline_aligned_in_smp; + +#define GET_RBDR_DESC(RING, idx)\ + (&(((struct rbdr_entry_t *)((RING)->desc))[idx])) +#define GET_SQ_DESC(RING, idx)\ + (&(((struct sq_hdr_subdesc *)((RING)->desc))[idx])) +#define GET_CQ_DESC(RING, idx)\ + (&(((union cq_desc_t *)((RING)->desc))[idx])) + +/* CQ status bits */ +#define CQ_WR_FULL BIT(26) +#define CQ_WR_DISABLE BIT(25) +#define CQ_WR_FAULT BIT(24) +#define CQ_CQE_COUNT (0xFFFF << 0) + +#define CQ_ERR_MASK (CQ_WR_FULL | CQ_WR_DISABLE | CQ_WR_FAULT) + +int nicvf_set_qset_resources(struct nicvf *nic); +int nicvf_config_data_transfer(struct nicvf *nic, bool enable); +void nicvf_qset_config(struct nicvf *nic, bool enable); +void nicvf_cmp_queue_config(struct nicvf *nic, struct queue_set *qs, + int qidx, bool enable); + +void nicvf_sq_enable(struct nicvf *nic, struct snd_queue *sq, int qidx); +void nicvf_sq_disable(struct nicvf *nic, int qidx); +void nicvf_put_sq_desc(struct snd_queue *sq, int desc_cnt); +void nicvf_sq_free_used_descs(struct net_device *netdev, + struct snd_queue *sq, int qidx); +int nicvf_sq_append_skb(struct nicvf *nic, struct sk_buff *skb); + +struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic, struct cqe_rx_t *cqe_rx); +void nicvf_rbdr_task(unsigned long data); +void nicvf_rbdr_work(struct work_struct *work); + +void nicvf_enable_intr(struct nicvf *nic, int int_type, int q_idx); +void nicvf_disable_intr(struct nicvf *nic, int int_type, int q_idx); +void nicvf_clear_intr(struct nicvf *nic, int int_type, int q_idx); +int nicvf_is_intr_enabled(struct nicvf *nic, int int_type, int q_idx); + +/* Register access APIs */ +void nicvf_reg_write(struct nicvf *nic, u64 offset, u64 val); +u64 nicvf_reg_read(struct nicvf *nic, u64 offset); +void nicvf_qset_reg_write(struct nicvf *nic, u64 offset, u64 val); +u64 nicvf_qset_reg_read(struct nicvf *nic, u64 offset); +void nicvf_queue_reg_write(struct nicvf *nic, u64 offset, + u64 qidx, u64 val); +u64 nicvf_queue_reg_read(struct nicvf *nic, + u64 offset, u64 qidx); + +/* Stats */ +void nicvf_update_rq_stats(struct nicvf *nic, int rq_idx); +void nicvf_update_sq_stats(struct nicvf *nic, int sq_idx); +int nicvf_check_cqe_rx_errs(struct nicvf *nic, + struct cmp_queue *cq, struct cqe_rx_t *cqe_rx); +int nicvf_check_cqe_tx_errs(struct nicvf *nic, + struct cmp_queue *cq, struct cqe_send_t *cqe_tx); +#endif /* NICVF_QUEUES_H */ diff --git a/drivers/net/ethernet/cavium/thunder/q_struct.h b/drivers/net/ethernet/cavium/thunder/q_struct.h new file mode 100644 index 000000000000..3c1de97b1add --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/q_struct.h @@ -0,0 +1,701 @@ +/* + * This file contains HW queue descriptor formats, config register + * structures etc + * + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#ifndef Q_STRUCT_H +#define Q_STRUCT_H + +/* Load transaction types for reading segment bytes specified by + * NIC_SEND_GATHER_S[LD_TYPE]. + */ +enum nic_send_ld_type_e { + NIC_SEND_LD_TYPE_E_LDD = 0x0, + NIC_SEND_LD_TYPE_E_LDT = 0x1, + NIC_SEND_LD_TYPE_E_LDWB = 0x2, + NIC_SEND_LD_TYPE_E_ENUM_LAST = 0x3, +}; + +enum ether_type_algorithm { + ETYPE_ALG_NONE = 0x0, + ETYPE_ALG_SKIP = 0x1, + ETYPE_ALG_ENDPARSE = 0x2, + ETYPE_ALG_VLAN = 0x3, + ETYPE_ALG_VLAN_STRIP = 0x4, +}; + +enum layer3_type { + L3TYPE_NONE = 0x00, + L3TYPE_GRH = 0x01, + L3TYPE_IPV4 = 0x04, + L3TYPE_IPV4_OPTIONS = 0x05, + L3TYPE_IPV6 = 0x06, + L3TYPE_IPV6_OPTIONS = 0x07, + L3TYPE_ET_STOP = 0x0D, + L3TYPE_OTHER = 0x0E, +}; + +enum layer4_type { + L4TYPE_NONE = 0x00, + L4TYPE_IPSEC_ESP = 0x01, + L4TYPE_IPFRAG = 0x02, + L4TYPE_IPCOMP = 0x03, + L4TYPE_TCP = 0x04, + L4TYPE_UDP = 0x05, + L4TYPE_SCTP = 0x06, + L4TYPE_GRE = 0x07, + L4TYPE_ROCE_BTH = 0x08, + L4TYPE_OTHER = 0x0E, +}; + +/* CPI and RSSI configuration */ +enum cpi_algorithm_type { + CPI_ALG_NONE = 0x0, + CPI_ALG_VLAN = 0x1, + CPI_ALG_VLAN16 = 0x2, + CPI_ALG_DIFF = 0x3, +}; + +enum rss_algorithm_type { + RSS_ALG_NONE = 0x00, + RSS_ALG_PORT = 0x01, + RSS_ALG_IP = 0x02, + RSS_ALG_TCP_IP = 0x03, + RSS_ALG_UDP_IP = 0x04, + RSS_ALG_SCTP_IP = 0x05, + RSS_ALG_GRE_IP = 0x06, + RSS_ALG_ROCE = 0x07, +}; + +enum rss_hash_cfg { + RSS_HASH_L2ETC = 0x00, + RSS_HASH_IP = 0x01, + RSS_HASH_TCP = 0x02, + RSS_HASH_TCP_SYN_DIS = 0x03, + RSS_HASH_UDP = 0x04, + RSS_HASH_L4ETC = 0x05, + RSS_HASH_ROCE = 0x06, + RSS_L3_BIDI = 0x07, + RSS_L4_BIDI = 0x08, +}; + +/* Completion queue entry types */ +enum cqe_type { + CQE_TYPE_INVALID = 0x0, + CQE_TYPE_RX = 0x2, + CQE_TYPE_RX_SPLIT = 0x3, + CQE_TYPE_RX_TCP = 0x4, + CQE_TYPE_SEND = 0x8, + CQE_TYPE_SEND_PTP = 0x9, +}; + +enum cqe_rx_tcp_status { + CQE_RX_STATUS_VALID_TCP_CNXT = 0x00, + CQE_RX_STATUS_INVALID_TCP_CNXT = 0x0F, +}; + +enum cqe_send_status { + CQE_SEND_STATUS_GOOD = 0x00, + CQE_SEND_STATUS_DESC_FAULT = 0x01, + CQE_SEND_STATUS_HDR_CONS_ERR = 0x11, + CQE_SEND_STATUS_SUBDESC_ERR = 0x12, + CQE_SEND_STATUS_IMM_SIZE_OFLOW = 0x80, + CQE_SEND_STATUS_CRC_SEQ_ERR = 0x81, + CQE_SEND_STATUS_DATA_SEQ_ERR = 0x82, + CQE_SEND_STATUS_MEM_SEQ_ERR = 0x83, + CQE_SEND_STATUS_LOCK_VIOL = 0x84, + CQE_SEND_STATUS_LOCK_UFLOW = 0x85, + CQE_SEND_STATUS_DATA_FAULT = 0x86, + CQE_SEND_STATUS_TSTMP_CONFLICT = 0x87, + CQE_SEND_STATUS_TSTMP_TIMEOUT = 0x88, + CQE_SEND_STATUS_MEM_FAULT = 0x89, + CQE_SEND_STATUS_CSUM_OVERLAP = 0x8A, + CQE_SEND_STATUS_CSUM_OVERFLOW = 0x8B, +}; + +enum cqe_rx_tcp_end_reason { + CQE_RX_TCP_END_FIN_FLAG_DET = 0, + CQE_RX_TCP_END_INVALID_FLAG = 1, + CQE_RX_TCP_END_TIMEOUT = 2, + CQE_RX_TCP_END_OUT_OF_SEQ = 3, + CQE_RX_TCP_END_PKT_ERR = 4, + CQE_RX_TCP_END_QS_DISABLED = 0x0F, +}; + +/* Packet protocol level error enumeration */ +enum cqe_rx_err_level { + CQE_RX_ERRLVL_RE = 0x0, + CQE_RX_ERRLVL_L2 = 0x1, + CQE_RX_ERRLVL_L3 = 0x2, + CQE_RX_ERRLVL_L4 = 0x3, +}; + +/* Packet protocol level error type enumeration */ +enum cqe_rx_err_opcode { + CQE_RX_ERR_RE_NONE = 0x0, + CQE_RX_ERR_RE_PARTIAL = 0x1, + CQE_RX_ERR_RE_JABBER = 0x2, + CQE_RX_ERR_RE_FCS = 0x7, + CQE_RX_ERR_RE_TERMINATE = 0x9, + CQE_RX_ERR_RE_RX_CTL = 0xb, + CQE_RX_ERR_PREL2_ERR = 0x1f, + CQE_RX_ERR_L2_FRAGMENT = 0x20, + CQE_RX_ERR_L2_OVERRUN = 0x21, + CQE_RX_ERR_L2_PFCS = 0x22, + CQE_RX_ERR_L2_PUNY = 0x23, + CQE_RX_ERR_L2_MAL = 0x24, + CQE_RX_ERR_L2_OVERSIZE = 0x25, + CQE_RX_ERR_L2_UNDERSIZE = 0x26, + CQE_RX_ERR_L2_LENMISM = 0x27, + CQE_RX_ERR_L2_PCLP = 0x28, + CQE_RX_ERR_IP_NOT = 0x41, + CQE_RX_ERR_IP_CHK = 0x42, + CQE_RX_ERR_IP_MAL = 0x43, + CQE_RX_ERR_IP_MALD = 0x44, + CQE_RX_ERR_IP_HOP = 0x45, + CQE_RX_ERR_L3_ICRC = 0x46, + CQE_RX_ERR_L3_PCLP = 0x47, + CQE_RX_ERR_L4_MAL = 0x61, + CQE_RX_ERR_L4_CHK = 0x62, + CQE_RX_ERR_UDP_LEN = 0x63, + CQE_RX_ERR_L4_PORT = 0x64, + CQE_RX_ERR_TCP_FLAG = 0x65, + CQE_RX_ERR_TCP_OFFSET = 0x66, + CQE_RX_ERR_L4_PCLP = 0x67, + CQE_RX_ERR_RBDR_TRUNC = 0x70, +}; + +struct cqe_rx_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 cqe_type:4; /* W0 */ + u64 stdn_fault:1; + u64 rsvd0:1; + u64 rq_qs:7; + u64 rq_idx:3; + u64 rsvd1:12; + u64 rss_alg:4; + u64 rsvd2:4; + u64 rb_cnt:4; + u64 vlan_found:1; + u64 vlan_stripped:1; + u64 vlan2_found:1; + u64 vlan2_stripped:1; + u64 l4_type:4; + u64 l3_type:4; + u64 l2_present:1; + u64 err_level:3; + u64 err_opcode:8; + + u64 pkt_len:16; /* W1 */ + u64 l2_ptr:8; + u64 l3_ptr:8; + u64 l4_ptr:8; + u64 cq_pkt_len:8; + u64 align_pad:3; + u64 rsvd3:1; + u64 chan:12; + + u64 rss_tag:32; /* W2 */ + u64 vlan_tci:16; + u64 vlan_ptr:8; + u64 vlan2_ptr:8; + + u64 rb3_sz:16; /* W3 */ + u64 rb2_sz:16; + u64 rb1_sz:16; + u64 rb0_sz:16; + + u64 rb7_sz:16; /* W4 */ + u64 rb6_sz:16; + u64 rb5_sz:16; + u64 rb4_sz:16; + + u64 rb11_sz:16; /* W5 */ + u64 rb10_sz:16; + u64 rb9_sz:16; + u64 rb8_sz:16; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 err_opcode:8; + u64 err_level:3; + u64 l2_present:1; + u64 l3_type:4; + u64 l4_type:4; + u64 vlan2_stripped:1; + u64 vlan2_found:1; + u64 vlan_stripped:1; + u64 vlan_found:1; + u64 rb_cnt:4; + u64 rsvd2:4; + u64 rss_alg:4; + u64 rsvd1:12; + u64 rq_idx:3; + u64 rq_qs:7; + u64 rsvd0:1; + u64 stdn_fault:1; + u64 cqe_type:4; /* W0 */ + u64 chan:12; + u64 rsvd3:1; + u64 align_pad:3; + u64 cq_pkt_len:8; + u64 l4_ptr:8; + u64 l3_ptr:8; + u64 l2_ptr:8; + u64 pkt_len:16; /* W1 */ + u64 vlan2_ptr:8; + u64 vlan_ptr:8; + u64 vlan_tci:16; + u64 rss_tag:32; /* W2 */ + u64 rb0_sz:16; + u64 rb1_sz:16; + u64 rb2_sz:16; + u64 rb3_sz:16; /* W3 */ + u64 rb4_sz:16; + u64 rb5_sz:16; + u64 rb6_sz:16; + u64 rb7_sz:16; /* W4 */ + u64 rb8_sz:16; + u64 rb9_sz:16; + u64 rb10_sz:16; + u64 rb11_sz:16; /* W5 */ +#endif + u64 rb0_ptr:64; + u64 rb1_ptr:64; + u64 rb2_ptr:64; + u64 rb3_ptr:64; + u64 rb4_ptr:64; + u64 rb5_ptr:64; + u64 rb6_ptr:64; + u64 rb7_ptr:64; + u64 rb8_ptr:64; + u64 rb9_ptr:64; + u64 rb10_ptr:64; + u64 rb11_ptr:64; +}; + +struct cqe_rx_tcp_err_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 cqe_type:4; /* W0 */ + u64 rsvd0:60; + + u64 rsvd1:4; /* W1 */ + u64 partial_first:1; + u64 rsvd2:27; + u64 rbdr_bytes:8; + u64 rsvd3:24; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 rsvd0:60; + u64 cqe_type:4; + + u64 rsvd3:24; + u64 rbdr_bytes:8; + u64 rsvd2:27; + u64 partial_first:1; + u64 rsvd1:4; +#endif +}; + +struct cqe_rx_tcp_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 cqe_type:4; /* W0 */ + u64 rsvd0:52; + u64 cq_tcp_status:8; + + u64 rsvd1:32; /* W1 */ + u64 tcp_cntx_bytes:8; + u64 rsvd2:8; + u64 tcp_err_bytes:16; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 cq_tcp_status:8; + u64 rsvd0:52; + u64 cqe_type:4; /* W0 */ + + u64 tcp_err_bytes:16; + u64 rsvd2:8; + u64 tcp_cntx_bytes:8; + u64 rsvd1:32; /* W1 */ +#endif +}; + +struct cqe_send_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 cqe_type:4; /* W0 */ + u64 rsvd0:4; + u64 sqe_ptr:16; + u64 rsvd1:4; + u64 rsvd2:10; + u64 sq_qs:7; + u64 sq_idx:3; + u64 rsvd3:8; + u64 send_status:8; + + u64 ptp_timestamp:64; /* W1 */ +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 send_status:8; + u64 rsvd3:8; + u64 sq_idx:3; + u64 sq_qs:7; + u64 rsvd2:10; + u64 rsvd1:4; + u64 sqe_ptr:16; + u64 rsvd0:4; + u64 cqe_type:4; /* W0 */ + + u64 ptp_timestamp:64; /* W1 */ +#endif +}; + +union cq_desc_t { + u64 u[64]; + struct cqe_send_t snd_hdr; + struct cqe_rx_t rx_hdr; + struct cqe_rx_tcp_t rx_tcp_hdr; + struct cqe_rx_tcp_err_t rx_tcp_err_hdr; +}; + +struct rbdr_entry_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 rsvd0:15; + u64 buf_addr:42; + u64 cache_align:7; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 cache_align:7; + u64 buf_addr:42; + u64 rsvd0:15; +#endif +}; + +/* TCP reassembly context */ +struct rbe_tcp_cnxt_t { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 tcp_pkt_cnt:12; + u64 rsvd1:4; + u64 align_hdr_bytes:4; + u64 align_ptr_bytes:4; + u64 ptr_bytes:16; + u64 rsvd2:24; + u64 cqe_type:4; + u64 rsvd0:54; + u64 tcp_end_reason:2; + u64 tcp_status:4; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 tcp_status:4; + u64 tcp_end_reason:2; + u64 rsvd0:54; + u64 cqe_type:4; + u64 rsvd2:24; + u64 ptr_bytes:16; + u64 align_ptr_bytes:4; + u64 align_hdr_bytes:4; + u64 rsvd1:4; + u64 tcp_pkt_cnt:12; +#endif +}; + +/* Always Big endian */ +struct rx_hdr_t { + u64 opaque:32; + u64 rss_flow:8; + u64 skip_length:6; + u64 disable_rss:1; + u64 disable_tcp_reassembly:1; + u64 nodrop:1; + u64 dest_alg:2; + u64 rsvd0:2; + u64 dest_rq:11; +}; + +enum send_l4_csum_type { + SEND_L4_CSUM_DISABLE = 0x00, + SEND_L4_CSUM_UDP = 0x01, + SEND_L4_CSUM_TCP = 0x02, + SEND_L4_CSUM_SCTP = 0x03, +}; + +enum send_crc_alg { + SEND_CRCALG_CRC32 = 0x00, + SEND_CRCALG_CRC32C = 0x01, + SEND_CRCALG_ICRC = 0x02, +}; + +enum send_load_type { + SEND_LD_TYPE_LDD = 0x00, + SEND_LD_TYPE_LDT = 0x01, + SEND_LD_TYPE_LDWB = 0x02, +}; + +enum send_mem_alg_type { + SEND_MEMALG_SET = 0x00, + SEND_MEMALG_ADD = 0x08, + SEND_MEMALG_SUB = 0x09, + SEND_MEMALG_ADDLEN = 0x0A, + SEND_MEMALG_SUBLEN = 0x0B, +}; + +enum send_mem_dsz_type { + SEND_MEMDSZ_B64 = 0x00, + SEND_MEMDSZ_B32 = 0x01, + SEND_MEMDSZ_B8 = 0x03, +}; + +enum sq_subdesc_type { + SQ_DESC_TYPE_INVALID = 0x00, + SQ_DESC_TYPE_HEADER = 0x01, + SQ_DESC_TYPE_CRC = 0x02, + SQ_DESC_TYPE_IMMEDIATE = 0x03, + SQ_DESC_TYPE_GATHER = 0x04, + SQ_DESC_TYPE_MEMORY = 0x05, +}; + +struct sq_crc_subdesc { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 rsvd1:32; + u64 crc_ival:32; + u64 subdesc_type:4; + u64 crc_alg:2; + u64 rsvd0:10; + u64 crc_insert_pos:16; + u64 hdr_start:16; + u64 crc_len:16; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 crc_len:16; + u64 hdr_start:16; + u64 crc_insert_pos:16; + u64 rsvd0:10; + u64 crc_alg:2; + u64 subdesc_type:4; + u64 crc_ival:32; + u64 rsvd1:32; +#endif +}; + +struct sq_gather_subdesc { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 subdesc_type:4; /* W0 */ + u64 ld_type:2; + u64 rsvd0:42; + u64 size:16; + + u64 rsvd1:15; /* W1 */ + u64 addr:49; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 size:16; + u64 rsvd0:42; + u64 ld_type:2; + u64 subdesc_type:4; /* W0 */ + + u64 addr:49; + u64 rsvd1:15; /* W1 */ +#endif +}; + +/* SQ immediate subdescriptor */ +struct sq_imm_subdesc { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 subdesc_type:4; /* W0 */ + u64 rsvd0:46; + u64 len:14; + + u64 data:64; /* W1 */ +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 len:14; + u64 rsvd0:46; + u64 subdesc_type:4; /* W0 */ + + u64 data:64; /* W1 */ +#endif +}; + +struct sq_mem_subdesc { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 subdesc_type:4; /* W0 */ + u64 mem_alg:4; + u64 mem_dsz:2; + u64 wmem:1; + u64 rsvd0:21; + u64 offset:32; + + u64 rsvd1:15; /* W1 */ + u64 addr:49; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 offset:32; + u64 rsvd0:21; + u64 wmem:1; + u64 mem_dsz:2; + u64 mem_alg:4; + u64 subdesc_type:4; /* W0 */ + + u64 addr:49; + u64 rsvd1:15; /* W1 */ +#endif +}; + +struct sq_hdr_subdesc { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 subdesc_type:4; + u64 tso:1; + u64 post_cqe:1; /* Post CQE on no error also */ + u64 dont_send:1; + u64 tstmp:1; + u64 subdesc_cnt:8; + u64 csum_l4:2; + u64 csum_l3:1; + u64 rsvd0:5; + u64 l4_offset:8; + u64 l3_offset:8; + u64 rsvd1:4; + u64 tot_len:20; /* W0 */ + + u64 tso_sdc_cont:8; + u64 tso_sdc_first:8; + u64 tso_l4_offset:8; + u64 tso_flags_last:12; + u64 tso_flags_first:12; + u64 rsvd2:2; + u64 tso_max_paysize:14; /* W1 */ +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 tot_len:20; + u64 rsvd1:4; + u64 l3_offset:8; + u64 l4_offset:8; + u64 rsvd0:5; + u64 csum_l3:1; + u64 csum_l4:2; + u64 subdesc_cnt:8; + u64 tstmp:1; + u64 dont_send:1; + u64 post_cqe:1; /* Post CQE on no error also */ + u64 tso:1; + u64 subdesc_type:4; /* W0 */ + + u64 tso_max_paysize:14; + u64 rsvd2:2; + u64 tso_flags_first:12; + u64 tso_flags_last:12; + u64 tso_l4_offset:8; + u64 tso_sdc_first:8; + u64 tso_sdc_cont:8; /* W1 */ +#endif +}; + +/* Queue config register formats */ +struct rq_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_2_63:62; + u64 ena:1; + u64 tcp_ena:1; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 tcp_ena:1; + u64 ena:1; + u64 reserved_2_63:62; +#endif +}; + +struct cq_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_43_63:21; + u64 ena:1; + u64 reset:1; + u64 caching:1; + u64 reserved_35_39:5; + u64 qsize:3; + u64 reserved_25_31:7; + u64 avg_con:9; + u64 reserved_0_15:16; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 reserved_0_15:16; + u64 avg_con:9; + u64 reserved_25_31:7; + u64 qsize:3; + u64 reserved_35_39:5; + u64 caching:1; + u64 reset:1; + u64 ena:1; + u64 reserved_43_63:21; +#endif +}; + +struct sq_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_20_63:44; + u64 ena:1; + u64 reserved_18_18:1; + u64 reset:1; + u64 ldwb:1; + u64 reserved_11_15:5; + u64 qsize:3; + u64 reserved_3_7:5; + u64 tstmp_bgx_intf:3; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 tstmp_bgx_intf:3; + u64 reserved_3_7:5; + u64 qsize:3; + u64 reserved_11_15:5; + u64 ldwb:1; + u64 reset:1; + u64 reserved_18_18:1; + u64 ena:1; + u64 reserved_20_63:44; +#endif +}; + +struct rbdr_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_45_63:19; + u64 ena:1; + u64 reset:1; + u64 ldwb:1; + u64 reserved_36_41:6; + u64 qsize:4; + u64 reserved_25_31:7; + u64 avg_con:9; + u64 reserved_12_15:4; + u64 lines:12; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 lines:12; + u64 reserved_12_15:4; + u64 avg_con:9; + u64 reserved_25_31:7; + u64 qsize:4; + u64 reserved_36_41:6; + u64 ldwb:1; + u64 reset:1; + u64 ena: 1; + u64 reserved_45_63:19; +#endif +}; + +struct qs_cfg { +#if defined(__BIG_ENDIAN_BITFIELD) + u64 reserved_32_63:32; + u64 ena:1; + u64 reserved_27_30:4; + u64 sq_ins_ena:1; + u64 sq_ins_pos:6; + u64 lock_ena:1; + u64 lock_viol_cqe_ena:1; + u64 send_tstmp_ena:1; + u64 be:1; + u64 reserved_7_15:9; + u64 vnic:7; +#elif defined(__LITTLE_ENDIAN_BITFIELD) + u64 vnic:7; + u64 reserved_7_15:9; + u64 be:1; + u64 send_tstmp_ena:1; + u64 lock_viol_cqe_ena:1; + u64 lock_ena:1; + u64 sq_ins_pos:6; + u64 sq_ins_ena:1; + u64 reserved_27_30:4; + u64 ena:1; + u64 reserved_32_63:32; +#endif +}; + +#endif /* Q_STRUCT_H */ diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c new file mode 100644 index 000000000000..020e11cf3fdd --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c @@ -0,0 +1,966 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/phy.h> +#include <linux/of.h> +#include <linux/of_mdio.h> +#include <linux/of_net.h> + +#include "nic_reg.h" +#include "nic.h" +#include "thunder_bgx.h" + +#define DRV_NAME "thunder-BGX" +#define DRV_VERSION "1.0" + +struct lmac { + struct bgx *bgx; + int dmac; + unsigned char mac[ETH_ALEN]; + bool link_up; + int lmacid; /* ID within BGX */ + int lmacid_bd; /* ID on board */ + struct net_device netdev; + struct phy_device *phydev; + unsigned int last_duplex; + unsigned int last_link; + unsigned int last_speed; + bool is_sgmii; + struct delayed_work dwork; + struct workqueue_struct *check_link; +} lmac; + +struct bgx { + u8 bgx_id; + u8 qlm_mode; + struct lmac lmac[MAX_LMAC_PER_BGX]; + int lmac_count; + int lmac_type; + int lane_to_sds; + int use_training; + void __iomem *reg_base; + struct pci_dev *pdev; +} bgx; + +struct bgx *bgx_vnic[MAX_BGX_THUNDER]; +static int lmac_count; /* Total no of LMACs in system */ + +static int bgx_xaui_check_link(struct lmac *lmac); + +/* Supported devices */ +static const struct pci_device_id bgx_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_BGX) }, + { 0, } /* end of table */ +}; + +MODULE_AUTHOR("Cavium Inc"); +MODULE_DESCRIPTION("Cavium Thunder BGX/MAC Driver"); +MODULE_LICENSE("GPL v2"); +MODULE_VERSION(DRV_VERSION); +MODULE_DEVICE_TABLE(pci, bgx_id_table); + +/* The Cavium ThunderX network controller can *only* be found in SoCs + * containing the ThunderX ARM64 CPU implementation. All accesses to the device + * registers on this platform are implicitly strongly ordered with respect + * to memory accesses. So writeq_relaxed() and readq_relaxed() are safe to use + * with no memory barriers in this driver. The readq()/writeq() functions add + * explicit ordering operation which in this case are redundant, and only + * add overhead. + */ + +/* Register read/write APIs */ +static u64 bgx_reg_read(struct bgx *bgx, u8 lmac, u64 offset) +{ + void __iomem *addr = bgx->reg_base + ((u32)lmac << 20) + offset; + + return readq_relaxed(addr); +} + +static void bgx_reg_write(struct bgx *bgx, u8 lmac, u64 offset, u64 val) +{ + void __iomem *addr = bgx->reg_base + ((u32)lmac << 20) + offset; + + writeq_relaxed(val, addr); +} + +static void bgx_reg_modify(struct bgx *bgx, u8 lmac, u64 offset, u64 val) +{ + void __iomem *addr = bgx->reg_base + ((u32)lmac << 20) + offset; + + writeq_relaxed(val | readq_relaxed(addr), addr); +} + +static int bgx_poll_reg(struct bgx *bgx, u8 lmac, u64 reg, u64 mask, bool zero) +{ + int timeout = 100; + u64 reg_val; + + while (timeout) { + reg_val = bgx_reg_read(bgx, lmac, reg); + if (zero && !(reg_val & mask)) + return 0; + if (!zero && (reg_val & mask)) + return 0; + usleep_range(1000, 2000); + timeout--; + } + return 1; +} + +/* Return number of BGX present in HW */ +unsigned bgx_get_map(int node) +{ + int i; + unsigned map = 0; + + for (i = 0; i < MAX_BGX_PER_CN88XX; i++) { + if (bgx_vnic[(node * MAX_BGX_PER_CN88XX) + i]) + map |= (1 << i); + } + + return map; +} +EXPORT_SYMBOL(bgx_get_map); + +/* Return number of LMAC configured for this BGX */ +int bgx_get_lmac_count(int node, int bgx_idx) +{ + struct bgx *bgx; + + bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + if (bgx) + return bgx->lmac_count; + + return 0; +} +EXPORT_SYMBOL(bgx_get_lmac_count); + +/* Returns the current link status of LMAC */ +void bgx_get_lmac_link_state(int node, int bgx_idx, int lmacid, void *status) +{ + struct bgx_link_status *link = (struct bgx_link_status *)status; + struct bgx *bgx; + struct lmac *lmac; + + bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + if (!bgx) + return; + + lmac = &bgx->lmac[lmacid]; + link->link_up = lmac->link_up; + link->duplex = lmac->last_duplex; + link->speed = lmac->last_speed; +} +EXPORT_SYMBOL(bgx_get_lmac_link_state); + +const char *bgx_get_lmac_mac(int node, int bgx_idx, int lmacid) +{ + struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + + if (bgx) + return bgx->lmac[lmacid].mac; + + return NULL; +} +EXPORT_SYMBOL(bgx_get_lmac_mac); + +void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const char *mac) +{ + struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + + if (!bgx) + return; + + ether_addr_copy(bgx->lmac[lmacid].mac, mac); +} +EXPORT_SYMBOL(bgx_set_lmac_mac); + +static void bgx_sgmii_change_link_state(struct lmac *lmac) +{ + struct bgx *bgx = lmac->bgx; + u64 cmr_cfg; + u64 port_cfg = 0; + u64 misc_ctl = 0; + + cmr_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_CMRX_CFG); + cmr_cfg &= ~CMR_EN; + bgx_reg_write(bgx, lmac->lmacid, BGX_CMRX_CFG, cmr_cfg); + + port_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG); + misc_ctl = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_PCS_MISCX_CTL); + + if (lmac->link_up) { + misc_ctl &= ~PCS_MISC_CTL_GMX_ENO; + port_cfg &= ~GMI_PORT_CFG_DUPLEX; + port_cfg |= (lmac->last_duplex << 2); + } else { + misc_ctl |= PCS_MISC_CTL_GMX_ENO; + } + + switch (lmac->last_speed) { + case 10: + port_cfg &= ~GMI_PORT_CFG_SPEED; /* speed 0 */ + port_cfg |= GMI_PORT_CFG_SPEED_MSB; /* speed_msb 1 */ + port_cfg &= ~GMI_PORT_CFG_SLOT_TIME; /* slottime 0 */ + misc_ctl &= ~PCS_MISC_CTL_SAMP_PT_MASK; + misc_ctl |= 50; /* samp_pt */ + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_TXX_SLOT, 64); + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_TXX_BURST, 0); + break; + case 100: + port_cfg &= ~GMI_PORT_CFG_SPEED; /* speed 0 */ + port_cfg &= ~GMI_PORT_CFG_SPEED_MSB; /* speed_msb 0 */ + port_cfg &= ~GMI_PORT_CFG_SLOT_TIME; /* slottime 0 */ + misc_ctl &= ~PCS_MISC_CTL_SAMP_PT_MASK; + misc_ctl |= 5; /* samp_pt */ + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_TXX_SLOT, 64); + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_TXX_BURST, 0); + break; + case 1000: + port_cfg |= GMI_PORT_CFG_SPEED; /* speed 1 */ + port_cfg &= ~GMI_PORT_CFG_SPEED_MSB; /* speed_msb 0 */ + port_cfg |= GMI_PORT_CFG_SLOT_TIME; /* slottime 1 */ + misc_ctl &= ~PCS_MISC_CTL_SAMP_PT_MASK; + misc_ctl |= 1; /* samp_pt */ + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_TXX_SLOT, 512); + if (lmac->last_duplex) + bgx_reg_write(bgx, lmac->lmacid, + BGX_GMP_GMI_TXX_BURST, 0); + else + bgx_reg_write(bgx, lmac->lmacid, + BGX_GMP_GMI_TXX_BURST, 8192); + break; + default: + break; + } + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_PCS_MISCX_CTL, misc_ctl); + bgx_reg_write(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG, port_cfg); + + port_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG); + + /* renable lmac */ + cmr_cfg |= CMR_EN; + bgx_reg_write(bgx, lmac->lmacid, BGX_CMRX_CFG, cmr_cfg); +} + +void bgx_lmac_handler(struct net_device *netdev) +{ + struct lmac *lmac = container_of(netdev, struct lmac, netdev); + struct phy_device *phydev = lmac->phydev; + int link_changed = 0; + + if (!lmac) + return; + + if (!phydev->link && lmac->last_link) + link_changed = -1; + + if (phydev->link && + (lmac->last_duplex != phydev->duplex || + lmac->last_link != phydev->link || + lmac->last_speed != phydev->speed)) { + link_changed = 1; + } + + lmac->last_link = phydev->link; + lmac->last_speed = phydev->speed; + lmac->last_duplex = phydev->duplex; + + if (!link_changed) + return; + + if (link_changed > 0) + lmac->link_up = true; + else + lmac->link_up = false; + + if (lmac->is_sgmii) + bgx_sgmii_change_link_state(lmac); + else + bgx_xaui_check_link(lmac); +} + +u64 bgx_get_rx_stats(int node, int bgx_idx, int lmac, int idx) +{ + struct bgx *bgx; + + bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + if (!bgx) + return 0; + + if (idx > 8) + lmac = 0; + return bgx_reg_read(bgx, lmac, BGX_CMRX_RX_STAT0 + (idx * 8)); +} +EXPORT_SYMBOL(bgx_get_rx_stats); + +u64 bgx_get_tx_stats(int node, int bgx_idx, int lmac, int idx) +{ + struct bgx *bgx; + + bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + if (!bgx) + return 0; + + return bgx_reg_read(bgx, lmac, BGX_CMRX_TX_STAT0 + (idx * 8)); +} +EXPORT_SYMBOL(bgx_get_tx_stats); + +static void bgx_flush_dmac_addrs(struct bgx *bgx, int lmac) +{ + u64 offset; + + while (bgx->lmac[lmac].dmac > 0) { + offset = ((bgx->lmac[lmac].dmac - 1) * sizeof(u64)) + + (lmac * MAX_DMAC_PER_LMAC * sizeof(u64)); + bgx_reg_write(bgx, 0, BGX_CMR_RX_DMACX_CAM + offset, 0); + bgx->lmac[lmac].dmac--; + } +} + +static int bgx_lmac_sgmii_init(struct bgx *bgx, int lmacid) +{ + u64 cfg; + + bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_THRESH, 0x30); + /* max packet size */ + bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_RXX_JABBER, MAX_FRAME_SIZE); + + /* Disable frame alignment if using preamble */ + cfg = bgx_reg_read(bgx, lmacid, BGX_GMP_GMI_TXX_APPEND); + if (cfg & 1) + bgx_reg_write(bgx, lmacid, BGX_GMP_GMI_TXX_SGMII_CTL, 0); + + /* Enable lmac */ + bgx_reg_modify(bgx, lmacid, BGX_CMRX_CFG, CMR_EN); + + /* PCS reset */ + bgx_reg_modify(bgx, lmacid, BGX_GMP_PCS_MRX_CTL, PCS_MRX_CTL_RESET); + if (bgx_poll_reg(bgx, lmacid, BGX_GMP_PCS_MRX_CTL, + PCS_MRX_CTL_RESET, true)) { + dev_err(&bgx->pdev->dev, "BGX PCS reset not completed\n"); + return -1; + } + + /* power down, reset autoneg, autoneg enable */ + cfg = bgx_reg_read(bgx, lmacid, BGX_GMP_PCS_MRX_CTL); + cfg &= ~PCS_MRX_CTL_PWR_DN; + cfg |= (PCS_MRX_CTL_RST_AN | PCS_MRX_CTL_AN_EN); + bgx_reg_write(bgx, lmacid, BGX_GMP_PCS_MRX_CTL, cfg); + + if (bgx_poll_reg(bgx, lmacid, BGX_GMP_PCS_MRX_STATUS, + PCS_MRX_STATUS_AN_CPT, false)) { + dev_err(&bgx->pdev->dev, "BGX AN_CPT not completed\n"); + return -1; + } + + return 0; +} + +static int bgx_lmac_xaui_init(struct bgx *bgx, int lmacid, int lmac_type) +{ + u64 cfg; + + /* Reset SPU */ + bgx_reg_modify(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_RESET); + if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_RESET, true)) { + dev_err(&bgx->pdev->dev, "BGX SPU reset not completed\n"); + return -1; + } + + /* Disable LMAC */ + cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); + cfg &= ~CMR_EN; + bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); + + bgx_reg_modify(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_LOW_POWER); + /* Set interleaved running disparity for RXAUI */ + if (bgx->lmac_type != BGX_MODE_RXAUI) + bgx_reg_modify(bgx, lmacid, + BGX_SPUX_MISC_CONTROL, SPU_MISC_CTL_RX_DIS); + else + bgx_reg_modify(bgx, lmacid, BGX_SPUX_MISC_CONTROL, + SPU_MISC_CTL_RX_DIS | SPU_MISC_CTL_INTLV_RDISP); + + /* clear all interrupts */ + cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_RX_INT); + bgx_reg_write(bgx, lmacid, BGX_SMUX_RX_INT, cfg); + cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_TX_INT); + bgx_reg_write(bgx, lmacid, BGX_SMUX_TX_INT, cfg); + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); + bgx_reg_write(bgx, lmacid, BGX_SPUX_INT, cfg); + + if (bgx->use_training) { + bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LP_CUP, 0x00); + bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LD_CUP, 0x00); + bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LD_REP, 0x00); + /* training enable */ + bgx_reg_modify(bgx, lmacid, + BGX_SPUX_BR_PMD_CRTL, SPU_PMD_CRTL_TRAIN_EN); + } + + /* Append FCS to each packet */ + bgx_reg_modify(bgx, lmacid, BGX_SMUX_TX_APPEND, SMU_TX_APPEND_FCS_D); + + /* Disable forward error correction */ + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_FEC_CONTROL); + cfg &= ~SPU_FEC_CTL_FEC_EN; + bgx_reg_write(bgx, lmacid, BGX_SPUX_FEC_CONTROL, cfg); + + /* Disable autoneg */ + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_AN_CONTROL); + cfg = cfg & ~(SPU_AN_CTL_AN_EN | SPU_AN_CTL_XNP_EN); + bgx_reg_write(bgx, lmacid, BGX_SPUX_AN_CONTROL, cfg); + + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_AN_ADV); + if (bgx->lmac_type == BGX_MODE_10G_KR) + cfg |= (1 << 23); + else if (bgx->lmac_type == BGX_MODE_40G_KR) + cfg |= (1 << 24); + else + cfg &= ~((1 << 23) | (1 << 24)); + cfg = cfg & (~((1ULL << 25) | (1ULL << 22) | (1ULL << 12))); + bgx_reg_write(bgx, lmacid, BGX_SPUX_AN_ADV, cfg); + + cfg = bgx_reg_read(bgx, 0, BGX_SPU_DBG_CONTROL); + cfg &= ~SPU_DBG_CTL_AN_ARB_LINK_CHK_EN; + bgx_reg_write(bgx, 0, BGX_SPU_DBG_CONTROL, cfg); + + /* Enable lmac */ + bgx_reg_modify(bgx, lmacid, BGX_CMRX_CFG, CMR_EN); + + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_CONTROL1); + cfg &= ~SPU_CTL_LOW_POWER; + bgx_reg_write(bgx, lmacid, BGX_SPUX_CONTROL1, cfg); + + cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_TX_CTL); + cfg &= ~SMU_TX_CTL_UNI_EN; + cfg |= SMU_TX_CTL_DIC_EN; + bgx_reg_write(bgx, lmacid, BGX_SMUX_TX_CTL, cfg); + + /* take lmac_count into account */ + bgx_reg_modify(bgx, lmacid, BGX_SMUX_TX_THRESH, (0x100 - 1)); + /* max packet size */ + bgx_reg_modify(bgx, lmacid, BGX_SMUX_RX_JABBER, MAX_FRAME_SIZE); + + return 0; +} + +static int bgx_xaui_check_link(struct lmac *lmac) +{ + struct bgx *bgx = lmac->bgx; + int lmacid = lmac->lmacid; + int lmac_type = bgx->lmac_type; + u64 cfg; + + bgx_reg_modify(bgx, lmacid, BGX_SPUX_MISC_CONTROL, SPU_MISC_CTL_RX_DIS); + if (bgx->use_training) { + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); + if (!(cfg & (1ull << 13))) { + cfg = (1ull << 13) | (1ull << 14); + bgx_reg_write(bgx, lmacid, BGX_SPUX_INT, cfg); + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_BR_PMD_CRTL); + cfg |= (1ull << 0); + bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_CRTL, cfg); + return -1; + } + } + + /* wait for PCS to come out of reset */ + if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_RESET, true)) { + dev_err(&bgx->pdev->dev, "BGX SPU reset not completed\n"); + return -1; + } + + if ((lmac_type == BGX_MODE_10G_KR) || (lmac_type == BGX_MODE_XFI) || + (lmac_type == BGX_MODE_40G_KR) || (lmac_type == BGX_MODE_XLAUI)) { + if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_BR_STATUS1, + SPU_BR_STATUS_BLK_LOCK, false)) { + dev_err(&bgx->pdev->dev, + "SPU_BR_STATUS_BLK_LOCK not completed\n"); + return -1; + } + } else { + if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_BX_STATUS, + SPU_BX_STATUS_RX_ALIGN, false)) { + dev_err(&bgx->pdev->dev, + "SPU_BX_STATUS_RX_ALIGN not completed\n"); + return -1; + } + } + + /* Clear rcvflt bit (latching high) and read it back */ + bgx_reg_modify(bgx, lmacid, BGX_SPUX_STATUS2, SPU_STATUS2_RCVFLT); + if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) { + dev_err(&bgx->pdev->dev, "Receive fault, retry training\n"); + if (bgx->use_training) { + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); + if (!(cfg & (1ull << 13))) { + cfg = (1ull << 13) | (1ull << 14); + bgx_reg_write(bgx, lmacid, BGX_SPUX_INT, cfg); + cfg = bgx_reg_read(bgx, lmacid, + BGX_SPUX_BR_PMD_CRTL); + cfg |= (1ull << 0); + bgx_reg_write(bgx, lmacid, + BGX_SPUX_BR_PMD_CRTL, cfg); + return -1; + } + } + return -1; + } + + /* Wait for MAC RX to be ready */ + if (bgx_poll_reg(bgx, lmacid, BGX_SMUX_RX_CTL, + SMU_RX_CTL_STATUS, true)) { + dev_err(&bgx->pdev->dev, "SMU RX link not okay\n"); + return -1; + } + + /* Wait for BGX RX to be idle */ + if (bgx_poll_reg(bgx, lmacid, BGX_SMUX_CTL, SMU_CTL_RX_IDLE, false)) { + dev_err(&bgx->pdev->dev, "SMU RX not idle\n"); + return -1; + } + + /* Wait for BGX TX to be idle */ + if (bgx_poll_reg(bgx, lmacid, BGX_SMUX_CTL, SMU_CTL_TX_IDLE, false)) { + dev_err(&bgx->pdev->dev, "SMU TX not idle\n"); + return -1; + } + + if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) { + dev_err(&bgx->pdev->dev, "Receive fault\n"); + return -1; + } + + /* Receive link is latching low. Force it high and verify it */ + bgx_reg_modify(bgx, lmacid, BGX_SPUX_STATUS1, SPU_STATUS1_RCV_LNK); + if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_STATUS1, + SPU_STATUS1_RCV_LNK, false)) { + dev_err(&bgx->pdev->dev, "SPU receive link down\n"); + return -1; + } + + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_MISC_CONTROL); + cfg &= ~SPU_MISC_CTL_RX_DIS; + bgx_reg_write(bgx, lmacid, BGX_SPUX_MISC_CONTROL, cfg); + return 0; +} + +static void bgx_poll_for_link(struct work_struct *work) +{ + struct lmac *lmac; + u64 link; + + lmac = container_of(work, struct lmac, dwork.work); + + /* Receive link is latching low. Force it high and verify it */ + bgx_reg_modify(lmac->bgx, lmac->lmacid, + BGX_SPUX_STATUS1, SPU_STATUS1_RCV_LNK); + bgx_poll_reg(lmac->bgx, lmac->lmacid, BGX_SPUX_STATUS1, + SPU_STATUS1_RCV_LNK, false); + + link = bgx_reg_read(lmac->bgx, lmac->lmacid, BGX_SPUX_STATUS1); + if (link & SPU_STATUS1_RCV_LNK) { + lmac->link_up = 1; + if (lmac->bgx->lmac_type == BGX_MODE_XLAUI) + lmac->last_speed = 40000; + else + lmac->last_speed = 10000; + lmac->last_duplex = 1; + } else { + lmac->link_up = 0; + } + + if (lmac->last_link != lmac->link_up) { + lmac->last_link = lmac->link_up; + if (lmac->link_up) + bgx_xaui_check_link(lmac); + } + + queue_delayed_work(lmac->check_link, &lmac->dwork, HZ * 2); +} + +static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) +{ + struct lmac *lmac; + u64 cfg; + + lmac = &bgx->lmac[lmacid]; + lmac->bgx = bgx; + + if (bgx->lmac_type == BGX_MODE_SGMII) { + lmac->is_sgmii = 1; + if (bgx_lmac_sgmii_init(bgx, lmacid)) + return -1; + } else { + lmac->is_sgmii = 0; + if (bgx_lmac_xaui_init(bgx, lmacid, bgx->lmac_type)) + return -1; + } + + if (lmac->is_sgmii) { + cfg = bgx_reg_read(bgx, lmacid, BGX_GMP_GMI_TXX_APPEND); + cfg |= ((1ull << 2) | (1ull << 1)); /* FCS and PAD */ + bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_APPEND, cfg); + bgx_reg_write(bgx, lmacid, BGX_GMP_GMI_TXX_MIN_PKT, 60 - 1); + } else { + cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_TX_APPEND); + cfg |= ((1ull << 2) | (1ull << 1)); /* FCS and PAD */ + bgx_reg_modify(bgx, lmacid, BGX_SMUX_TX_APPEND, cfg); + bgx_reg_write(bgx, lmacid, BGX_SMUX_TX_MIN_PKT, 60 + 4); + } + + /* Enable lmac */ + bgx_reg_modify(bgx, lmacid, BGX_CMRX_CFG, + CMR_EN | CMR_PKT_RX_EN | CMR_PKT_TX_EN); + + /* Restore default cfg, incase low level firmware changed it */ + bgx_reg_write(bgx, lmacid, BGX_CMRX_RX_DMAC_CTL, 0x03); + + if ((bgx->lmac_type != BGX_MODE_XFI) && + (bgx->lmac_type != BGX_MODE_XLAUI) && + (bgx->lmac_type != BGX_MODE_40G_KR) && + (bgx->lmac_type != BGX_MODE_10G_KR)) { + if (!lmac->phydev) + return -ENODEV; + + lmac->phydev->dev_flags = 0; + + if (phy_connect_direct(&lmac->netdev, lmac->phydev, + bgx_lmac_handler, + PHY_INTERFACE_MODE_SGMII)) + return -ENODEV; + + phy_start_aneg(lmac->phydev); + } else { + lmac->check_link = alloc_workqueue("check_link", WQ_UNBOUND | + WQ_MEM_RECLAIM, 1); + if (!lmac->check_link) + return -ENOMEM; + INIT_DELAYED_WORK(&lmac->dwork, bgx_poll_for_link); + queue_delayed_work(lmac->check_link, &lmac->dwork, 0); + } + + return 0; +} + +void bgx_lmac_disable(struct bgx *bgx, u8 lmacid) +{ + struct lmac *lmac; + u64 cmrx_cfg; + + lmac = &bgx->lmac[lmacid]; + if (lmac->check_link) { + /* Destroy work queue */ + cancel_delayed_work(&lmac->dwork); + flush_workqueue(lmac->check_link); + destroy_workqueue(lmac->check_link); + } + + cmrx_cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); + cmrx_cfg &= ~(1 << 15); + bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cmrx_cfg); + bgx_flush_dmac_addrs(bgx, lmacid); + + if (lmac->phydev) + phy_disconnect(lmac->phydev); + + lmac->phydev = NULL; +} + +static void bgx_set_num_ports(struct bgx *bgx) +{ + u64 lmac_count; + + switch (bgx->qlm_mode) { + case QLM_MODE_SGMII: + bgx->lmac_count = 4; + bgx->lmac_type = BGX_MODE_SGMII; + bgx->lane_to_sds = 0; + break; + case QLM_MODE_XAUI_1X4: + bgx->lmac_count = 1; + bgx->lmac_type = BGX_MODE_XAUI; + bgx->lane_to_sds = 0xE4; + break; + case QLM_MODE_RXAUI_2X2: + bgx->lmac_count = 2; + bgx->lmac_type = BGX_MODE_RXAUI; + bgx->lane_to_sds = 0xE4; + break; + case QLM_MODE_XFI_4X1: + bgx->lmac_count = 4; + bgx->lmac_type = BGX_MODE_XFI; + bgx->lane_to_sds = 0; + break; + case QLM_MODE_XLAUI_1X4: + bgx->lmac_count = 1; + bgx->lmac_type = BGX_MODE_XLAUI; + bgx->lane_to_sds = 0xE4; + break; + case QLM_MODE_10G_KR_4X1: + bgx->lmac_count = 4; + bgx->lmac_type = BGX_MODE_10G_KR; + bgx->lane_to_sds = 0; + bgx->use_training = 1; + break; + case QLM_MODE_40G_KR4_1X4: + bgx->lmac_count = 1; + bgx->lmac_type = BGX_MODE_40G_KR; + bgx->lane_to_sds = 0xE4; + bgx->use_training = 1; + break; + default: + bgx->lmac_count = 0; + break; + } + + /* Check if low level firmware has programmed LMAC count + * based on board type, if yes consider that otherwise + * the default static values + */ + lmac_count = bgx_reg_read(bgx, 0, BGX_CMR_RX_LMACS) & 0x7; + if (lmac_count != 4) + bgx->lmac_count = lmac_count; +} + +static void bgx_init_hw(struct bgx *bgx) +{ + int i; + + bgx_set_num_ports(bgx); + + bgx_reg_modify(bgx, 0, BGX_CMR_GLOBAL_CFG, CMR_GLOBAL_CFG_FCS_STRIP); + if (bgx_reg_read(bgx, 0, BGX_CMR_BIST_STATUS)) + dev_err(&bgx->pdev->dev, "BGX%d BIST failed\n", bgx->bgx_id); + + /* Set lmac type and lane2serdes mapping */ + for (i = 0; i < bgx->lmac_count; i++) { + if (bgx->lmac_type == BGX_MODE_RXAUI) { + if (i) + bgx->lane_to_sds = 0x0e; + else + bgx->lane_to_sds = 0x04; + bgx_reg_write(bgx, i, BGX_CMRX_CFG, + (bgx->lmac_type << 8) | bgx->lane_to_sds); + continue; + } + bgx_reg_write(bgx, i, BGX_CMRX_CFG, + (bgx->lmac_type << 8) | (bgx->lane_to_sds + i)); + bgx->lmac[i].lmacid_bd = lmac_count; + lmac_count++; + } + + bgx_reg_write(bgx, 0, BGX_CMR_TX_LMACS, bgx->lmac_count); + bgx_reg_write(bgx, 0, BGX_CMR_RX_LMACS, bgx->lmac_count); + + /* Set the backpressure AND mask */ + for (i = 0; i < bgx->lmac_count; i++) + bgx_reg_modify(bgx, 0, BGX_CMR_CHAN_MSK_AND, + ((1ULL << MAX_BGX_CHANS_PER_LMAC) - 1) << + (i * MAX_BGX_CHANS_PER_LMAC)); + + /* Disable all MAC filtering */ + for (i = 0; i < RX_DMAC_COUNT; i++) + bgx_reg_write(bgx, 0, BGX_CMR_RX_DMACX_CAM + (i * 8), 0x00); + + /* Disable MAC steering (NCSI traffic) */ + for (i = 0; i < RX_TRAFFIC_STEER_RULE_COUNT; i++) + bgx_reg_write(bgx, 0, BGX_CMR_RX_STREERING + (i * 8), 0x00); +} + +static void bgx_get_qlm_mode(struct bgx *bgx) +{ + struct device *dev = &bgx->pdev->dev; + int lmac_type; + int train_en; + + /* Read LMAC0 type to figure out QLM mode + * This is configured by low level firmware + */ + lmac_type = bgx_reg_read(bgx, 0, BGX_CMRX_CFG); + lmac_type = (lmac_type >> 8) & 0x07; + + train_en = bgx_reg_read(bgx, 0, BGX_SPUX_BR_PMD_CRTL) & + SPU_PMD_CRTL_TRAIN_EN; + + switch (lmac_type) { + case BGX_MODE_SGMII: + bgx->qlm_mode = QLM_MODE_SGMII; + dev_info(dev, "BGX%d QLM mode: SGMII\n", bgx->bgx_id); + break; + case BGX_MODE_XAUI: + bgx->qlm_mode = QLM_MODE_XAUI_1X4; + dev_info(dev, "BGX%d QLM mode: XAUI\n", bgx->bgx_id); + break; + case BGX_MODE_RXAUI: + bgx->qlm_mode = QLM_MODE_RXAUI_2X2; + dev_info(dev, "BGX%d QLM mode: RXAUI\n", bgx->bgx_id); + break; + case BGX_MODE_XFI: + if (!train_en) { + bgx->qlm_mode = QLM_MODE_XFI_4X1; + dev_info(dev, "BGX%d QLM mode: XFI\n", bgx->bgx_id); + } else { + bgx->qlm_mode = QLM_MODE_10G_KR_4X1; + dev_info(dev, "BGX%d QLM mode: 10G_KR\n", bgx->bgx_id); + } + break; + case BGX_MODE_XLAUI: + if (!train_en) { + bgx->qlm_mode = QLM_MODE_XLAUI_1X4; + dev_info(dev, "BGX%d QLM mode: XLAUI\n", bgx->bgx_id); + } else { + bgx->qlm_mode = QLM_MODE_40G_KR4_1X4; + dev_info(dev, "BGX%d QLM mode: 40G_KR4\n", bgx->bgx_id); + } + break; + default: + bgx->qlm_mode = QLM_MODE_SGMII; + dev_info(dev, "BGX%d QLM default mode: SGMII\n", bgx->bgx_id); + } +} + +static void bgx_init_of(struct bgx *bgx, struct device_node *np) +{ + struct device_node *np_child; + u8 lmac = 0; + + for_each_child_of_node(np, np_child) { + struct device_node *phy_np; + const char *mac; + + phy_np = of_parse_phandle(np_child, "phy-handle", 0); + if (phy_np) + bgx->lmac[lmac].phydev = of_phy_find_device(phy_np); + + mac = of_get_mac_address(np_child); + if (mac) + ether_addr_copy(bgx->lmac[lmac].mac, mac); + + SET_NETDEV_DEV(&bgx->lmac[lmac].netdev, &bgx->pdev->dev); + bgx->lmac[lmac].lmacid = lmac; + lmac++; + if (lmac == MAX_LMAC_PER_BGX) + break; + } +} + +static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + int err; + struct device *dev = &pdev->dev; + struct bgx *bgx = NULL; + struct device_node *np; + char bgx_sel[5]; + u8 lmac; + + bgx = devm_kzalloc(dev, sizeof(*bgx), GFP_KERNEL); + if (!bgx) + return -ENOMEM; + bgx->pdev = pdev; + + pci_set_drvdata(pdev, bgx); + + err = pci_enable_device(pdev); + if (err) { + dev_err(dev, "Failed to enable PCI device\n"); + pci_set_drvdata(pdev, NULL); + return err; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + dev_err(dev, "PCI request regions failed 0x%x\n", err); + goto err_disable_device; + } + + /* MAP configuration registers */ + bgx->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); + if (!bgx->reg_base) { + dev_err(dev, "BGX: Cannot map CSR memory space, aborting\n"); + err = -ENOMEM; + goto err_release_regions; + } + bgx->bgx_id = (pci_resource_start(pdev, PCI_CFG_REG_BAR_NUM) >> 24) & 1; + bgx->bgx_id += NODE_ID(pci_resource_start(pdev, PCI_CFG_REG_BAR_NUM)) + * MAX_BGX_PER_CN88XX; + bgx_vnic[bgx->bgx_id] = bgx; + bgx_get_qlm_mode(bgx); + + snprintf(bgx_sel, 5, "bgx%d", bgx->bgx_id); + np = of_find_node_by_name(NULL, bgx_sel); + if (np) + bgx_init_of(bgx, np); + + bgx_init_hw(bgx); + + /* Enable all LMACs */ + for (lmac = 0; lmac < bgx->lmac_count; lmac++) { + err = bgx_lmac_enable(bgx, lmac); + if (err) { + dev_err(dev, "BGX%d failed to enable lmac%d\n", + bgx->bgx_id, lmac); + goto err_enable; + } + } + + return 0; + +err_enable: + bgx_vnic[bgx->bgx_id] = NULL; +err_release_regions: + pci_release_regions(pdev); +err_disable_device: + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + return err; +} + +static void bgx_remove(struct pci_dev *pdev) +{ + struct bgx *bgx = pci_get_drvdata(pdev); + u8 lmac; + + /* Disable all LMACs */ + for (lmac = 0; lmac < bgx->lmac_count; lmac++) + bgx_lmac_disable(bgx, lmac); + + bgx_vnic[bgx->bgx_id] = NULL; + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); +} + +static struct pci_driver bgx_driver = { + .name = DRV_NAME, + .id_table = bgx_id_table, + .probe = bgx_probe, + .remove = bgx_remove, +}; + +static int __init bgx_init_module(void) +{ + pr_info("%s, ver %s\n", DRV_NAME, DRV_VERSION); + + return pci_register_driver(&bgx_driver); +} + +static void __exit bgx_cleanup_module(void) +{ + pci_unregister_driver(&bgx_driver); +} + +module_init(bgx_init_module); +module_exit(bgx_cleanup_module); diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h new file mode 100644 index 000000000000..9d91ce44f8d7 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h @@ -0,0 +1,223 @@ +/* + * Copyright (C) 2015 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#ifndef THUNDER_BGX_H +#define THUNDER_BGX_H + +#define MAX_BGX_THUNDER 8 /* Max 4 nodes, 2 per node */ +#define MAX_BGX_PER_CN88XX 2 +#define MAX_LMAC_PER_BGX 4 +#define MAX_BGX_CHANS_PER_LMAC 16 +#define MAX_DMAC_PER_LMAC 8 +#define MAX_FRAME_SIZE 9216 + +#define MAX_DMAC_PER_LMAC_TNS_BYPASS_MODE 2 + +#define MAX_LMAC (MAX_BGX_PER_CN88XX * MAX_LMAC_PER_BGX) + +#define NODE_ID_MASK 0x300000000000 +#define NODE_ID(x) ((x & NODE_ID_MASK) >> 44) + +/* Registers */ +#define BGX_CMRX_CFG 0x00 +#define CMR_PKT_TX_EN BIT_ULL(13) +#define CMR_PKT_RX_EN BIT_ULL(14) +#define CMR_EN BIT_ULL(15) +#define BGX_CMR_GLOBAL_CFG 0x08 +#define CMR_GLOBAL_CFG_FCS_STRIP BIT_ULL(6) +#define BGX_CMRX_RX_ID_MAP 0x60 +#define BGX_CMRX_RX_STAT0 0x70 +#define BGX_CMRX_RX_STAT1 0x78 +#define BGX_CMRX_RX_STAT2 0x80 +#define BGX_CMRX_RX_STAT3 0x88 +#define BGX_CMRX_RX_STAT4 0x90 +#define BGX_CMRX_RX_STAT5 0x98 +#define BGX_CMRX_RX_STAT6 0xA0 +#define BGX_CMRX_RX_STAT7 0xA8 +#define BGX_CMRX_RX_STAT8 0xB0 +#define BGX_CMRX_RX_STAT9 0xB8 +#define BGX_CMRX_RX_STAT10 0xC0 +#define BGX_CMRX_RX_BP_DROP 0xC8 +#define BGX_CMRX_RX_DMAC_CTL 0x0E8 +#define BGX_CMR_RX_DMACX_CAM 0x200 +#define RX_DMACX_CAM_EN BIT_ULL(48) +#define RX_DMACX_CAM_LMACID(x) (x << 49) +#define RX_DMAC_COUNT 32 +#define BGX_CMR_RX_STREERING 0x300 +#define RX_TRAFFIC_STEER_RULE_COUNT 8 +#define BGX_CMR_CHAN_MSK_AND 0x450 +#define BGX_CMR_BIST_STATUS 0x460 +#define BGX_CMR_RX_LMACS 0x468 +#define BGX_CMRX_TX_STAT0 0x600 +#define BGX_CMRX_TX_STAT1 0x608 +#define BGX_CMRX_TX_STAT2 0x610 +#define BGX_CMRX_TX_STAT3 0x618 +#define BGX_CMRX_TX_STAT4 0x620 +#define BGX_CMRX_TX_STAT5 0x628 +#define BGX_CMRX_TX_STAT6 0x630 +#define BGX_CMRX_TX_STAT7 0x638 +#define BGX_CMRX_TX_STAT8 0x640 +#define BGX_CMRX_TX_STAT9 0x648 +#define BGX_CMRX_TX_STAT10 0x650 +#define BGX_CMRX_TX_STAT11 0x658 +#define BGX_CMRX_TX_STAT12 0x660 +#define BGX_CMRX_TX_STAT13 0x668 +#define BGX_CMRX_TX_STAT14 0x670 +#define BGX_CMRX_TX_STAT15 0x678 +#define BGX_CMRX_TX_STAT16 0x680 +#define BGX_CMRX_TX_STAT17 0x688 +#define BGX_CMR_TX_LMACS 0x1000 + +#define BGX_SPUX_CONTROL1 0x10000 +#define SPU_CTL_LOW_POWER BIT_ULL(11) +#define SPU_CTL_RESET BIT_ULL(15) +#define BGX_SPUX_STATUS1 0x10008 +#define SPU_STATUS1_RCV_LNK BIT_ULL(2) +#define BGX_SPUX_STATUS2 0x10020 +#define SPU_STATUS2_RCVFLT BIT_ULL(10) +#define BGX_SPUX_BX_STATUS 0x10028 +#define SPU_BX_STATUS_RX_ALIGN BIT_ULL(12) +#define BGX_SPUX_BR_STATUS1 0x10030 +#define SPU_BR_STATUS_BLK_LOCK BIT_ULL(0) +#define SPU_BR_STATUS_RCV_LNK BIT_ULL(12) +#define BGX_SPUX_BR_PMD_CRTL 0x10068 +#define SPU_PMD_CRTL_TRAIN_EN BIT_ULL(1) +#define BGX_SPUX_BR_PMD_LP_CUP 0x10078 +#define BGX_SPUX_BR_PMD_LD_CUP 0x10088 +#define BGX_SPUX_BR_PMD_LD_REP 0x10090 +#define BGX_SPUX_FEC_CONTROL 0x100A0 +#define SPU_FEC_CTL_FEC_EN BIT_ULL(0) +#define SPU_FEC_CTL_ERR_EN BIT_ULL(1) +#define BGX_SPUX_AN_CONTROL 0x100C8 +#define SPU_AN_CTL_AN_EN BIT_ULL(12) +#define SPU_AN_CTL_XNP_EN BIT_ULL(13) +#define BGX_SPUX_AN_ADV 0x100D8 +#define BGX_SPUX_MISC_CONTROL 0x10218 +#define SPU_MISC_CTL_INTLV_RDISP BIT_ULL(10) +#define SPU_MISC_CTL_RX_DIS BIT_ULL(12) +#define BGX_SPUX_INT 0x10220 /* +(0..3) << 20 */ +#define BGX_SPUX_INT_W1S 0x10228 +#define BGX_SPUX_INT_ENA_W1C 0x10230 +#define BGX_SPUX_INT_ENA_W1S 0x10238 +#define BGX_SPU_DBG_CONTROL 0x10300 +#define SPU_DBG_CTL_AN_ARB_LINK_CHK_EN BIT_ULL(18) +#define SPU_DBG_CTL_AN_NONCE_MCT_DIS BIT_ULL(29) + +#define BGX_SMUX_RX_INT 0x20000 +#define BGX_SMUX_RX_JABBER 0x20030 +#define BGX_SMUX_RX_CTL 0x20048 +#define SMU_RX_CTL_STATUS (3ull << 0) +#define BGX_SMUX_TX_APPEND 0x20100 +#define SMU_TX_APPEND_FCS_D BIT_ULL(2) +#define BGX_SMUX_TX_MIN_PKT 0x20118 +#define BGX_SMUX_TX_INT 0x20140 +#define BGX_SMUX_TX_CTL 0x20178 +#define SMU_TX_CTL_DIC_EN BIT_ULL(0) +#define SMU_TX_CTL_UNI_EN BIT_ULL(1) +#define SMU_TX_CTL_LNK_STATUS (3ull << 4) +#define BGX_SMUX_TX_THRESH 0x20180 +#define BGX_SMUX_CTL 0x20200 +#define SMU_CTL_RX_IDLE BIT_ULL(0) +#define SMU_CTL_TX_IDLE BIT_ULL(1) + +#define BGX_GMP_PCS_MRX_CTL 0x30000 +#define PCS_MRX_CTL_RST_AN BIT_ULL(9) +#define PCS_MRX_CTL_PWR_DN BIT_ULL(11) +#define PCS_MRX_CTL_AN_EN BIT_ULL(12) +#define PCS_MRX_CTL_RESET BIT_ULL(15) +#define BGX_GMP_PCS_MRX_STATUS 0x30008 +#define PCS_MRX_STATUS_AN_CPT BIT_ULL(5) +#define BGX_GMP_PCS_ANX_AN_RESULTS 0x30020 +#define BGX_GMP_PCS_SGM_AN_ADV 0x30068 +#define BGX_GMP_PCS_MISCX_CTL 0x30078 +#define PCS_MISC_CTL_GMX_ENO BIT_ULL(11) +#define PCS_MISC_CTL_SAMP_PT_MASK 0x7Full +#define BGX_GMP_GMI_PRTX_CFG 0x38020 +#define GMI_PORT_CFG_SPEED BIT_ULL(1) +#define GMI_PORT_CFG_DUPLEX BIT_ULL(2) +#define GMI_PORT_CFG_SLOT_TIME BIT_ULL(3) +#define GMI_PORT_CFG_SPEED_MSB BIT_ULL(8) +#define BGX_GMP_GMI_RXX_JABBER 0x38038 +#define BGX_GMP_GMI_TXX_THRESH 0x38210 +#define BGX_GMP_GMI_TXX_APPEND 0x38218 +#define BGX_GMP_GMI_TXX_SLOT 0x38220 +#define BGX_GMP_GMI_TXX_BURST 0x38228 +#define BGX_GMP_GMI_TXX_MIN_PKT 0x38240 +#define BGX_GMP_GMI_TXX_SGMII_CTL 0x38300 + +#define BGX_MSIX_VEC_0_29_ADDR 0x400000 /* +(0..29) << 4 */ +#define BGX_MSIX_VEC_0_29_CTL 0x400008 +#define BGX_MSIX_PBA_0 0x4F0000 + +/* MSI-X interrupts */ +#define BGX_MSIX_VECTORS 30 +#define BGX_LMAC_VEC_OFFSET 7 +#define BGX_MSIX_VEC_SHIFT 4 + +#define CMRX_INT 0 +#define SPUX_INT 1 +#define SMUX_RX_INT 2 +#define SMUX_TX_INT 3 +#define GMPX_PCS_INT 4 +#define GMPX_GMI_RX_INT 5 +#define GMPX_GMI_TX_INT 6 +#define CMR_MEM_INT 28 +#define SPU_MEM_INT 29 + +#define LMAC_INTR_LINK_UP BIT(0) +#define LMAC_INTR_LINK_DOWN BIT(1) + +/* RX_DMAC_CTL configuration*/ +enum MCAST_MODE { + MCAST_MODE_REJECT, + MCAST_MODE_ACCEPT, + MCAST_MODE_CAM_FILTER, + RSVD +}; + +#define BCAST_ACCEPT 1 +#define CAM_ACCEPT 1 + +void bgx_add_dmac_addr(u64 dmac, int node, int bgx_idx, int lmac); +unsigned bgx_get_map(int node); +int bgx_get_lmac_count(int node, int bgx); +const char *bgx_get_lmac_mac(int node, int bgx_idx, int lmacid); +void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const char *mac); +void bgx_get_lmac_link_state(int node, int bgx_idx, int lmacid, void *status); +u64 bgx_get_rx_stats(int node, int bgx_idx, int lmac, int idx); +u64 bgx_get_tx_stats(int node, int bgx_idx, int lmac, int idx); +#define BGX_RX_STATS_COUNT 11 +#define BGX_TX_STATS_COUNT 18 + +struct bgx_stats { + u64 rx_stats[BGX_RX_STATS_COUNT]; + u64 tx_stats[BGX_TX_STATS_COUNT]; +}; + +enum LMAC_TYPE { + BGX_MODE_SGMII = 0, /* 1 lane, 1.250 Gbaud */ + BGX_MODE_XAUI = 1, /* 4 lanes, 3.125 Gbaud */ + BGX_MODE_DXAUI = 1, /* 4 lanes, 6.250 Gbaud */ + BGX_MODE_RXAUI = 2, /* 2 lanes, 6.250 Gbaud */ + BGX_MODE_XFI = 3, /* 1 lane, 10.3125 Gbaud */ + BGX_MODE_XLAUI = 4, /* 4 lanes, 10.3125 Gbaud */ + BGX_MODE_10G_KR = 3,/* 1 lane, 10.3125 Gbaud */ + BGX_MODE_40G_KR = 4,/* 4 lanes, 10.3125 Gbaud */ +}; + +enum qlm_mode { + QLM_MODE_SGMII, /* SGMII, each lane independent */ + QLM_MODE_XAUI_1X4, /* 1 XAUI or DXAUI, 4 lanes */ + QLM_MODE_RXAUI_2X2, /* 2 RXAUI, 2 lanes each */ + QLM_MODE_XFI_4X1, /* 4 XFI, 1 lane each */ + QLM_MODE_XLAUI_1X4, /* 1 XLAUI, 4 lanes each */ + QLM_MODE_10G_KR_4X1, /* 4 10GBASE-KR, 1 lane each */ + QLM_MODE_40G_KR4_1X4, /* 1 40GBASE-KR4, 4 lanes each */ +}; + +#endif /* THUNDER_BGX_H */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 524d11098c56..4d3a8c20eb12 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -224,7 +224,6 @@ struct sge_params { }; struct tp_params { - unsigned int ntxchan; /* # of Tx channels */ unsigned int tre; /* log2 of core clocks per TP tick */ unsigned int la_mask; /* what events are recorded by TP LA */ unsigned short tx_modq_map; /* TX modulation scheduler queue to */ @@ -273,6 +272,7 @@ struct pci_params { #define CHELSIO_T4 0x4 #define CHELSIO_T5 0x5 +#define CHELSIO_T6 0x6 enum chip_type { T4_A1 = CHELSIO_CHIP_CODE(CHELSIO_T4, 1), @@ -284,6 +284,10 @@ enum chip_type { T5_A1 = CHELSIO_CHIP_CODE(CHELSIO_T5, 1), T5_FIRST_REV = T5_A0, T5_LAST_REV = T5_A1, + + T6_A0 = CHELSIO_CHIP_CODE(CHELSIO_T6, 0), + T6_FIRST_REV = T6_A0, + T6_LAST_REV = T6_A0, }; struct devlog_params { @@ -292,6 +296,15 @@ struct devlog_params { u32 size; /* size of log */ }; +/* Stores chip specific parameters */ +struct arch_specific_params { + u8 nchan; + u16 mps_rplc_size; + u16 vfcount; + u32 sge_fl_db; + u16 mps_tcam_size; +}; + struct adapter_params { struct sge_params sge; struct tp_params tp; @@ -317,6 +330,7 @@ struct adapter_params { unsigned char nports; /* # of ethernet ports */ unsigned char portvec; enum chip_type chip; /* chip code */ + struct arch_specific_params arch; /* chip specific params */ unsigned char offload; unsigned char bypass; @@ -328,6 +342,17 @@ struct adapter_params { unsigned int max_ird_adapter; /* Max read depth per adapter */ }; +/* State needed to monitor the forward progress of SGE Ingress DMA activities + * and possible hangs. + */ +struct sge_idma_monitor_state { + unsigned int idma_1s_thresh; /* 1s threshold in Core Clock ticks */ + unsigned int idma_stalled[2]; /* synthesized stalled timers in HZ */ + unsigned int idma_state[2]; /* IDMA Hang detect state */ + unsigned int idma_qid[2]; /* IDMA Hung Ingress Queue ID */ + unsigned int idma_warn[2]; /* time to warning in HZ */ +}; + #include "t4fw_api.h" #define FW_VERSION(chip) ( \ @@ -630,12 +655,7 @@ struct sge { u32 fl_align; /* response queue message alignment */ u32 fl_starve_thres; /* Free List starvation threshold */ - /* State variables for detecting an SGE Ingress DMA hang */ - unsigned int idma_1s_thresh;/* SGE same State Counter 1s threshold */ - unsigned int idma_stalled[2];/* SGE synthesized stalled timers in HZ */ - unsigned int idma_state[2]; /* SGE IDMA Hang detect state */ - unsigned int idma_qid[2]; /* SGE IDMA Hung Ingress Queue ID */ - + struct sge_idma_monitor_state idma_monitor; unsigned int egr_start; unsigned int egr_sz; unsigned int ingr_start; @@ -644,6 +664,7 @@ struct sge { struct sge_rspq **ingr_map; /* qid->queue ingress queue map */ unsigned long *starving_fl; unsigned long *txq_maperr; + unsigned long *blocked_fl; struct timer_list rx_timer; /* refills starving FLs */ struct timer_list tx_timer; /* checks Tx queues */ }; @@ -672,7 +693,7 @@ struct adapter { struct pci_dev *pdev; struct device *pdev_dev; unsigned int mbox; - unsigned int fn; + unsigned int pf; unsigned int flags; enum chip_type chip; @@ -843,6 +864,11 @@ enum { VLAN_REWRITE }; +static inline int is_t6(enum chip_type chip) +{ + return CHELSIO_CHIP_VERSION(chip) == CHELSIO_T6; +} + static inline int is_t5(enum chip_type chip) { return CHELSIO_CHIP_VERSION(chip) == CHELSIO_T5; @@ -1055,7 +1081,7 @@ int t4_mgmt_tx(struct adapter *adap, struct sk_buff *skb); int t4_ofld_send(struct adapter *adap, struct sk_buff *skb); int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, struct net_device *dev, int intr_idx, - struct sge_fl *fl, rspq_handler_t hnd); + struct sge_fl *fl, rspq_handler_t hnd, int cong); int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq, struct net_device *dev, struct netdev_queue *netdevq, unsigned int iqid); @@ -1095,6 +1121,19 @@ static inline int is_bypass_device(int device) } } +static inline int is_10gbt_device(int device) +{ + /* this should be set based upon device capabilities */ + switch (device) { + case 0x4409: + case 0x4486: + return 1; + + default: + return 0; + } +} + static inline unsigned int core_ticks_per_usec(const struct adapter *adap) { return adap->params.vpd.cclk / 1000; @@ -1117,9 +1156,19 @@ static inline unsigned int core_ticks_to_us(const struct adapter *adapter, void t4_set_reg_field(struct adapter *adap, unsigned int addr, u32 mask, u32 val); +int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd, + int size, void *rpl, bool sleep_ok, int timeout); int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, void *rpl, bool sleep_ok); +static inline int t4_wr_mbox_timeout(struct adapter *adap, int mbox, + const void *cmd, int size, void *rpl, + int timeout) +{ + return t4_wr_mbox_meat_timeout(adap, mbox, cmd, size, rpl, true, + timeout); +} + static inline int t4_wr_mbox(struct adapter *adap, int mbox, const void *cmd, int size, void *rpl) { @@ -1151,6 +1200,10 @@ int t4_link_start(struct adapter *adap, unsigned int mbox, unsigned int port, struct link_config *lc); int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port); +u32 t4_read_pcie_cfg4(struct adapter *adap, int reg); +u32 t4_get_util_window(struct adapter *adap); +void t4_setup_memwin(struct adapter *adap, u32 memwin_base, u32 window); + #define T4_MEMORY_WRITE 0 #define T4_MEMORY_READ 1 int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, u32 len, @@ -1169,6 +1222,11 @@ int get_vpd_params(struct adapter *adapter, struct vpd_params *p); int t4_read_flash(struct adapter *adapter, unsigned int addr, unsigned int nwords, u32 *data, int byte_oriented); int t4_load_fw(struct adapter *adapter, const u8 *fw_data, unsigned int size); +int t4_load_phy_fw(struct adapter *adap, + int win, spinlock_t *lock, + int (*phy_fw_version)(const u8 *, size_t), + const u8 *phy_fw_data, size_t phy_fw_size); +int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver); int t4_fwcache(struct adapter *adap, enum fw_params_param_dev_fwcache op); int t4_fw_upgrade(struct adapter *adap, unsigned int mbox, const u8 *fw_data, unsigned int size, int force); @@ -1182,7 +1240,7 @@ int t4_prep_fw(struct adapter *adap, struct fw_info *fw_info, int t4_prep_adapter(struct adapter *adapter); enum t4_bar2_qtype { T4_BAR2_QTYPE_EGRESS, T4_BAR2_QTYPE_INGRESS }; -int cxgb4_t4_bar2_sge_qregs(struct adapter *adapter, +int t4_bar2_sge_qregs(struct adapter *adapter, unsigned int qid, enum t4_bar2_qtype qtype, u64 *pbar2_qoffset, @@ -1195,12 +1253,15 @@ int t4_init_devlog_params(struct adapter *adapter); int t4_init_sge_params(struct adapter *adapter); int t4_init_tp_params(struct adapter *adap); int t4_filter_field_shift(const struct adapter *adap, int filter_sel); +int t4_init_rss_mode(struct adapter *adap, int mbox); int t4_port_init(struct adapter *adap, int mbox, int pf, int vf); void t4_fatal_err(struct adapter *adapter); int t4_config_rss_range(struct adapter *adapter, int mbox, unsigned int viid, int start, int n, const u16 *rspq, unsigned int nrspq); int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode, unsigned int flags); +int t4_config_vi_rss(struct adapter *adapter, int mbox, unsigned int viid, + unsigned int flags, unsigned int defq); int t4_read_rss(struct adapter *adapter, u16 *entries); void t4_read_rss_key(struct adapter *adapter, u32 *key); void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx); @@ -1211,10 +1272,7 @@ void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index, u32 t4_read_rss_pf_map(struct adapter *adapter); u32 t4_read_rss_pf_mask(struct adapter *adapter); -int t4_mc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, - u64 *parity); -int t4_edc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, - u64 *parity); +unsigned int t4_get_mps_bg_map(struct adapter *adapter, int idx); void t4_pmtx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]); void t4_pmrx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]); int t4_read_cim_ibq(struct adapter *adap, unsigned int qid, u32 *data, @@ -1259,13 +1317,16 @@ int t4_fw_initialize(struct adapter *adap, unsigned int mbox); int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int nparams, const u32 *params, u32 *val); +int t4_query_params_rw(struct adapter *adap, unsigned int mbox, unsigned int pf, + unsigned int vf, unsigned int nparams, const u32 *params, + u32 *val, int rw); +int t4_set_params_timeout(struct adapter *adap, unsigned int mbox, + unsigned int pf, unsigned int vf, + unsigned int nparams, const u32 *params, + const u32 *val, int timeout); int t4_set_params(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int nparams, const u32 *params, const u32 *val); -int t4_set_params_nosleep(struct adapter *adap, unsigned int mbox, - unsigned int pf, unsigned int vf, - unsigned int nparams, const u32 *params, - const u32 *val); int t4_cfg_pfvf(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int txq, unsigned int txq_eth_ctrl, unsigned int rxqi, unsigned int rxq, unsigned int tc, @@ -1310,4 +1371,9 @@ int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox, u32 addr, u32 val); void t4_sge_decode_idma_state(struct adapter *adapter, int state); void t4_free_mem(void *addr); +void t4_idma_monitor_init(struct adapter *adapter, + struct sge_idma_monitor_state *idma); +void t4_idma_monitor(struct adapter *adapter, + struct sge_idma_monitor_state *idma, + int hz, int ticks); #endif /* __CXGB4_H__ */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index 371f75e782e5..3719807efddd 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -1084,41 +1084,89 @@ static inline void tcamxy2valmask(u64 x, u64 y, u8 *addr, u64 *mask) static int mps_tcam_show(struct seq_file *seq, void *v) { - if (v == SEQ_START_TOKEN) - seq_puts(seq, "Idx Ethernet address Mask Vld Ports PF" - " VF Replication " - "P0 P1 P2 P3 ML\n"); - else { + struct adapter *adap = seq->private; + unsigned int chip_ver = CHELSIO_CHIP_VERSION(adap->params.chip); + + if (v == SEQ_START_TOKEN) { + if (adap->params.arch.mps_rplc_size > 128) + seq_puts(seq, "Idx Ethernet address Mask " + "Vld Ports PF VF " + "Replication " + " P0 P1 P2 P3 ML\n"); + else + seq_puts(seq, "Idx Ethernet address Mask " + "Vld Ports PF VF Replication" + " P0 P1 P2 P3 ML\n"); + } else { u64 mask; u8 addr[ETH_ALEN]; - struct adapter *adap = seq->private; + bool replicate; unsigned int idx = (uintptr_t)v - 2; - u64 tcamy = t4_read_reg64(adap, MPS_CLS_TCAM_Y_L(idx)); - u64 tcamx = t4_read_reg64(adap, MPS_CLS_TCAM_X_L(idx)); - u32 cls_lo = t4_read_reg(adap, MPS_CLS_SRAM_L(idx)); - u32 cls_hi = t4_read_reg(adap, MPS_CLS_SRAM_H(idx)); - u32 rplc[4] = {0, 0, 0, 0}; + u64 tcamy, tcamx, val; + u32 cls_lo, cls_hi, ctl; + u32 rplc[8] = {0}; + + if (chip_ver > CHELSIO_T5) { + /* CtlCmdType - 0: Read, 1: Write + * CtlTcamSel - 0: TCAM0, 1: TCAM1 + * CtlXYBitSel- 0: Y bit, 1: X bit + */ + + /* Read tcamy */ + ctl = CTLCMDTYPE_V(0) | CTLXYBITSEL_V(0); + if (idx < 256) + ctl |= CTLTCAMINDEX_V(idx) | CTLTCAMSEL_V(0); + else + ctl |= CTLTCAMINDEX_V(idx - 256) | + CTLTCAMSEL_V(1); + t4_write_reg(adap, MPS_CLS_TCAM_DATA2_CTL_A, ctl); + val = t4_read_reg(adap, MPS_CLS_TCAM_DATA1_A); + tcamy = DMACH_G(val) << 32; + tcamy |= t4_read_reg(adap, MPS_CLS_TCAM_DATA0_A); + + /* Read tcamx. Change the control param */ + ctl |= CTLXYBITSEL_V(1); + t4_write_reg(adap, MPS_CLS_TCAM_DATA2_CTL_A, ctl); + val = t4_read_reg(adap, MPS_CLS_TCAM_DATA1_A); + tcamx = DMACH_G(val) << 32; + tcamx |= t4_read_reg(adap, MPS_CLS_TCAM_DATA0_A); + } else { + tcamy = t4_read_reg64(adap, MPS_CLS_TCAM_Y_L(idx)); + tcamx = t4_read_reg64(adap, MPS_CLS_TCAM_X_L(idx)); + } + + cls_lo = t4_read_reg(adap, MPS_CLS_SRAM_L(idx)); + cls_hi = t4_read_reg(adap, MPS_CLS_SRAM_H(idx)); if (tcamx & tcamy) { seq_printf(seq, "%3u -\n", idx); goto out; } - if (cls_lo & REPLICATE_F) { + rplc[0] = rplc[1] = rplc[2] = rplc[3] = 0; + if (chip_ver > CHELSIO_T5) + replicate = (cls_lo & T6_REPLICATE_F); + else + replicate = (cls_lo & REPLICATE_F); + + if (replicate) { struct fw_ldst_cmd ldst_cmd; int ret; + struct fw_ldst_mps_rplc mps_rplc; + u32 ldst_addrspc; memset(&ldst_cmd, 0, sizeof(ldst_cmd)); + ldst_addrspc = + FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MPS); ldst_cmd.op_to_addrspace = htonl(FW_CMD_OP_V(FW_LDST_CMD) | FW_CMD_REQUEST_F | FW_CMD_READ_F | - FW_LDST_CMD_ADDRSPACE_V( - FW_LDST_ADDRSPC_MPS)); + ldst_addrspc); ldst_cmd.cycles_to_len16 = htonl(FW_LEN16(ldst_cmd)); - ldst_cmd.u.mps.fid_ctl = + ldst_cmd.u.mps.rplc.fid_idx = htons(FW_LDST_CMD_FID_V(FW_LDST_MPS_RPLC) | - FW_LDST_CMD_CTL_V(idx)); + FW_LDST_CMD_IDX_V(idx)); ret = t4_wr_mbox(adap, adap->mbox, &ldst_cmd, sizeof(ldst_cmd), &ldst_cmd); if (ret) @@ -1126,30 +1174,69 @@ static int mps_tcam_show(struct seq_file *seq, void *v) "replication map for idx %d: %d\n", idx, -ret); else { - rplc[0] = ntohl(ldst_cmd.u.mps.rplc31_0); - rplc[1] = ntohl(ldst_cmd.u.mps.rplc63_32); - rplc[2] = ntohl(ldst_cmd.u.mps.rplc95_64); - rplc[3] = ntohl(ldst_cmd.u.mps.rplc127_96); + mps_rplc = ldst_cmd.u.mps.rplc; + rplc[0] = ntohl(mps_rplc.rplc31_0); + rplc[1] = ntohl(mps_rplc.rplc63_32); + rplc[2] = ntohl(mps_rplc.rplc95_64); + rplc[3] = ntohl(mps_rplc.rplc127_96); + if (adap->params.arch.mps_rplc_size > 128) { + rplc[4] = ntohl(mps_rplc.rplc159_128); + rplc[5] = ntohl(mps_rplc.rplc191_160); + rplc[6] = ntohl(mps_rplc.rplc223_192); + rplc[7] = ntohl(mps_rplc.rplc255_224); + } } } tcamxy2valmask(tcamx, tcamy, addr, &mask); - seq_printf(seq, "%3u %02x:%02x:%02x:%02x:%02x:%02x %012llx" - "%3c %#x%4u%4d", - idx, addr[0], addr[1], addr[2], addr[3], addr[4], - addr[5], (unsigned long long)mask, - (cls_lo & SRAM_VLD_F) ? 'Y' : 'N', PORTMAP_G(cls_hi), - PF_G(cls_lo), - (cls_lo & VF_VALID_F) ? VF_G(cls_lo) : -1); - if (cls_lo & REPLICATE_F) - seq_printf(seq, " %08x %08x %08x %08x", - rplc[3], rplc[2], rplc[1], rplc[0]); + if (chip_ver > CHELSIO_T5) + seq_printf(seq, "%3u %02x:%02x:%02x:%02x:%02x:%02x " + "%012llx%3c %#x%4u%4d", + idx, addr[0], addr[1], addr[2], addr[3], + addr[4], addr[5], (unsigned long long)mask, + (cls_lo & T6_SRAM_VLD_F) ? 'Y' : 'N', + PORTMAP_G(cls_hi), + T6_PF_G(cls_lo), + (cls_lo & T6_VF_VALID_F) ? + T6_VF_G(cls_lo) : -1); else - seq_printf(seq, "%36c", ' '); - seq_printf(seq, "%4u%3u%3u%3u %#x\n", - SRAM_PRIO0_G(cls_lo), SRAM_PRIO1_G(cls_lo), - SRAM_PRIO2_G(cls_lo), SRAM_PRIO3_G(cls_lo), - (cls_lo >> MULTILISTEN0_S) & 0xf); + seq_printf(seq, "%3u %02x:%02x:%02x:%02x:%02x:%02x " + "%012llx%3c %#x%4u%4d", + idx, addr[0], addr[1], addr[2], addr[3], + addr[4], addr[5], (unsigned long long)mask, + (cls_lo & SRAM_VLD_F) ? 'Y' : 'N', + PORTMAP_G(cls_hi), + PF_G(cls_lo), + (cls_lo & VF_VALID_F) ? VF_G(cls_lo) : -1); + + if (replicate) { + if (adap->params.arch.mps_rplc_size > 128) + seq_printf(seq, " %08x %08x %08x %08x " + "%08x %08x %08x %08x", + rplc[7], rplc[6], rplc[5], rplc[4], + rplc[3], rplc[2], rplc[1], rplc[0]); + else + seq_printf(seq, " %08x %08x %08x %08x", + rplc[3], rplc[2], rplc[1], rplc[0]); + } else { + if (adap->params.arch.mps_rplc_size > 128) + seq_printf(seq, "%72c", ' '); + else + seq_printf(seq, "%36c", ' '); + } + + if (chip_ver > CHELSIO_T5) + seq_printf(seq, "%4u%3u%3u%3u %#x\n", + T6_SRAM_PRIO0_G(cls_lo), + T6_SRAM_PRIO1_G(cls_lo), + T6_SRAM_PRIO2_G(cls_lo), + T6_SRAM_PRIO3_G(cls_lo), + (cls_lo >> T6_MULTILISTEN0_S) & 0xf); + else + seq_printf(seq, "%4u%3u%3u%3u %#x\n", + SRAM_PRIO0_G(cls_lo), SRAM_PRIO1_G(cls_lo), + SRAM_PRIO2_G(cls_lo), SRAM_PRIO3_G(cls_lo), + (cls_lo >> MULTILISTEN0_S) & 0xf); } out: return 0; } @@ -1222,7 +1309,7 @@ static int sensors_show(struct seq_file *seq, void *v) param[1] = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_DIAG) | FW_PARAMS_PARAM_Y_V(FW_PARAM_DEV_DIAG_VDD)); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, param, val); if (ret < 0 || val[0] == 0) @@ -1416,6 +1503,9 @@ static int rss_config_show(struct seq_file *seq, void *v) seq_printf(seq, " HashDelay: %3d\n", HASHDELAY_G(rssconf)); if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) seq_printf(seq, " VfWrAddr: %3d\n", VFWRADDR_G(rssconf)); + else + seq_printf(seq, " VfWrAddr: %3d\n", + T6_VFWRADDR_G(rssconf)); seq_printf(seq, " KeyMode: %s\n", keymode[KEYMODE_G(rssconf)]); seq_printf(seq, " VfWrEn: %3s\n", yesno(rssconf & VFWREN_F)); seq_printf(seq, " KeyWrEn: %3s\n", yesno(rssconf & KEYWREN_F)); @@ -1634,14 +1724,14 @@ static int rss_vf_config_open(struct inode *inode, struct file *file) struct adapter *adapter = inode->i_private; struct seq_tab *p; struct rss_vf_conf *vfconf; - int vf; + int vf, vfcount = adapter->params.arch.vfcount; - p = seq_open_tab(file, 128, sizeof(*vfconf), 1, rss_vf_config_show); + p = seq_open_tab(file, vfcount, sizeof(*vfconf), 1, rss_vf_config_show); if (!p) return -ENOMEM; vfconf = (struct rss_vf_conf *)p->data; - for (vf = 0; vf < 128; vf++) { + for (vf = 0; vf < vfcount; vf++) { t4_read_rss_vf_config(adapter, vf, &vfconf[vf].rss_vf_vfl, &vfconf[vf].rss_vf_vfh); } @@ -1959,6 +2049,61 @@ static void add_debugfs_mem(struct adapter *adap, const char *name, size_mb << 20); } +static int blocked_fl_open(struct inode *inode, struct file *file) +{ + file->private_data = inode->i_private; + return 0; +} + +static ssize_t blocked_fl_read(struct file *filp, char __user *ubuf, + size_t count, loff_t *ppos) +{ + int len; + const struct adapter *adap = filp->private_data; + char *buf; + ssize_t size = (adap->sge.egr_sz + 3) / 4 + + adap->sge.egr_sz / 32 + 2; /* includes ,/\n/\0 */ + + buf = kzalloc(size, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + len = snprintf(buf, size - 1, "%*pb\n", + adap->sge.egr_sz, adap->sge.blocked_fl); + len += sprintf(buf + len, "\n"); + size = simple_read_from_buffer(ubuf, count, ppos, buf, len); + t4_free_mem(buf); + return size; +} + +static ssize_t blocked_fl_write(struct file *filp, const char __user *ubuf, + size_t count, loff_t *ppos) +{ + int err; + unsigned long *t; + struct adapter *adap = filp->private_data; + + t = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz), sizeof(long), GFP_KERNEL); + if (!t) + return -ENOMEM; + + err = bitmap_parse_user(ubuf, count, t, adap->sge.egr_sz); + if (err) + return err; + + bitmap_copy(adap->sge.blocked_fl, t, adap->sge.egr_sz); + t4_free_mem(t); + return count; +} + +static const struct file_operations blocked_fl_fops = { + .owner = THIS_MODULE, + .open = blocked_fl_open, + .read = blocked_fl_read, + .write = blocked_fl_write, + .llseek = generic_file_llseek, +}; + /* Add an array of Debug FS files. */ void add_debugfs_files(struct adapter *adap, @@ -1978,7 +2123,7 @@ void add_debugfs_files(struct adapter *adap, int t4_setup_debugfs(struct adapter *adap) { int i; - u32 size; + u32 size = 0; struct dentry *de; static struct t4_debugfs_entry t4_debugfs_files[] = { @@ -2022,6 +2167,7 @@ int t4_setup_debugfs(struct adapter *adap) #if IS_ENABLED(CONFIG_IPV6) { "clip_tbl", &clip_tbl_debugfs_fops, S_IRUSR, 0 }, #endif + { "blocked_fl", &blocked_fl_fops, S_IRUSR | S_IWUSR, 0 }, }; /* Debug FS nodes common to all T5 and later adapters. @@ -2048,12 +2194,7 @@ int t4_setup_debugfs(struct adapter *adap) size = t4_read_reg(adap, MA_EDRAM1_BAR_A); add_debugfs_mem(adap, "edc1", MEM_EDC1, EDRAM1_SIZE_G(size)); } - if (is_t4(adap->params.chip)) { - size = t4_read_reg(adap, MA_EXT_MEMORY_BAR_A); - if (i & EXT_MEM_ENABLE_F) - add_debugfs_mem(adap, "mc", MEM_MC, - EXT_MEM_SIZE_G(size)); - } else { + if (is_t5(adap->params.chip)) { if (i & EXT_MEM0_ENABLE_F) { size = t4_read_reg(adap, MA_EXT_MEMORY0_BAR_A); add_debugfs_mem(adap, "mc0", MEM_MC0, @@ -2064,6 +2205,11 @@ int t4_setup_debugfs(struct adapter *adap) add_debugfs_mem(adap, "mc1", MEM_MC1, EXT_MEM1_SIZE_G(size)); } + } else { + if (i & EXT_MEM_ENABLE_F) + size = t4_read_reg(adap, MA_EXT_MEMORY_BAR_A); + add_debugfs_mem(adap, "mc", MEM_MC, + EXT_MEM_SIZE_G(size)); } de = debugfs_create_file_size("flash", S_IRUSR, adap->debugfs_root, adap, diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index 10d82b51d7ef..13d5101a0d9f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -250,7 +250,7 @@ static int restart_autoneg(struct net_device *dev) return -EAGAIN; if (p->link_cfg.autoneg != AUTONEG_ENABLE) return -EINVAL; - t4_restart_aneg(p->adapter, p->adapter->fn, p->tx_chan); + t4_restart_aneg(p->adapter, p->adapter->pf, p->tx_chan); return 0; } @@ -267,7 +267,7 @@ static int identify_port(struct net_device *dev, else return -EINVAL; - return t4_identify_port(adap, adap->fn, netdev2pinfo(dev)->viid, val); + return t4_identify_port(adap, adap->pf, netdev2pinfo(dev)->viid, val); } static unsigned int from_fw_linkcaps(enum fw_port_type type, unsigned int caps) @@ -439,7 +439,7 @@ static int set_settings(struct net_device *dev, struct ethtool_cmd *cmd) lc->autoneg = cmd->autoneg; if (netif_running(dev)) - return t4_link_start(p->adapter, p->adapter->fn, p->tx_chan, + return t4_link_start(p->adapter, p->adapter->pf, p->tx_chan, lc); return 0; } @@ -472,7 +472,7 @@ static int set_pauseparam(struct net_device *dev, if (epause->tx_pause) lc->requested_fc |= PAUSE_TX; if (netif_running(dev)) - return t4_link_start(p->adapter, p->adapter->fn, p->tx_chan, + return t4_link_start(p->adapter, p->adapter->pf, p->tx_chan, lc); return 0; } @@ -578,7 +578,7 @@ static int get_coalesce(struct net_device *dev, struct ethtool_coalesce *c) const struct sge_rspq *rq = &adap->sge.ethrxq[pi->first_qset].rspq; c->rx_coalesce_usecs = qtimer_val(adap, rq); - c->rx_max_coalesced_frames = (rq->intr_params & QINTR_CNT_EN) ? + c->rx_max_coalesced_frames = (rq->intr_params & QINTR_CNT_EN_F) ? adap->sge.counter_val[rq->pktcnt_idx] : 0; c->use_adaptive_rx_coalesce = get_adaptive_rx_setting(dev); return 0; @@ -617,7 +617,7 @@ static int eeprom_ptov(unsigned int phys_addr, unsigned int fn, unsigned int sz) */ static int eeprom_rd_phys(struct adapter *adap, unsigned int phys_addr, u32 *v) { - int vaddr = eeprom_ptov(phys_addr, adap->fn, EEPROMPFSIZE); + int vaddr = eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE); if (vaddr >= 0) vaddr = pci_read_vpd(adap->pdev, vaddr, sizeof(u32), v); @@ -626,7 +626,7 @@ static int eeprom_rd_phys(struct adapter *adap, unsigned int phys_addr, u32 *v) static int eeprom_wr_phys(struct adapter *adap, unsigned int phys_addr, u32 v) { - int vaddr = eeprom_ptov(phys_addr, adap->fn, EEPROMPFSIZE); + int vaddr = eeprom_ptov(phys_addr, adap->pf, EEPROMPFSIZE); if (vaddr >= 0) vaddr = pci_write_vpd(adap->pdev, vaddr, sizeof(u32), &v); @@ -669,8 +669,8 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, aligned_offset = eeprom->offset & ~3; aligned_len = (eeprom->len + (eeprom->offset & 3) + 3) & ~3; - if (adapter->fn > 0) { - u32 start = 1024 + adapter->fn * EEPROMPFSIZE; + if (adapter->pf > 0) { + u32 start = 1024 + adapter->pf * EEPROMPFSIZE; if (aligned_offset < start || aligned_offset + aligned_len > start + EEPROMPFSIZE) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 803d91beec6f..a589591e5b63 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -135,8 +135,14 @@ struct filter_entry { #define FW4_FNAME "cxgb4/t4fw.bin" #define FW5_FNAME "cxgb4/t5fw.bin" +#define FW6_FNAME "cxgb4/t6fw.bin" #define FW4_CFNAME "cxgb4/t4-config.txt" #define FW5_CFNAME "cxgb4/t5-config.txt" +#define FW6_CFNAME "cxgb4/t6-config.txt" +#define PHY_AQ1202_FIRMWARE "cxgb4/aq1202_fw.cld" +#define PHY_BCM84834_FIRMWARE "cxgb4/bcm8483.bin" +#define PHY_AQ1202_DEVICEID 0x4409 +#define PHY_BCM84834_DEVICEID 0x4486 MODULE_DESCRIPTION(DRV_DESC); MODULE_AUTHOR("Chelsio Communications"); @@ -318,8 +324,9 @@ static void dcb_tx_queue_prio_enable(struct net_device *dev, int enable) * level") we need to issue the Set Parameters Commannd * without sleeping (timeout < 0). */ - err = t4_set_params_nosleep(adap, adap->mbox, adap->fn, 0, 1, - &name, &value); + err = t4_set_params_timeout(adap, adap->mbox, adap->pf, 0, 1, + &name, &value, + -FW_CMD_MAX_TIMEOUT); if (err) dev_err(adap->pdev_dev, @@ -382,7 +389,7 @@ static int set_addr_filters(const struct net_device *dev, bool sleep) int uc_cnt = netdev_uc_count(dev); int mc_cnt = netdev_mc_count(dev); const struct port_info *pi = netdev_priv(dev); - unsigned int mb = pi->adapter->fn; + unsigned int mb = pi->adapter->pf; /* first do the secondary unicast addresses */ netdev_for_each_uc_addr(ha, dev) { @@ -439,7 +446,7 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok) ret = set_addr_filters(dev, sleep_ok); if (ret == 0) - ret = t4_set_rxmode(pi->adapter, pi->adapter->fn, pi->viid, mtu, + ret = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, mtu, (dev->flags & IFF_PROMISC) ? 1 : 0, (dev->flags & IFF_ALLMULTI) ? 1 : 0, 1, -1, sleep_ok); @@ -456,7 +463,7 @@ static int link_start(struct net_device *dev) { int ret; struct port_info *pi = netdev_priv(dev); - unsigned int mb = pi->adapter->fn; + unsigned int mb = pi->adapter->pf; /* * We do not set address filters and promiscuity here, the stack does @@ -856,23 +863,39 @@ static void free_msix_queue_irqs(struct adapter *adap) * * Sets up the portion of the HW RSS table for the port's VI to distribute * packets to the Rx queues in @queues. + * Should never be called before setting up sge eth rx queues */ int cxgb4_write_rss(const struct port_info *pi, const u16 *queues) { u16 *rss; int i, err; - const struct sge_eth_rxq *q = &pi->adapter->sge.ethrxq[pi->first_qset]; + struct adapter *adapter = pi->adapter; + const struct sge_eth_rxq *rxq; + rxq = &adapter->sge.ethrxq[pi->first_qset]; rss = kmalloc(pi->rss_size * sizeof(u16), GFP_KERNEL); if (!rss) return -ENOMEM; /* map the queue indices to queue ids */ for (i = 0; i < pi->rss_size; i++, queues++) - rss[i] = q[*queues].rspq.abs_id; + rss[i] = rxq[*queues].rspq.abs_id; - err = t4_config_rss_range(pi->adapter, pi->adapter->fn, pi->viid, 0, + err = t4_config_rss_range(adapter, adapter->pf, pi->viid, 0, pi->rss_size, rss, pi->rss_size); + /* If Tunnel All Lookup isn't specified in the global RSS + * Configuration, then we need to specify a default Ingress + * Queue for any ingress packets which aren't hashed. We'll + * use our first ingress queue ... + */ + if (!err) + err = t4_config_vi_rss(adapter, adapter->mbox, pi->viid, + FW_RSS_VI_CONFIG_CMD_IP6FOURTUPEN_F | + FW_RSS_VI_CONFIG_CMD_IP6TWOTUPEN_F | + FW_RSS_VI_CONFIG_CMD_IP4FOURTUPEN_F | + FW_RSS_VI_CONFIG_CMD_IP4TWOTUPEN_F | + FW_RSS_VI_CONFIG_CMD_UDPEN_F, + rss[0]); kfree(rss); return err; } @@ -885,11 +908,15 @@ int cxgb4_write_rss(const struct port_info *pi, const u16 *queues) */ static int setup_rss(struct adapter *adap) { - int i, err; + int i, j, err; for_each_port(adap, i) { const struct port_info *pi = adap2pinfo(adap, i); + /* Fill default values with equal distribution */ + for (j = 0; j < pi->rss_size; j++) + pi->rss[j] = j % pi->nqsets; + err = cxgb4_write_rss(pi, pi->rss); if (err) return err; @@ -977,7 +1004,7 @@ static int alloc_ofld_rxqs(struct adapter *adap, struct sge_ofld_rxq *q, err = t4_sge_alloc_rxq(adap, &q->rspq, false, adap->port[i / per_chan], msi_idx, q->fl.size ? &q->fl : NULL, - uldrx_handler); + uldrx_handler, 0); if (err) return err; memset(&q->stats, 0, sizeof(q->stats)); @@ -1007,7 +1034,7 @@ static int setup_sge_queues(struct adapter *adap) msi_idx = 1; /* vector 0 is for non-queue interrupts */ else { err = t4_sge_alloc_rxq(adap, &s->intrq, false, adap->port[0], 0, - NULL, NULL); + NULL, NULL, -1); if (err) return err; msi_idx = -((int)s->intrq.abs_id + 1); @@ -1027,7 +1054,7 @@ static int setup_sge_queues(struct adapter *adap) * new/deleted queues. */ err = t4_sge_alloc_rxq(adap, &s->fw_evtq, true, adap->port[0], - msi_idx, NULL, fwevtq_handler); + msi_idx, NULL, fwevtq_handler, -1); if (err) { freeout: t4_free_sge_resources(adap); return err; @@ -1044,7 +1071,9 @@ freeout: t4_free_sge_resources(adap); msi_idx++; err = t4_sge_alloc_rxq(adap, &q->rspq, false, dev, msi_idx, &q->fl, - t4_ethrx_handler); + t4_ethrx_handler, + t4_get_mps_bg_map(adap, + pi->tx_chan)); if (err) goto freeout; q->rspq.idx = j; @@ -1389,8 +1418,8 @@ int cxgb4_set_rspq_intr_params(struct sge_rspq *q, FW_PARAMS_PARAM_X_V( FW_PARAMS_PARAM_DMAQ_IQ_INTCNTTHRESH) | FW_PARAMS_PARAM_YZ_V(q->cntxt_id); - err = t4_set_params(adap, adap->fn, adap->fn, 0, 1, &v, - &new_idx); + err = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, + &v, &new_idx); if (err) return err; } @@ -1398,7 +1427,7 @@ int cxgb4_set_rspq_intr_params(struct sge_rspq *q, } us = us == 0 ? 6 : closest_timer(&adap->sge, us); - q->intr_params = QINTR_TIMER_IDX(us) | (cnt > 0 ? QINTR_CNT_EN : 0); + q->intr_params = QINTR_TIMER_IDX_V(us) | QINTR_CNT_EN_V(cnt > 0); return 0; } @@ -1411,7 +1440,7 @@ static int cxgb_set_features(struct net_device *dev, netdev_features_t features) if (!(changed & NETIF_F_HW_VLAN_CTAG_RX)) return 0; - err = t4_set_rxmode(pi->adapter, pi->adapter->fn, pi->viid, -1, + err = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, -1, -1, -1, -1, !!(features & NETIF_F_HW_VLAN_CTAG_RX), true); if (unlikely(err)) @@ -1694,7 +1723,7 @@ static int tid_init(struct tid_info *t) bitmap_zero(t->stid_bmap, t->nstids + t->nsftids); /* Reserve stid 0 for T4/T5 adapters */ if (!t->stid_base && - (is_t4(adap->params.chip) || is_t5(adap->params.chip))) + (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5)) __set_bit(0, t->stid_bmap); return 0; @@ -2042,25 +2071,6 @@ out: } EXPORT_SYMBOL(cxgb4_sync_txq_pidx); -void cxgb4_disable_db_coalescing(struct net_device *dev) -{ - struct adapter *adap; - - adap = netdev2adap(dev); - t4_set_reg_field(adap, SGE_DOORBELL_CONTROL_A, NOCOALESCE_F, - NOCOALESCE_F); -} -EXPORT_SYMBOL(cxgb4_disable_db_coalescing); - -void cxgb4_enable_db_coalescing(struct net_device *dev) -{ - struct adapter *adap; - - adap = netdev2adap(dev); - t4_set_reg_field(adap, SGE_DOORBELL_CONTROL_A, NOCOALESCE_F, 0); -} -EXPORT_SYMBOL(cxgb4_enable_db_coalescing); - int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte) { struct adapter *adap; @@ -2100,10 +2110,7 @@ int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte) if (offset < mc0_end) { memtype = MEM_MC0; memaddr = offset - edc1_end; - } else if (is_t4(adap->params.chip)) { - /* T4 only has a single memory channel */ - goto err; - } else { + } else if (is_t5(adap->params.chip)) { size = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A); mc1_size = EXT_MEM1_SIZE_G(size) << 20; mc1_end = mc0_end + mc1_size; @@ -2114,6 +2121,9 @@ int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte) /* offset beyond the end of any memory */ goto err; } + } else { + /* T4/T6 only has a single memory channel */ + goto err; } } @@ -2148,7 +2158,7 @@ int cxgb4_bar2_sge_qregs(struct net_device *dev, u64 *pbar2_qoffset, unsigned int *pbar2_qid) { - return cxgb4_t4_bar2_sge_qregs(netdev2adap(dev), + return t4_bar2_sge_qregs(netdev2adap(dev), qid, (qtype == CXGB4_BAR2_QTYPE_EGRESS ? T4_BAR2_QTYPE_EGRESS @@ -2278,9 +2288,13 @@ static void process_db_full(struct work_struct *work) drain_db_fifo(adap, dbfifo_drain_delay); enable_dbs(adap); notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY); - t4_set_reg_field(adap, SGE_INT_ENABLE3_A, - DBFIFO_HP_INT_F | DBFIFO_LP_INT_F, - DBFIFO_HP_INT_F | DBFIFO_LP_INT_F); + if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) + t4_set_reg_field(adap, SGE_INT_ENABLE3_A, + DBFIFO_HP_INT_F | DBFIFO_LP_INT_F, + DBFIFO_HP_INT_F | DBFIFO_LP_INT_F); + else + t4_set_reg_field(adap, SGE_INT_ENABLE3_A, + DBFIFO_LP_INT_F, DBFIFO_LP_INT_F); } static void sync_txq_pidx(struct adapter *adap, struct sge_txq *q) @@ -2342,7 +2356,7 @@ static void process_db_drop(struct work_struct *work) drain_db_fifo(adap, dbfifo_drain_delay); enable_dbs(adap); notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY); - } else { + } else if (is_t5(adap->params.chip)) { u32 dropped_db = t4_read_reg(adap, 0x010ac); u16 qid = (dropped_db >> 15) & 0x1ffff; u16 pidx_inc = dropped_db & 0x1fff; @@ -2350,7 +2364,7 @@ static void process_db_drop(struct work_struct *work) unsigned int bar2_qid; int ret; - ret = cxgb4_t4_bar2_sge_qregs(adap, qid, T4_BAR2_QTYPE_EGRESS, + ret = t4_bar2_sge_qregs(adap, qid, T4_BAR2_QTYPE_EGRESS, &bar2_qoffset, &bar2_qid); if (ret) dev_err(adap->pdev_dev, "doorbell drop recovery: " @@ -2363,7 +2377,8 @@ static void process_db_drop(struct work_struct *work) t4_set_reg_field(adap, 0x10b0, 1<<15, 1<<15); } - t4_set_reg_field(adap, SGE_DOORBELL_CONTROL_A, DROPPED_DB_F, 0); + if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) + t4_set_reg_field(adap, SGE_DOORBELL_CONTROL_A, DROPPED_DB_F, 0); } void t4_db_full(struct adapter *adap) @@ -2393,7 +2408,7 @@ static void uld_attach(struct adapter *adap, unsigned int uld) unsigned short i; lli.pdev = adap->pdev; - lli.pf = adap->fn; + lli.pf = adap->pf; lli.l2t = adap->l2t; lli.tids = &adap->tids; lli.ports = adap->port; @@ -2432,6 +2447,7 @@ static void uld_attach(struct adapter *adap, unsigned int uld) lli.max_ordird_qp = adap->params.max_ordird_qp; lli.max_ird_adapter = adap->params.max_ird_adapter; lli.ulptx_memwrite_dsgl = adap->params.ulptx_memwrite_dsgl; + lli.nodeid = dev_to_node(adap->pdev_dev); handle = ulds[uld].add(&lli); if (IS_ERR(handle)) { @@ -2729,7 +2745,7 @@ static int cxgb_close(struct net_device *dev) netif_tx_stop_all_queues(dev); netif_carrier_off(dev); - return t4_enable_vi(adapter, adapter->fn, pi->viid, false, false); + return t4_enable_vi(adapter, adapter->pf, pi->viid, false, false); } /* Return an error number if the indicated filter isn't writable ... @@ -2932,7 +2948,7 @@ static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd) } else return -EINVAL; - mbox = pi->adapter->fn; + mbox = pi->adapter->pf; if (cmd == SIOCGMIIREG) ret = t4_mdio_rd(pi->adapter, mbox, prtad, devad, data->reg_num, &data->val_out); @@ -2959,7 +2975,7 @@ static int cxgb_change_mtu(struct net_device *dev, int new_mtu) if (new_mtu < 81 || new_mtu > MAX_MTU) /* accommodate SACK */ return -EINVAL; - ret = t4_set_rxmode(pi->adapter, pi->adapter->fn, pi->viid, new_mtu, -1, + ret = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, new_mtu, -1, -1, -1, -1, true); if (!ret) dev->mtu = new_mtu; @@ -2975,7 +2991,7 @@ static int cxgb_set_mac_addr(struct net_device *dev, void *p) if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; - ret = t4_change_mac(pi->adapter, pi->adapter->fn, pi->viid, + ret = t4_change_mac(pi->adapter, pi->adapter->pf, pi->viid, pi->xact_addr_filt, addr->sa_data, true, true); if (ret < 0) return ret; @@ -3034,86 +3050,11 @@ void t4_fatal_err(struct adapter *adap) dev_alert(adap->pdev_dev, "encountered fatal error, adapter stopped\n"); } -/* Return the specified PCI-E Configuration Space register from our Physical - * Function. We try first via a Firmware LDST Command since we prefer to let - * the firmware own all of these registers, but if that fails we go for it - * directly ourselves. - */ -static u32 t4_read_pcie_cfg4(struct adapter *adap, int reg) -{ - struct fw_ldst_cmd ldst_cmd; - u32 val; - int ret; - - /* Construct and send the Firmware LDST Command to retrieve the - * specified PCI-E Configuration Space register. - */ - memset(&ldst_cmd, 0, sizeof(ldst_cmd)); - ldst_cmd.op_to_addrspace = - htonl(FW_CMD_OP_V(FW_LDST_CMD) | - FW_CMD_REQUEST_F | - FW_CMD_READ_F | - FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_FUNC_PCIE)); - ldst_cmd.cycles_to_len16 = htonl(FW_LEN16(ldst_cmd)); - ldst_cmd.u.pcie.select_naccess = FW_LDST_CMD_NACCESS_V(1); - ldst_cmd.u.pcie.ctrl_to_fn = - (FW_LDST_CMD_LC_F | FW_LDST_CMD_FN_V(adap->fn)); - ldst_cmd.u.pcie.r = reg; - ret = t4_wr_mbox(adap, adap->mbox, &ldst_cmd, sizeof(ldst_cmd), - &ldst_cmd); - - /* If the LDST Command suucceeded, exctract the returned register - * value. Otherwise read it directly ourself. - */ - if (ret == 0) - val = ntohl(ldst_cmd.u.pcie.data[0]); - else - t4_hw_pci_read_cfg4(adap, reg, &val); - - return val; -} - static void setup_memwin(struct adapter *adap) { - u32 mem_win0_base, mem_win1_base, mem_win2_base, mem_win2_aperture; + u32 nic_win_base = t4_get_util_window(adap); - if (is_t4(adap->params.chip)) { - u32 bar0; - - /* Truncation intentional: we only read the bottom 32-bits of - * the 64-bit BAR0/BAR1 ... We use the hardware backdoor - * mechanism to read BAR0 instead of using - * pci_resource_start() because we could be operating from - * within a Virtual Machine which is trapping our accesses to - * our Configuration Space and we need to set up the PCI-E - * Memory Window decoders with the actual addresses which will - * be coming across the PCI-E link. - */ - bar0 = t4_read_pcie_cfg4(adap, PCI_BASE_ADDRESS_0); - bar0 &= PCI_BASE_ADDRESS_MEM_MASK; - adap->t4_bar0 = bar0; - - mem_win0_base = bar0 + MEMWIN0_BASE; - mem_win1_base = bar0 + MEMWIN1_BASE; - mem_win2_base = bar0 + MEMWIN2_BASE; - mem_win2_aperture = MEMWIN2_APERTURE; - } else { - /* For T5, only relative offset inside the PCIe BAR is passed */ - mem_win0_base = MEMWIN0_BASE; - mem_win1_base = MEMWIN1_BASE; - mem_win2_base = MEMWIN2_BASE_T5; - mem_win2_aperture = MEMWIN2_APERTURE_T5; - } - t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, 0), - mem_win0_base | BIR_V(0) | - WINDOW_V(ilog2(MEMWIN0_APERTURE) - 10)); - t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, 1), - mem_win1_base | BIR_V(0) | - WINDOW_V(ilog2(MEMWIN1_APERTURE) - 10)); - t4_write_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, 2), - mem_win2_base | BIR_V(0) | - WINDOW_V(ilog2(mem_win2_aperture) - 10)); - t4_read_reg(adap, PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, 2)); + t4_setup_memwin(adap, nic_win_base, MEMWIN_NIC); } static void setup_memwin_rdma(struct adapter *adap) @@ -3147,7 +3088,7 @@ static int adap_init1(struct adapter *adap, struct fw_caps_config_cmd *c) c->op_to_write = htonl(FW_CMD_OP_V(FW_CAPS_CONFIG_CMD) | FW_CMD_REQUEST_F | FW_CMD_READ_F); c->cfvalid_to_len16 = htonl(FW_LEN16(*c)); - ret = t4_wr_mbox(adap, adap->fn, c, sizeof(*c), c); + ret = t4_wr_mbox(adap, adap->mbox, c, sizeof(*c), c); if (ret < 0) return ret; @@ -3163,18 +3104,18 @@ static int adap_init1(struct adapter *adap, struct fw_caps_config_cmd *c) } c->op_to_write = htonl(FW_CMD_OP_V(FW_CAPS_CONFIG_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F); - ret = t4_wr_mbox(adap, adap->fn, c, sizeof(*c), NULL); + ret = t4_wr_mbox(adap, adap->mbox, c, sizeof(*c), NULL); if (ret < 0) return ret; - ret = t4_config_glbl_rss(adap, adap->fn, + ret = t4_config_glbl_rss(adap, adap->pf, FW_RSS_GLB_CONFIG_CMD_MODE_BASICVIRTUAL, FW_RSS_GLB_CONFIG_CMD_TNLMAPEN_F | FW_RSS_GLB_CONFIG_CMD_TNLALLLKP_F); if (ret < 0) return ret; - ret = t4_cfg_pfvf(adap, adap->fn, adap->fn, 0, adap->sge.egr_sz, 64, + ret = t4_cfg_pfvf(adap, adap->mbox, adap->pf, 0, adap->sge.egr_sz, 64, MAX_INGQ, 0, 0, 4, 0xf, 0xf, 16, FW_CMD_CAP_PF, FW_CMD_CAP_PF); if (ret < 0) @@ -3218,7 +3159,7 @@ static int adap_init1(struct adapter *adap, struct fw_caps_config_cmd *c) } /* get basic stuff going */ - return t4_early_init(adap, adap->fn); + return t4_early_init(adap, adap->pf); } /* @@ -3274,6 +3215,142 @@ static int adap_init0_tweaks(struct adapter *adapter) return 0; } +/* 10Gb/s-BT PHY Support. chip-external 10Gb/s-BT PHYs are complex chips + * unto themselves and they contain their own firmware to perform their + * tasks ... + */ +static int phy_aq1202_version(const u8 *phy_fw_data, + size_t phy_fw_size) +{ + int offset; + + /* At offset 0x8 you're looking for the primary image's + * starting offset which is 3 Bytes wide + * + * At offset 0xa of the primary image, you look for the offset + * of the DRAM segment which is 3 Bytes wide. + * + * The FW version is at offset 0x27e of the DRAM and is 2 Bytes + * wide + */ + #define be16(__p) (((__p)[0] << 8) | (__p)[1]) + #define le16(__p) ((__p)[0] | ((__p)[1] << 8)) + #define le24(__p) (le16(__p) | ((__p)[2] << 16)) + + offset = le24(phy_fw_data + 0x8) << 12; + offset = le24(phy_fw_data + offset + 0xa); + return be16(phy_fw_data + offset + 0x27e); + + #undef be16 + #undef le16 + #undef le24 +} + +static struct info_10gbt_phy_fw { + unsigned int phy_fw_id; /* PCI Device ID */ + char *phy_fw_file; /* /lib/firmware/ PHY Firmware file */ + int (*phy_fw_version)(const u8 *phy_fw_data, size_t phy_fw_size); + int phy_flash; /* Has FLASH for PHY Firmware */ +} phy_info_array[] = { + { + PHY_AQ1202_DEVICEID, + PHY_AQ1202_FIRMWARE, + phy_aq1202_version, + 1, + }, + { + PHY_BCM84834_DEVICEID, + PHY_BCM84834_FIRMWARE, + NULL, + 0, + }, + { 0, NULL, NULL }, +}; + +static struct info_10gbt_phy_fw *find_phy_info(int devid) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(phy_info_array); i++) { + if (phy_info_array[i].phy_fw_id == devid) + return &phy_info_array[i]; + } + return NULL; +} + +/* Handle updating of chip-external 10Gb/s-BT PHY firmware. This needs to + * happen after the FW_RESET_CMD but before the FW_INITIALIZE_CMD. On error + * we return a negative error number. If we transfer new firmware we return 1 + * (from t4_load_phy_fw()). If we don't do anything we return 0. + */ +static int adap_init0_phy(struct adapter *adap) +{ + const struct firmware *phyf; + int ret; + struct info_10gbt_phy_fw *phy_info; + + /* Use the device ID to determine which PHY file to flash. + */ + phy_info = find_phy_info(adap->pdev->device); + if (!phy_info) { + dev_warn(adap->pdev_dev, + "No PHY Firmware file found for this PHY\n"); + return -EOPNOTSUPP; + } + + /* If we have a T4 PHY firmware file under /lib/firmware/cxgb4/, then + * use that. The adapter firmware provides us with a memory buffer + * where we can load a PHY firmware file from the host if we want to + * override the PHY firmware File in flash. + */ + ret = request_firmware_direct(&phyf, phy_info->phy_fw_file, + adap->pdev_dev); + if (ret < 0) { + /* For adapters without FLASH attached to PHY for their + * firmware, it's obviously a fatal error if we can't get the + * firmware to the adapter. For adapters with PHY firmware + * FLASH storage, it's worth a warning if we can't find the + * PHY Firmware but we'll neuter the error ... + */ + dev_err(adap->pdev_dev, "unable to find PHY Firmware image " + "/lib/firmware/%s, error %d\n", + phy_info->phy_fw_file, -ret); + if (phy_info->phy_flash) { + int cur_phy_fw_ver = 0; + + t4_phy_fw_ver(adap, &cur_phy_fw_ver); + dev_warn(adap->pdev_dev, "continuing with, on-adapter " + "FLASH copy, version %#x\n", cur_phy_fw_ver); + ret = 0; + } + + return ret; + } + + /* Load PHY Firmware onto adapter. + */ + ret = t4_load_phy_fw(adap, MEMWIN_NIC, &adap->win0_lock, + phy_info->phy_fw_version, + (u8 *)phyf->data, phyf->size); + if (ret < 0) + dev_err(adap->pdev_dev, "PHY Firmware transfer error %d\n", + -ret); + else if (ret > 0) { + int new_phy_fw_ver = 0; + + if (phy_info->phy_fw_version) + new_phy_fw_ver = phy_info->phy_fw_version(phyf->data, + phyf->size); + dev_info(adap->pdev_dev, "Successfully transferred PHY " + "Firmware /lib/firmware/%s, version %#x\n", + phy_info->phy_fw_file, new_phy_fw_ver); + } + + release_firmware(phyf); + + return ret; +} + /* * Attempt to initialize the adapter via a Firmware Configuration File. */ @@ -3298,6 +3375,16 @@ static int adap_init0_config(struct adapter *adapter, int reset) goto bye; } + /* If this is a 10Gb/s-BT adapter make sure the chip-external + * 10Gb/s-BT PHYs have up-to-date firmware. Note that this step needs + * to be performed after any global adapter RESET above since some + * PHYs only have local RAM copies of the PHY firmware. + */ + if (is_10gbt_device(adapter->pdev->device)) { + ret = adap_init0_phy(adapter); + if (ret < 0) + goto bye; + } /* * If we have a T4 configuration file under /lib/firmware/cxgb4/, * then use that. Otherwise, use the configuration file stored @@ -3310,6 +3397,9 @@ static int adap_init0_config(struct adapter *adapter, int reset) case CHELSIO_T5: fw_config_file = FW5_CFNAME; break; + case CHELSIO_T6: + fw_config_file = FW6_CFNAME; + break; default: dev_err(adapter->pdev_dev, "Device %d is not supported\n", adapter->pdev->device); @@ -3335,7 +3425,7 @@ static int adap_init0_config(struct adapter *adapter, int reset) params[0] = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_CF)); ret = t4_query_params(adapter, adapter->mbox, - adapter->fn, 0, 1, params, val); + adapter->pf, 0, 1, params, val); if (ret == 0) { /* * For t4_memory_rw() below addresses and @@ -3506,7 +3596,24 @@ static struct fw_info fw_info_array[] = { .intfver_iscsi = FW_INTFVER(T5, ISCSI), .intfver_fcoe = FW_INTFVER(T5, FCOE), }, + }, { + .chip = CHELSIO_T6, + .fs_name = FW6_CFNAME, + .fw_mod_name = FW6_FNAME, + .fw_hdr = { + .chip = FW_HDR_CHIP_T6, + .fw_ver = __cpu_to_be32(FW_VERSION(T6)), + .intfver_nic = FW_INTFVER(T6, NIC), + .intfver_vnic = FW_INTFVER(T6, VNIC), + .intfver_ofld = FW_INTFVER(T6, OFLD), + .intfver_ri = FW_INTFVER(T6, RI), + .intfver_iscsipdu = FW_INTFVER(T6, ISCSIPDU), + .intfver_iscsi = FW_INTFVER(T6, ISCSI), + .intfver_fcoepdu = FW_INTFVER(T6, FCOEPDU), + .intfver_fcoe = FW_INTFVER(T6, FCOE), + }, } + }; static struct fw_info *find_fw_info(int chip) @@ -3624,7 +3731,7 @@ static int adap_init0(struct adapter *adap) v = FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 1, &v, &port_vec); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &v, &port_vec); if (ret < 0) goto bye; @@ -3647,7 +3754,7 @@ static int adap_init0(struct adapter *adap) */ params[0] = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_CF)); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 1, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, params, val); /* If the firmware doesn't support Configuration Files, @@ -3706,7 +3813,7 @@ static int adap_init0(struct adapter *adap) params[3] = FW_PARAM_PFVF(FILTER_START); params[4] = FW_PARAM_PFVF(FILTER_END); params[5] = FW_PARAM_PFVF(IQFLINT_START); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 6, params, val); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 6, params, val); if (ret < 0) goto bye; adap->sge.egr_start = val[0]; @@ -3724,7 +3831,7 @@ static int adap_init0(struct adapter *adap) */ params[0] = FW_PARAM_PFVF(EQ_END); params[1] = FW_PARAM_PFVF(IQFLINT_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, params, val); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, params, val); if (ret < 0) goto bye; adap->sge.egr_sz = val[0] - adap->sge.egr_start + 1; @@ -3745,7 +3852,7 @@ static int adap_init0(struct adapter *adap) } /* Allocate the memory for the vaious egress queue bitmaps - * ie starving_fl and txq_maperr. + * ie starving_fl, txq_maperr and blocked_fl. */ adap->sge.starving_fl = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz), sizeof(long), GFP_KERNEL); @@ -3761,9 +3868,18 @@ static int adap_init0(struct adapter *adap) goto bye; } +#ifdef CONFIG_DEBUG_FS + adap->sge.blocked_fl = kcalloc(BITS_TO_LONGS(adap->sge.egr_sz), + sizeof(long), GFP_KERNEL); + if (!adap->sge.blocked_fl) { + ret = -ENOMEM; + goto bye; + } +#endif + params[0] = FW_PARAM_PFVF(CLIP_START); params[1] = FW_PARAM_PFVF(CLIP_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, params, val); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, params, val); if (ret < 0) goto bye; adap->clipt_start = val[0]; @@ -3772,7 +3888,7 @@ static int adap_init0(struct adapter *adap) /* query params related to active filter region */ params[0] = FW_PARAM_PFVF(ACTIVE_FILTER_START); params[1] = FW_PARAM_PFVF(ACTIVE_FILTER_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, params, val); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, params, val); /* If Active filter size is set we enable establishing * offload connection through firmware work request */ @@ -3789,7 +3905,7 @@ static int adap_init0(struct adapter *adap) */ params[0] = FW_PARAM_PFVF(CPLFW4MSG_ENCAP); val[0] = 1; - (void) t4_set_params(adap, adap->mbox, adap->fn, 0, 1, params, val); + (void)t4_set_params(adap, adap->mbox, adap->pf, 0, 1, params, val); /* * Find out whether we're allowed to use the T5+ ULPTX MEMWRITE DSGL @@ -3801,7 +3917,7 @@ static int adap_init0(struct adapter *adap) adap->params.ulptx_memwrite_dsgl = false; } else { params[0] = FW_PARAM_DEV(ULPTX_MEMWRITE_DSGL); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, params, val); adap->params.ulptx_memwrite_dsgl = (ret == 0 && val[0] != 0); } @@ -3827,7 +3943,7 @@ static int adap_init0(struct adapter *adap) params[3] = FW_PARAM_PFVF(TDDP_START); params[4] = FW_PARAM_PFVF(TDDP_END); params[5] = FW_PARAM_DEV(FLOWC_BUFFIFO_SZ); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 6, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 6, params, val); if (ret < 0) goto bye; @@ -3865,7 +3981,7 @@ static int adap_init0(struct adapter *adap) params[3] = FW_PARAM_PFVF(RQ_END); params[4] = FW_PARAM_PFVF(PBL_START); params[5] = FW_PARAM_PFVF(PBL_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 6, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 6, params, val); if (ret < 0) goto bye; @@ -3882,7 +3998,7 @@ static int adap_init0(struct adapter *adap) params[3] = FW_PARAM_PFVF(CQ_END); params[4] = FW_PARAM_PFVF(OCQ_START); params[5] = FW_PARAM_PFVF(OCQ_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 6, params, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 6, params, val); if (ret < 0) goto bye; @@ -3895,7 +4011,7 @@ static int adap_init0(struct adapter *adap) params[0] = FW_PARAM_DEV(MAXORDIRD_QP); params[1] = FW_PARAM_DEV(MAXIRD_ADAPTER); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, params, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, params, val); if (ret < 0) { adap->params.max_ordird_qp = 8; @@ -3913,7 +4029,7 @@ static int adap_init0(struct adapter *adap) if (caps_cmd.iscsicaps) { params[0] = FW_PARAM_PFVF(ISCSI_START); params[1] = FW_PARAM_PFVF(ISCSI_END); - ret = t4_query_params(adap, adap->mbox, adap->fn, 0, 2, + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 2, params, val); if (ret < 0) goto bye; @@ -3973,6 +4089,9 @@ bye: kfree(adap->sge.ingr_map); kfree(adap->sge.starving_fl); kfree(adap->sge.txq_maperr); +#ifdef CONFIG_DEBUG_FS + kfree(adap->sge.blocked_fl); +#endif if (ret != -ETIMEDOUT && ret != -EIO) t4_fw_bye(adap, adap->mbox); return ret; @@ -4040,7 +4159,7 @@ static pci_ers_result_t eeh_slot_reset(struct pci_dev *pdev) if (t4_wait_dev_ready(adap->regs) < 0) return PCI_ERS_RESULT_DISCONNECT; - if (t4_fw_hello(adap, adap->fn, adap->fn, MASTER_MUST, NULL) < 0) + if (t4_fw_hello(adap, adap->mbox, adap->pf, MASTER_MUST, NULL) < 0) return PCI_ERS_RESULT_DISCONNECT; adap->flags |= FW_OK; if (adap_init1(adap, &c)) @@ -4049,7 +4168,7 @@ static pci_ers_result_t eeh_slot_reset(struct pci_dev *pdev) for_each_port(adap, i) { struct port_info *p = adap2pinfo(adap, i); - ret = t4_alloc_vi(adap, adap->fn, p->tx_chan, adap->fn, 0, 1, + ret = t4_alloc_vi(adap, adap->mbox, p->tx_chan, adap->pf, 0, 1, NULL, NULL); if (ret < 0) return PCI_ERS_RESULT_DISCONNECT; @@ -4340,7 +4459,12 @@ static int enable_msix(struct adapter *adap) static int init_rss(struct adapter *adap) { - unsigned int i, j; + unsigned int i; + int err; + + err = t4_init_rss_mode(adap, adap->mbox); + if (err) + return err; for_each_port(adap, i) { struct port_info *pi = adap2pinfo(adap, i); @@ -4348,8 +4472,6 @@ static int init_rss(struct adapter *adap) pi->rss = kcalloc(pi->rss_size, sizeof(u16), GFP_KERNEL); if (!pi->rss) return -ENOMEM; - for (j = 0; j < pi->rss_size; j++) - pi->rss[j] = ethtool_rxfh_indir_default(j, pi->nqsets); } return 0; } @@ -4413,6 +4535,9 @@ static void free_some_resources(struct adapter *adapter) kfree(adapter->sge.ingr_map); kfree(adapter->sge.starving_fl); kfree(adapter->sge.txq_maperr); +#ifdef CONFIG_DEBUG_FS + kfree(adapter->sge.blocked_fl); +#endif disable_msi(adapter); for_each_port(adapter, i) @@ -4421,7 +4546,7 @@ static void free_some_resources(struct adapter *adapter) free_netdev(adapter->port[i]); } if (adapter->flags & FW_OK) - t4_fw_bye(adapter, adapter->fn); + t4_fw_bye(adapter, adapter->pf); } #define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN) @@ -4512,7 +4637,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) adapter->pdev = pdev; adapter->pdev_dev = &pdev->dev; adapter->mbox = func; - adapter->fn = func; + adapter->pf = func; adapter->msg_enable = dflt_msg_enable; memset(adapter->chan_map, 0xff, sizeof(adapter->chan_map)); @@ -4532,7 +4657,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (!is_t4(adapter->params.chip)) { s_qpp = (QUEUESPERPAGEPF0_S + (QUEUESPERPAGEPF1_S - QUEUESPERPAGEPF0_S) * - adapter->fn); + adapter->pf); qpp = 1 << QUEUESPERPAGEPF0_G(t4_read_reg(adapter, SGE_EGRESS_QUEUES_PER_PAGE_PF_A) >> s_qpp); num_seg = PAGE_SIZE / SEGMENT_SIZE; @@ -4559,6 +4684,9 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) setup_memwin(adapter); err = adap_init0(adapter); +#ifdef CONFIG_DEBUG_FS + bitmap_zero(adapter->sge.blocked_fl, adapter->sge.egr_sz); +#endif setup_memwin_rdma(adapter); if (err) goto out_unmap_bar; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h index 78ab4d406ce2..14e8110b5dbb 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h @@ -264,6 +264,7 @@ struct cxgb4_lld_info { unsigned int max_ordird_qp; /* Max ORD/IRD depth per RDMA QP */ unsigned int max_ird_adapter; /* Max IRD memory per adapter */ bool ulptx_memwrite_dsgl; /* use of T5 DSGL allowed */ + int nodeid; /* device numa node id */ }; struct cxgb4_uld_info { @@ -297,8 +298,6 @@ struct sk_buff *cxgb4_pktgl_to_skb(const struct pkt_gl *gl, unsigned int skb_len, unsigned int pull_len); int cxgb4_sync_txq_pidx(struct net_device *dev, u16 qid, u16 pidx, u16 size); int cxgb4_flush_eq_cache(struct net_device *dev); -void cxgb4_disable_db_coalescing(struct net_device *dev); -void cxgb4_enable_db_coalescing(struct net_device *dev); int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte); u64 cxgb4_read_sge_timestamp(struct net_device *dev); diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 0d2eddab04ef..6b7c37fd0252 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -100,16 +100,6 @@ */ #define TX_QCHECK_PERIOD (HZ / 2) -/* SGE Hung Ingress DMA Threshold Warning time (in Hz) and Warning Repeat Rate - * (in RX_QCHECK_PERIOD multiples). If we find one of the SGE Ingress DMA - * State Machines in the same state for this amount of time (in HZ) then we'll - * issue a warning about a potential hang. We'll repeat the warning as the - * SGE Ingress DMA Channel appears to be hung every N RX_QCHECK_PERIODs till - * the situation clears. If the situation clears, we'll note that as well. - */ -#define SGE_IDMA_WARN_THRESH (1 * HZ) -#define SGE_IDMA_WARN_REPEAT (20 * RX_QCHECK_PERIOD) - /* * Max number of Tx descriptors to be reclaimed by the Tx timer. */ @@ -532,14 +522,17 @@ static void unmap_rx_buf(struct adapter *adap, struct sge_fl *q) static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q) { - u32 val; if (q->pend_cred >= 8) { + u32 val = adap->params.arch.sge_fl_db; + if (is_t4(adap->params.chip)) - val = PIDX_V(q->pend_cred / 8); + val |= PIDX_V(q->pend_cred / 8); else - val = PIDX_T5_V(q->pend_cred / 8) | - DBTYPE_F; - val |= DBPRIO_F; + val |= PIDX_T5_V(q->pend_cred / 8); + + /* Make sure all memory writes to the Free List queue are + * committed before we tell the hardware about them. + */ wmb(); /* If we don't have access to the new User Doorbell (T5+), use @@ -594,6 +587,11 @@ static unsigned int refill_fl(struct adapter *adap, struct sge_fl *q, int n, struct rx_sw_desc *sd = &q->sdesc[q->pidx]; int node; +#ifdef CONFIG_DEBUG_FS + if (test_bit(q->cntxt_id - adap->sge.egr_start, adap->sge.blocked_fl)) + goto out; +#endif + gfp |= __GFP_NOWARN; node = dev_to_node(adap->pdev_dev); @@ -930,7 +928,10 @@ static void cxgb_pio_copy(u64 __iomem *dst, u64 *src) */ static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n) { - wmb(); /* write descriptors before telling HW */ + /* Make sure that all writes to the TX Descriptors are committed + * before we tell the hardware about them. + */ + wmb(); /* If we don't have access to the new User Doorbell (T5+), use the old * doorbell mechanism; otherwise use the new BAR2 mechanism. @@ -1032,7 +1033,7 @@ static void inline_tx_skb(const struct sk_buff *skb, const struct sge_txq *q, * Figure out what HW csum a packet wants and return the appropriate control * bits. */ -static u64 hwcsum(const struct sk_buff *skb) +static u64 hwcsum(enum chip_type chip, const struct sk_buff *skb) { int csum_type; const struct iphdr *iph = ip_hdr(skb); @@ -1047,7 +1048,7 @@ nocsum: /* * unknown protocol, disable HW csum * and hope a bad packet is detected */ - return TXPKT_L4CSUM_DIS; + return TXPKT_L4CSUM_DIS_F; } } else { /* @@ -1063,15 +1064,21 @@ nocsum: /* goto nocsum; } - if (likely(csum_type >= TX_CSUM_TCPIP)) - return TXPKT_CSUM_TYPE(csum_type) | - TXPKT_IPHDR_LEN(skb_network_header_len(skb)) | - TXPKT_ETHHDR_LEN(skb_network_offset(skb) - ETH_HLEN); - else { + if (likely(csum_type >= TX_CSUM_TCPIP)) { + u64 hdr_len = TXPKT_IPHDR_LEN_V(skb_network_header_len(skb)); + int eth_hdr_len = skb_network_offset(skb) - ETH_HLEN; + + if (CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5) + hdr_len |= TXPKT_ETHHDR_LEN_V(eth_hdr_len); + else + hdr_len |= T6_TXPKT_ETHHDR_LEN_V(eth_hdr_len); + return TXPKT_CSUM_TYPE_V(csum_type) | hdr_len; + } else { int start = skb_transport_offset(skb); - return TXPKT_CSUM_TYPE(csum_type) | TXPKT_CSUM_START(start) | - TXPKT_CSUM_LOC(start + skb->csum_offset); + return TXPKT_CSUM_TYPE_V(csum_type) | + TXPKT_CSUM_START_V(start) | + TXPKT_CSUM_LOC_V(start + skb->csum_offset); } } @@ -1112,11 +1119,11 @@ cxgb_fcoe_offload(struct sk_buff *skb, struct adapter *adap, return -ENOTSUPP; /* FC CRC offload */ - *cntrl = TXPKT_CSUM_TYPE(TX_CSUM_FCOE) | - TXPKT_L4CSUM_DIS | TXPKT_IPCSUM_DIS | - TXPKT_CSUM_START(CXGB_FCOE_TXPKT_CSUM_START) | - TXPKT_CSUM_END(CXGB_FCOE_TXPKT_CSUM_END) | - TXPKT_CSUM_LOC(CXGB_FCOE_TXPKT_CSUM_END); + *cntrl = TXPKT_CSUM_TYPE_V(TX_CSUM_FCOE) | + TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F | + TXPKT_CSUM_START_V(CXGB_FCOE_TXPKT_CSUM_START) | + TXPKT_CSUM_END_V(CXGB_FCOE_TXPKT_CSUM_END) | + TXPKT_CSUM_LOC_V(CXGB_FCOE_TXPKT_CSUM_END); return 0; } #endif /* CONFIG_CHELSIO_T4_FCOE */ @@ -1130,7 +1137,6 @@ cxgb_fcoe_offload(struct sk_buff *skb, struct adapter *adap, */ netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev) { - int len; u32 wr_mid; u64 cntrl, *end; int qidx, credits; @@ -1143,6 +1149,7 @@ netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev) const struct skb_shared_info *ssi; dma_addr_t addr[MAX_SKB_FRAGS + 1]; bool immediate = false; + int len, max_pkt_len; #ifdef CONFIG_CHELSIO_T4_FCOE int err; #endif /* CONFIG_CHELSIO_T4_FCOE */ @@ -1156,13 +1163,20 @@ out_free: dev_kfree_skb_any(skb); return NETDEV_TX_OK; } + /* Discard the packet if the length is greater than mtu */ + max_pkt_len = ETH_HLEN + dev->mtu; + if (skb_vlan_tag_present(skb)) + max_pkt_len += VLAN_HLEN; + if (!skb_shinfo(skb)->gso_size && (unlikely(skb->len > max_pkt_len))) + goto out_free; + pi = netdev_priv(dev); adap = pi->adapter; qidx = skb_get_queue_mapping(skb); q = &adap->sge.ethtxq[qidx + pi->first_qset]; reclaim_completed_tx(adap, &q->q, true); - cntrl = TXPKT_L4CSUM_DIS | TXPKT_IPCSUM_DIS; + cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F; #ifdef CONFIG_CHELSIO_T4_FCOE err = cxgb_fcoe_offload(skb, adap, pi, &cntrl); @@ -1213,23 +1227,29 @@ out_free: dev_kfree_skb_any(skb); len += sizeof(*lso); wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) | FW_WR_IMMDLEN_V(len)); - lso->c.lso_ctrl = htonl(LSO_OPCODE(CPL_TX_PKT_LSO) | - LSO_FIRST_SLICE | LSO_LAST_SLICE | - LSO_IPV6(v6) | - LSO_ETHHDR_LEN(eth_xtra_len / 4) | - LSO_IPHDR_LEN(l3hdr_len / 4) | - LSO_TCPHDR_LEN(tcp_hdr(skb)->doff)); + lso->c.lso_ctrl = htonl(LSO_OPCODE_V(CPL_TX_PKT_LSO) | + LSO_FIRST_SLICE_F | LSO_LAST_SLICE_F | + LSO_IPV6_V(v6) | + LSO_ETHHDR_LEN_V(eth_xtra_len / 4) | + LSO_IPHDR_LEN_V(l3hdr_len / 4) | + LSO_TCPHDR_LEN_V(tcp_hdr(skb)->doff)); lso->c.ipid_ofst = htons(0); lso->c.mss = htons(ssi->gso_size); lso->c.seqno_offset = htonl(0); if (is_t4(adap->params.chip)) lso->c.len = htonl(skb->len); else - lso->c.len = htonl(LSO_T5_XFER_SIZE(skb->len)); + lso->c.len = htonl(LSO_T5_XFER_SIZE_V(skb->len)); cpl = (void *)(lso + 1); - cntrl = TXPKT_CSUM_TYPE(v6 ? TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | - TXPKT_IPHDR_LEN(l3hdr_len) | - TXPKT_ETHHDR_LEN(eth_xtra_len); + + if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) + cntrl = TXPKT_ETHHDR_LEN_V(eth_xtra_len); + else + cntrl = T6_TXPKT_ETHHDR_LEN_V(eth_xtra_len); + + cntrl |= TXPKT_CSUM_TYPE_V(v6 ? + TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | + TXPKT_IPHDR_LEN_V(l3hdr_len); q->tso++; q->tx_cso += ssi->gso_segs; } else { @@ -1238,23 +1258,25 @@ out_free: dev_kfree_skb_any(skb); FW_WR_IMMDLEN_V(len)); cpl = (void *)(wr + 1); if (skb->ip_summed == CHECKSUM_PARTIAL) { - cntrl = hwcsum(skb) | TXPKT_IPCSUM_DIS; + cntrl = hwcsum(adap->params.chip, skb) | + TXPKT_IPCSUM_DIS_F; q->tx_cso++; } } if (skb_vlan_tag_present(skb)) { q->vlan_ins++; - cntrl |= TXPKT_VLAN_VLD | TXPKT_VLAN(skb_vlan_tag_get(skb)); + cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb)); #ifdef CONFIG_CHELSIO_T4_FCOE if (skb->protocol == htons(ETH_P_FCOE)) - cntrl |= TXPKT_VLAN( + cntrl |= TXPKT_VLAN_V( ((skb->priority & 0x7) << VLAN_PRIO_SHIFT)); #endif /* CONFIG_CHELSIO_T4_FCOE */ } - cpl->ctrl0 = htonl(TXPKT_OPCODE(CPL_TX_PKT_XT) | - TXPKT_INTF(pi->tx_chan) | TXPKT_PF(adap->fn)); + cpl->ctrl0 = htonl(TXPKT_OPCODE_V(CPL_TX_PKT_XT) | + TXPKT_INTF_V(pi->tx_chan) | + TXPKT_PF_V(adap->pf)); cpl->pack = htons(0); cpl->len = htons(skb->len); cpl->ctrl1 = cpu_to_be64(cntrl); @@ -1964,7 +1986,7 @@ static void restore_rx_bufs(const struct pkt_gl *si, struct sge_fl *q, static inline bool is_new_response(const struct rsp_ctrl *r, const struct sge_rspq *q) { - return RSPD_GEN(r->type_gen) == q->gen; + return (r->type_gen >> RSPD_GEN_S) == q->gen; } /** @@ -2011,19 +2033,19 @@ static int process_responses(struct sge_rspq *q, int budget) break; dma_rmb(); - rsp_type = RSPD_TYPE(rc->type_gen); - if (likely(rsp_type == RSP_TYPE_FLBUF)) { + rsp_type = RSPD_TYPE_G(rc->type_gen); + if (likely(rsp_type == RSPD_TYPE_FLBUF_X)) { struct page_frag *fp; struct pkt_gl si; const struct rx_sw_desc *rsd; u32 len = ntohl(rc->pldbuflen_qid), bufsz, frags; - if (len & RSPD_NEWBUF) { + if (len & RSPD_NEWBUF_F) { if (likely(q->offset > 0)) { free_rx_bufs(q->adap, &rxq->fl, 1); q->offset = 0; } - len = RSPD_LEN(len); + len = RSPD_LEN_G(len); } si.tot_len = len; @@ -2058,7 +2080,7 @@ static int process_responses(struct sge_rspq *q, int budget) q->offset += ALIGN(fp->size, s->fl_align); else restore_rx_bufs(&si, &rxq->fl, frags); - } else if (likely(rsp_type == RSP_TYPE_CPL)) { + } else if (likely(rsp_type == RSPD_TYPE_CPL_X)) { ret = q->handler(q, q->cur_desc, NULL); } else { ret = q->handler(q, (const __be64 *)rc, CXGB4_MSG_AN); @@ -2066,7 +2088,7 @@ static int process_responses(struct sge_rspq *q, int budget) if (unlikely(ret)) { /* couldn't process descriptor, back off for recovery */ - q->next_intr_params = QINTR_TIMER_IDX(NOMEM_TMR_IDX); + q->next_intr_params = QINTR_TIMER_IDX_V(NOMEM_TMR_IDX); break; } @@ -2090,7 +2112,7 @@ int cxgb_busy_poll(struct napi_struct *napi) return LL_FLUSH_BUSY; work_done = process_responses(q, 4); - params = QINTR_TIMER_IDX(TIMERREG_COUNTER0_X) | QINTR_CNT_EN; + params = QINTR_TIMER_IDX_V(TIMERREG_COUNTER0_X) | QINTR_CNT_EN_V(1); q->next_intr_params = params; val = CIDXINC_V(work_done) | SEINTARM_V(params); @@ -2137,7 +2159,7 @@ static int napi_rx_handler(struct napi_struct *napi, int budget) int timer_index; napi_complete(napi); - timer_index = QINTR_TIMER_IDX_GET(q->next_intr_params); + timer_index = QINTR_TIMER_IDX_G(q->next_intr_params); if (q->adaptive_rx) { if (work_done > max(timer_pkt_quota[timer_index], @@ -2147,15 +2169,16 @@ static int napi_rx_handler(struct napi_struct *napi, int budget) timer_index = timer_index - 1; timer_index = clamp(timer_index, 0, SGE_TIMERREGS - 1); - q->next_intr_params = QINTR_TIMER_IDX(timer_index) | - V_QINTR_CNT_EN; + q->next_intr_params = + QINTR_TIMER_IDX_V(timer_index) | + QINTR_CNT_EN_V(0); params = q->next_intr_params; } else { params = q->next_intr_params; q->next_intr_params = q->intr_params; } } else - params = QINTR_TIMER_IDX(7); + params = QINTR_TIMER_IDX_V(7); val = CIDXINC_V(work_done) | SEINTARM_V(params); @@ -2203,7 +2226,7 @@ static unsigned int process_intrq(struct adapter *adap) break; dma_rmb(); - if (RSPD_TYPE(rc->type_gen) == RSP_TYPE_INTR) { + if (RSPD_TYPE_G(rc->type_gen) == RSPD_TYPE_INTR_X) { unsigned int qid = ntohl(rc->pldbuflen_qid); qid -= adap->sge.ingr_start; @@ -2279,7 +2302,7 @@ irq_handler_t t4_intr_handler(struct adapter *adap) static void sge_rx_timer_cb(unsigned long data) { unsigned long m; - unsigned int i, idma_same_state_cnt[2]; + unsigned int i; struct adapter *adap = (struct adapter *)data; struct sge *s = &adap->sge; @@ -2300,67 +2323,16 @@ static void sge_rx_timer_cb(unsigned long data) set_bit(id, s->starving_fl); } } + /* The remainder of the SGE RX Timer Callback routine is dedicated to + * global Master PF activities like checking for chip ingress stalls, + * etc. + */ + if (!(adap->flags & MASTER_PF)) + goto done; - t4_write_reg(adap, SGE_DEBUG_INDEX_A, 13); - idma_same_state_cnt[0] = t4_read_reg(adap, SGE_DEBUG_DATA_HIGH_A); - idma_same_state_cnt[1] = t4_read_reg(adap, SGE_DEBUG_DATA_LOW_A); - - for (i = 0; i < 2; i++) { - u32 debug0, debug11; - - /* If the Ingress DMA Same State Counter ("timer") is less - * than 1s, then we can reset our synthesized Stall Timer and - * continue. If we have previously emitted warnings about a - * potential stalled Ingress Queue, issue a note indicating - * that the Ingress Queue has resumed forward progress. - */ - if (idma_same_state_cnt[i] < s->idma_1s_thresh) { - if (s->idma_stalled[i] >= SGE_IDMA_WARN_THRESH) - CH_WARN(adap, "SGE idma%d, queue%u,resumed after %d sec\n", - i, s->idma_qid[i], - s->idma_stalled[i]/HZ); - s->idma_stalled[i] = 0; - continue; - } - - /* Synthesize an SGE Ingress DMA Same State Timer in the Hz - * domain. The first time we get here it'll be because we - * passed the 1s Threshold; each additional time it'll be - * because the RX Timer Callback is being fired on its regular - * schedule. - * - * If the stall is below our Potential Hung Ingress Queue - * Warning Threshold, continue. - */ - if (s->idma_stalled[i] == 0) - s->idma_stalled[i] = HZ; - else - s->idma_stalled[i] += RX_QCHECK_PERIOD; - - if (s->idma_stalled[i] < SGE_IDMA_WARN_THRESH) - continue; - - /* We'll issue a warning every SGE_IDMA_WARN_REPEAT Hz */ - if (((s->idma_stalled[i] - HZ) % SGE_IDMA_WARN_REPEAT) != 0) - continue; - - /* Read and save the SGE IDMA State and Queue ID information. - * We do this every time in case it changes across time ... - */ - t4_write_reg(adap, SGE_DEBUG_INDEX_A, 0); - debug0 = t4_read_reg(adap, SGE_DEBUG_DATA_LOW_A); - s->idma_state[i] = (debug0 >> (i * 9)) & 0x3f; - - t4_write_reg(adap, SGE_DEBUG_INDEX_A, 11); - debug11 = t4_read_reg(adap, SGE_DEBUG_DATA_LOW_A); - s->idma_qid[i] = (debug11 >> (i * 16)) & 0xffff; - - CH_WARN(adap, "SGE idma%u, queue%u, maybe stuck state%u %dsecs (debug0=%#x, debug11=%#x)\n", - i, s->idma_qid[i], s->idma_state[i], - s->idma_stalled[i]/HZ, debug0, debug11); - t4_sge_decode_idma_state(adap, s->idma_state[i]); - } + t4_idma_monitor(adap, &s->idma_monitor, HZ, RX_QCHECK_PERIOD); +done: mod_timer(&s->rx_timer, jiffies + RX_QCHECK_PERIOD); } @@ -2429,7 +2401,7 @@ static void __iomem *bar2_address(struct adapter *adapter, u64 bar2_qoffset; int ret; - ret = cxgb4_t4_bar2_sge_qregs(adapter, qid, qtype, + ret = t4_bar2_sge_qregs(adapter, qid, qtype, &bar2_qoffset, pbar2_qid); if (ret) return NULL; @@ -2437,9 +2409,12 @@ static void __iomem *bar2_address(struct adapter *adapter, return adapter->bar2 + bar2_qoffset; } +/* @intr_idx: MSI/MSI-X vector if >=0, -(absolute qid + 1) if < 0 + * @cong: < 0 -> no congestion feedback, >= 0 -> congestion channel map + */ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, struct net_device *dev, int intr_idx, - struct sge_fl *fl, rspq_handler_t hnd) + struct sge_fl *fl, rspq_handler_t hnd, int cong) { int ret, flsz = 0; struct fw_iq_cmd c; @@ -2457,12 +2432,13 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, memset(&c, 0, sizeof(c)); c.op_to_vfn = htonl(FW_CMD_OP_V(FW_IQ_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | FW_CMD_EXEC_F | - FW_IQ_CMD_PFN_V(adap->fn) | FW_IQ_CMD_VFN_V(0)); + FW_IQ_CMD_PFN_V(adap->pf) | FW_IQ_CMD_VFN_V(0)); c.alloc_to_len16 = htonl(FW_IQ_CMD_ALLOC_F | FW_IQ_CMD_IQSTART_F | FW_LEN16(c)); c.type_to_iqandstindex = htonl(FW_IQ_CMD_TYPE_V(FW_IQ_TYPE_FL_INT_CAP) | FW_IQ_CMD_IQASYNCH_V(fwevtq) | FW_IQ_CMD_VIID_V(pi->viid) | - FW_IQ_CMD_IQANDST_V(intr_idx < 0) | FW_IQ_CMD_IQANUD_V(1) | + FW_IQ_CMD_IQANDST_V(intr_idx < 0) | + FW_IQ_CMD_IQANUD_V(UPDATEDELIVERY_INTERRUPT_X) | FW_IQ_CMD_IQANDSTINDEX_V(intr_idx >= 0 ? intr_idx : -intr_idx - 1)); c.iqdroprss_to_iqesize = htons(FW_IQ_CMD_IQPCIECH_V(pi->tx_chan) | @@ -2471,8 +2447,21 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, FW_IQ_CMD_IQESIZE_V(ilog2(iq->iqe_len) - 4)); c.iqsize = htons(iq->size); c.iqaddr = cpu_to_be64(iq->phys_addr); + if (cong >= 0) + c.iqns_to_fl0congen = htonl(FW_IQ_CMD_IQFLINTCONGEN_F); if (fl) { + enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip); + + /* Allocate the ring for the hardware free list (with space + * for its status page) along with the associated software + * descriptor ring. The free list size needs to be a multiple + * of the Egress Queue Unit and at least 2 Egress Units larger + * than the SGE's Egress Congrestion Threshold + * (fl_starve_thres - 1). + */ + if (fl->size < s->fl_starve_thres - 1 + 2 * 8) + fl->size = s->fl_starve_thres - 1 + 2 * 8; fl->size = roundup(fl->size, 8); fl->desc = alloc_ring(adap->pdev_dev, fl->size, sizeof(__be64), sizeof(struct rx_sw_desc), &fl->addr, @@ -2481,17 +2470,25 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, goto fl_nomem; flsz = fl->size / 8 + s->stat_len / sizeof(struct tx_desc); - c.iqns_to_fl0congen = htonl(FW_IQ_CMD_FL0PACKEN_F | - FW_IQ_CMD_FL0FETCHRO_F | - FW_IQ_CMD_FL0DATARO_F | - FW_IQ_CMD_FL0PADEN_F); - c.fl0dcaen_to_fl0cidxfthresh = htons(FW_IQ_CMD_FL0FBMIN_V(2) | - FW_IQ_CMD_FL0FBMAX_V(3)); + c.iqns_to_fl0congen |= htonl(FW_IQ_CMD_FL0PACKEN_F | + FW_IQ_CMD_FL0FETCHRO_F | + FW_IQ_CMD_FL0DATARO_F | + FW_IQ_CMD_FL0PADEN_F); + if (cong >= 0) + c.iqns_to_fl0congen |= + htonl(FW_IQ_CMD_FL0CNGCHMAP_V(cong) | + FW_IQ_CMD_FL0CONGCIF_F | + FW_IQ_CMD_FL0CONGEN_F); + c.fl0dcaen_to_fl0cidxfthresh = + htons(FW_IQ_CMD_FL0FBMIN_V(FETCHBURSTMIN_64B_X) | + FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ? + FETCHBURSTMAX_512B_X : + FETCHBURSTMAX_256B_X)); c.fl0size = htons(flsz); c.fl0addr = cpu_to_be64(fl->addr); } - ret = t4_wr_mbox(adap, adap->fn, &c, sizeof(c), &c); + ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c); if (ret) goto err; @@ -2532,6 +2529,41 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, &fl->bar2_qid); refill_fl(adap, fl, fl_cap(fl), GFP_KERNEL); } + + /* For T5 and later we attempt to set up the Congestion Manager values + * of the new RX Ethernet Queue. This should really be handled by + * firmware because it's more complex than any host driver wants to + * get involved with and it's different per chip and this is almost + * certainly wrong. Firmware would be wrong as well, but it would be + * a lot easier to fix in one place ... For now we do something very + * simple (and hopefully less wrong). + */ + if (!is_t4(adap->params.chip) && cong >= 0) { + u32 param, val; + int i; + + param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DMAQ) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DMAQ_CONM_CTXT) | + FW_PARAMS_PARAM_YZ_V(iq->cntxt_id)); + if (cong == 0) { + val = CONMCTXT_CNGTPMODE_V(CONMCTXT_CNGTPMODE_QUEUE_X); + } else { + val = + CONMCTXT_CNGTPMODE_V(CONMCTXT_CNGTPMODE_CHANNEL_X); + for (i = 0; i < 4; i++) { + if (cong & (1 << i)) + val |= + CONMCTXT_CNGCHMAP_V(1 << (i << 2)); + } + } + ret = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, + ¶m, &val); + if (ret) + dev_warn(adap->pdev_dev, "Failed to set Congestion" + " Manager Context for Ingress Queue %d: %d\n", + iq->cntxt_id, -ret); + } + return 0; fl_nomem: @@ -2589,23 +2621,24 @@ int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq, memset(&c, 0, sizeof(c)); c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_ETH_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | FW_CMD_EXEC_F | - FW_EQ_ETH_CMD_PFN_V(adap->fn) | + FW_EQ_ETH_CMD_PFN_V(adap->pf) | FW_EQ_ETH_CMD_VFN_V(0)); c.alloc_to_len16 = htonl(FW_EQ_ETH_CMD_ALLOC_F | FW_EQ_ETH_CMD_EQSTART_F | FW_LEN16(c)); c.viid_pkd = htonl(FW_EQ_ETH_CMD_AUTOEQUEQE_F | FW_EQ_ETH_CMD_VIID_V(pi->viid)); - c.fetchszm_to_iqid = htonl(FW_EQ_ETH_CMD_HOSTFCMODE_V(2) | - FW_EQ_ETH_CMD_PCIECHN_V(pi->tx_chan) | - FW_EQ_ETH_CMD_FETCHRO_V(1) | - FW_EQ_ETH_CMD_IQID_V(iqid)); - c.dcaen_to_eqsize = htonl(FW_EQ_ETH_CMD_FBMIN_V(2) | - FW_EQ_ETH_CMD_FBMAX_V(3) | - FW_EQ_ETH_CMD_CIDXFTHRESH_V(5) | - FW_EQ_ETH_CMD_EQSIZE_V(nentries)); + c.fetchszm_to_iqid = + htonl(FW_EQ_ETH_CMD_HOSTFCMODE_V(HOSTFCMODE_STATUS_PAGE_X) | + FW_EQ_ETH_CMD_PCIECHN_V(pi->tx_chan) | + FW_EQ_ETH_CMD_FETCHRO_F | FW_EQ_ETH_CMD_IQID_V(iqid)); + c.dcaen_to_eqsize = + htonl(FW_EQ_ETH_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) | + FW_EQ_ETH_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) | + FW_EQ_ETH_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) | + FW_EQ_ETH_CMD_EQSIZE_V(nentries)); c.eqaddr = cpu_to_be64(txq->q.phys_addr); - ret = t4_wr_mbox(adap, adap->fn, &c, sizeof(c), &c); + ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c); if (ret) { kfree(txq->q.sdesc); txq->q.sdesc = NULL; @@ -2637,29 +2670,30 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq, txq->q.desc = alloc_ring(adap->pdev_dev, nentries, sizeof(struct tx_desc), 0, &txq->q.phys_addr, - NULL, 0, NUMA_NO_NODE); + NULL, 0, dev_to_node(adap->pdev_dev)); if (!txq->q.desc) return -ENOMEM; c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_CTRL_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | FW_CMD_EXEC_F | - FW_EQ_CTRL_CMD_PFN_V(adap->fn) | + FW_EQ_CTRL_CMD_PFN_V(adap->pf) | FW_EQ_CTRL_CMD_VFN_V(0)); c.alloc_to_len16 = htonl(FW_EQ_CTRL_CMD_ALLOC_F | FW_EQ_CTRL_CMD_EQSTART_F | FW_LEN16(c)); c.cmpliqid_eqid = htonl(FW_EQ_CTRL_CMD_CMPLIQID_V(cmplqid)); c.physeqid_pkd = htonl(0); - c.fetchszm_to_iqid = htonl(FW_EQ_CTRL_CMD_HOSTFCMODE_V(2) | - FW_EQ_CTRL_CMD_PCIECHN_V(pi->tx_chan) | - FW_EQ_CTRL_CMD_FETCHRO_F | - FW_EQ_CTRL_CMD_IQID_V(iqid)); - c.dcaen_to_eqsize = htonl(FW_EQ_CTRL_CMD_FBMIN_V(2) | - FW_EQ_CTRL_CMD_FBMAX_V(3) | - FW_EQ_CTRL_CMD_CIDXFTHRESH_V(5) | - FW_EQ_CTRL_CMD_EQSIZE_V(nentries)); + c.fetchszm_to_iqid = + htonl(FW_EQ_CTRL_CMD_HOSTFCMODE_V(HOSTFCMODE_STATUS_PAGE_X) | + FW_EQ_CTRL_CMD_PCIECHN_V(pi->tx_chan) | + FW_EQ_CTRL_CMD_FETCHRO_F | FW_EQ_CTRL_CMD_IQID_V(iqid)); + c.dcaen_to_eqsize = + htonl(FW_EQ_CTRL_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) | + FW_EQ_CTRL_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) | + FW_EQ_CTRL_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) | + FW_EQ_CTRL_CMD_EQSIZE_V(nentries)); c.eqaddr = cpu_to_be64(txq->q.phys_addr); - ret = t4_wr_mbox(adap, adap->fn, &c, sizeof(c), &c); + ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c); if (ret) { dma_free_coherent(adap->pdev_dev, nentries * sizeof(struct tx_desc), @@ -2697,21 +2731,22 @@ int t4_sge_alloc_ofld_txq(struct adapter *adap, struct sge_ofld_txq *txq, memset(&c, 0, sizeof(c)); c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_OFLD_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | FW_CMD_EXEC_F | - FW_EQ_OFLD_CMD_PFN_V(adap->fn) | + FW_EQ_OFLD_CMD_PFN_V(adap->pf) | FW_EQ_OFLD_CMD_VFN_V(0)); c.alloc_to_len16 = htonl(FW_EQ_OFLD_CMD_ALLOC_F | FW_EQ_OFLD_CMD_EQSTART_F | FW_LEN16(c)); - c.fetchszm_to_iqid = htonl(FW_EQ_OFLD_CMD_HOSTFCMODE_V(2) | - FW_EQ_OFLD_CMD_PCIECHN_V(pi->tx_chan) | - FW_EQ_OFLD_CMD_FETCHRO_F | - FW_EQ_OFLD_CMD_IQID_V(iqid)); - c.dcaen_to_eqsize = htonl(FW_EQ_OFLD_CMD_FBMIN_V(2) | - FW_EQ_OFLD_CMD_FBMAX_V(3) | - FW_EQ_OFLD_CMD_CIDXFTHRESH_V(5) | - FW_EQ_OFLD_CMD_EQSIZE_V(nentries)); + c.fetchszm_to_iqid = + htonl(FW_EQ_OFLD_CMD_HOSTFCMODE_V(HOSTFCMODE_STATUS_PAGE_X) | + FW_EQ_OFLD_CMD_PCIECHN_V(pi->tx_chan) | + FW_EQ_OFLD_CMD_FETCHRO_F | FW_EQ_OFLD_CMD_IQID_V(iqid)); + c.dcaen_to_eqsize = + htonl(FW_EQ_OFLD_CMD_FBMIN_V(FETCHBURSTMIN_64B_X) | + FW_EQ_OFLD_CMD_FBMAX_V(FETCHBURSTMAX_512B_X) | + FW_EQ_OFLD_CMD_CIDXFTHRESH_V(CIDXFLUSHTHRESH_32_X) | + FW_EQ_OFLD_CMD_EQSIZE_V(nentries)); c.eqaddr = cpu_to_be64(txq->q.phys_addr); - ret = t4_wr_mbox(adap, adap->fn, &c, sizeof(c), &c); + ret = t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), &c); if (ret) { kfree(txq->q.sdesc); txq->q.sdesc = NULL; @@ -2750,7 +2785,7 @@ static void free_rspq_fl(struct adapter *adap, struct sge_rspq *rq, unsigned int fl_id = fl ? fl->cntxt_id : 0xffff; adap->sge.ingr_map[rq->cntxt_id - adap->sge.ingr_start] = NULL; - t4_iq_free(adap, adap->fn, adap->fn, 0, FW_IQ_TYPE_FL_INT_CAP, + t4_iq_free(adap, adap->mbox, adap->pf, 0, FW_IQ_TYPE_FL_INT_CAP, rq->cntxt_id, fl_id, 0xffff); dma_free_coherent(adap->pdev_dev, (rq->size + 1) * rq->iqe_len, rq->desc, rq->phys_addr); @@ -2805,7 +2840,7 @@ void t4_free_sge_resources(struct adapter *adap) free_rspq_fl(adap, &eq->rspq, eq->fl.size ? &eq->fl : NULL); if (etq->q.desc) { - t4_eth_eq_free(adap, adap->fn, adap->fn, 0, + t4_eth_eq_free(adap, adap->mbox, adap->pf, 0, etq->q.cntxt_id); free_tx_desc(adap, &etq->q, etq->q.in_use, true); kfree(etq->q.sdesc); @@ -2824,7 +2859,7 @@ void t4_free_sge_resources(struct adapter *adap) if (q->q.desc) { tasklet_kill(&q->qresume_tsk); - t4_ofld_eq_free(adap, adap->fn, adap->fn, 0, + t4_ofld_eq_free(adap, adap->mbox, adap->pf, 0, q->q.cntxt_id); free_tx_desc(adap, &q->q, q->q.in_use, false); kfree(q->q.sdesc); @@ -2839,7 +2874,7 @@ void t4_free_sge_resources(struct adapter *adap) if (cq->q.desc) { tasklet_kill(&cq->qresume_tsk); - t4_ctrl_eq_free(adap, adap->fn, adap->fn, 0, + t4_ctrl_eq_free(adap, adap->mbox, adap->pf, 0, cq->q.cntxt_id); __skb_queue_purge(&cq->sendq); free_txq(adap, &cq->q); @@ -3023,7 +3058,11 @@ int t4_sge_init(struct adapter *adap) * Packing Boundary. T5 introduced the ability to specify these * separately. The actual Ingress Packet Data alignment boundary * within Packed Buffer Mode is the maximum of these two - * specifications. + * specifications. (Note that it makes no real practical sense to + * have the Pading Boudary be larger than the Packing Boundary but you + * could set the chip up that way and, in fact, legacy T4 code would + * end doing this because it would initialize the Padding Boundary and + * leave the Packing Boundary initialized to 0 (16 bytes).) */ ingpadboundary = 1 << (INGPADBOUNDARY_G(sge_control) + INGPADBOUNDARY_SHIFT_X); @@ -3067,11 +3106,14 @@ int t4_sge_init(struct adapter *adap) egress_threshold = EGRTHRESHOLDPACKING_G(sge_conm_ctrl); s->fl_starve_thres = 2*egress_threshold + 1; + t4_idma_monitor_init(adap, &s->idma_monitor); + + /* Set up timers used for recuring callbacks to process RX and TX + * administrative tasks. + */ setup_timer(&s->rx_timer, sge_rx_timer_cb, (unsigned long)adap); setup_timer(&s->tx_timer, sge_tx_timer_cb, (unsigned long)adap); - s->idma_1s_thresh = core_ticks_per_usec(adap) * 1000000; /* 1 s */ - s->idma_stalled[0] = 0; - s->idma_stalled[1] = 0; + spin_lock_init(&s->intrq_lock); return 0; diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index e8578a742f2a..35a44db29347 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -150,7 +150,12 @@ void t4_write_indirect(struct adapter *adap, unsigned int addr_reg, */ void t4_hw_pci_read_cfg4(struct adapter *adap, int reg, u32 *val) { - u32 req = ENABLE_F | FUNCTION_V(adap->fn) | REGISTER_V(reg); + u32 req = FUNCTION_V(adap->pf) | REGISTER_V(reg); + + if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) + req |= ENABLE_F; + else + req |= T6_ENABLE_F; if (is_t4(adap->params.chip)) req |= LOCALCFG_F; @@ -214,8 +219,8 @@ static void fw_asrt(struct adapter *adap, u32 mbox_addr) get_mbox_rpl(adap, (__be64 *)&asrt, sizeof(asrt) / 8, mbox_addr); dev_alert(adap->pdev_dev, "FW assertion at %.16s:%u, val0 %#x, val1 %#x\n", - asrt.u.assert.filename_0_7, ntohl(asrt.u.assert.line), - ntohl(asrt.u.assert.x), ntohl(asrt.u.assert.y)); + asrt.u.assert.filename_0_7, be32_to_cpu(asrt.u.assert.line), + be32_to_cpu(asrt.u.assert.x), be32_to_cpu(asrt.u.assert.y)); } static void dump_mbox(struct adapter *adap, int mbox, u32 data_reg) @@ -233,13 +238,14 @@ static void dump_mbox(struct adapter *adap, int mbox, u32 data_reg) } /** - * t4_wr_mbox_meat - send a command to FW through the given mailbox + * t4_wr_mbox_meat_timeout - send a command to FW through the given mailbox * @adap: the adapter * @mbox: index of the mailbox to use * @cmd: the command to write * @size: command length in bytes * @rpl: where to optionally store the reply * @sleep_ok: if true we may sleep while awaiting command completion + * @timeout: time to wait for command to finish before timing out * * Sends the given command to FW through the selected mailbox and waits * for the FW to execute the command. If @rpl is not %NULL it is used to @@ -254,8 +260,8 @@ static void dump_mbox(struct adapter *adap, int mbox, u32 data_reg) * command or FW executes it but signals an error. In the latter case * the return value is the error code indicated by FW (negated). */ -int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, - void *rpl, bool sleep_ok) +int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd, + int size, void *rpl, bool sleep_ok, int timeout) { static const int delay[] = { 1, 1, 3, 5, 10, 10, 20, 50, 100, 200 @@ -294,7 +300,7 @@ int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, delay_idx = 0; ms = delay[0]; - for (i = 0; i < FW_CMD_MAX_TIMEOUT; i += ms) { + for (i = 0; i < timeout; i += ms) { if (sleep_ok) { ms = delay[delay_idx]; /* last element may repeat */ if (delay_idx < ARRAY_SIZE(delay) - 1) @@ -332,114 +338,11 @@ int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, return -ETIMEDOUT; } -/** - * t4_mc_read - read from MC through backdoor accesses - * @adap: the adapter - * @addr: address of first byte requested - * @idx: which MC to access - * @data: 64 bytes of data containing the requested address - * @ecc: where to store the corresponding 64-bit ECC word - * - * Read 64 bytes of data from MC starting at a 64-byte-aligned address - * that covers the requested address @addr. If @parity is not %NULL it - * is assigned the 64-bit ECC word for the read data. - */ -int t4_mc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, u64 *ecc) -{ - int i; - u32 mc_bist_cmd, mc_bist_cmd_addr, mc_bist_cmd_len; - u32 mc_bist_status_rdata, mc_bist_data_pattern; - - if (is_t4(adap->params.chip)) { - mc_bist_cmd = MC_BIST_CMD_A; - mc_bist_cmd_addr = MC_BIST_CMD_ADDR_A; - mc_bist_cmd_len = MC_BIST_CMD_LEN_A; - mc_bist_status_rdata = MC_BIST_STATUS_RDATA_A; - mc_bist_data_pattern = MC_BIST_DATA_PATTERN_A; - } else { - mc_bist_cmd = MC_REG(MC_P_BIST_CMD_A, idx); - mc_bist_cmd_addr = MC_REG(MC_P_BIST_CMD_ADDR_A, idx); - mc_bist_cmd_len = MC_REG(MC_P_BIST_CMD_LEN_A, idx); - mc_bist_status_rdata = MC_REG(MC_P_BIST_STATUS_RDATA_A, idx); - mc_bist_data_pattern = MC_REG(MC_P_BIST_DATA_PATTERN_A, idx); - } - - if (t4_read_reg(adap, mc_bist_cmd) & START_BIST_F) - return -EBUSY; - t4_write_reg(adap, mc_bist_cmd_addr, addr & ~0x3fU); - t4_write_reg(adap, mc_bist_cmd_len, 64); - t4_write_reg(adap, mc_bist_data_pattern, 0xc); - t4_write_reg(adap, mc_bist_cmd, BIST_OPCODE_V(1) | START_BIST_F | - BIST_CMD_GAP_V(1)); - i = t4_wait_op_done(adap, mc_bist_cmd, START_BIST_F, 0, 10, 1); - if (i) - return i; - -#define MC_DATA(i) MC_BIST_STATUS_REG(mc_bist_status_rdata, i) - - for (i = 15; i >= 0; i--) - *data++ = htonl(t4_read_reg(adap, MC_DATA(i))); - if (ecc) - *ecc = t4_read_reg64(adap, MC_DATA(16)); -#undef MC_DATA - return 0; -} - -/** - * t4_edc_read - read from EDC through backdoor accesses - * @adap: the adapter - * @idx: which EDC to access - * @addr: address of first byte requested - * @data: 64 bytes of data containing the requested address - * @ecc: where to store the corresponding 64-bit ECC word - * - * Read 64 bytes of data from EDC starting at a 64-byte-aligned address - * that covers the requested address @addr. If @parity is not %NULL it - * is assigned the 64-bit ECC word for the read data. - */ -int t4_edc_read(struct adapter *adap, int idx, u32 addr, __be32 *data, u64 *ecc) +int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size, + void *rpl, bool sleep_ok) { - int i; - u32 edc_bist_cmd, edc_bist_cmd_addr, edc_bist_cmd_len; - u32 edc_bist_cmd_data_pattern, edc_bist_status_rdata; - - if (is_t4(adap->params.chip)) { - edc_bist_cmd = EDC_REG(EDC_BIST_CMD_A, idx); - edc_bist_cmd_addr = EDC_REG(EDC_BIST_CMD_ADDR_A, idx); - edc_bist_cmd_len = EDC_REG(EDC_BIST_CMD_LEN_A, idx); - edc_bist_cmd_data_pattern = EDC_REG(EDC_BIST_DATA_PATTERN_A, - idx); - edc_bist_status_rdata = EDC_REG(EDC_BIST_STATUS_RDATA_A, - idx); - } else { - edc_bist_cmd = EDC_REG_T5(EDC_H_BIST_CMD_A, idx); - edc_bist_cmd_addr = EDC_REG_T5(EDC_H_BIST_CMD_ADDR_A, idx); - edc_bist_cmd_len = EDC_REG_T5(EDC_H_BIST_CMD_LEN_A, idx); - edc_bist_cmd_data_pattern = - EDC_REG_T5(EDC_H_BIST_DATA_PATTERN_A, idx); - edc_bist_status_rdata = - EDC_REG_T5(EDC_H_BIST_STATUS_RDATA_A, idx); - } - - if (t4_read_reg(adap, edc_bist_cmd) & START_BIST_F) - return -EBUSY; - t4_write_reg(adap, edc_bist_cmd_addr, addr & ~0x3fU); - t4_write_reg(adap, edc_bist_cmd_len, 64); - t4_write_reg(adap, edc_bist_cmd_data_pattern, 0xc); - t4_write_reg(adap, edc_bist_cmd, - BIST_OPCODE_V(1) | BIST_CMD_GAP_V(1) | START_BIST_F); - i = t4_wait_op_done(adap, edc_bist_cmd, START_BIST_F, 0, 10, 1); - if (i) - return i; - -#define EDC_DATA(i) (EDC_BIST_STATUS_REG(edc_bist_status_rdata, i)) - - for (i = 15; i >= 0; i--) - *data++ = htonl(t4_read_reg(adap, EDC_DATA(i))); - if (ecc) - *ecc = t4_read_reg64(adap, EDC_DATA(16)); -#undef EDC_DATA - return 0; + return t4_wr_mbox_meat_timeout(adap, mbox, cmd, size, rpl, sleep_ok, + FW_CMD_MAX_TIMEOUT); } /** @@ -483,9 +386,8 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, /* Offset into the region of memory which is being accessed * MEM_EDC0 = 0 * MEM_EDC1 = 1 - * MEM_MC = 2 -- T4 - * MEM_MC0 = 2 -- For T5 - * MEM_MC1 = 3 -- For T5 + * MEM_MC = 2 -- MEM_MC for chips with only 1 memory controller + * MEM_MC1 = 3 -- for chips with 2 memory controllers (e.g. T5) */ edc_size = EDRAM0_SIZE_G(t4_read_reg(adap, MA_EDRAM0_BAR_A)); if (mtype != MEM_MC1) @@ -514,7 +416,7 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, mem_base = PCIEOFST_G(mem_reg) << PCIEOFST_SHIFT_X; if (is_t4(adap->params.chip)) mem_base -= adap->t4_bar0; - win_pf = is_t4(adap->params.chip) ? 0 : PFNUM_V(adap->fn); + win_pf = is_t4(adap->params.chip) ? 0 : PFNUM_V(adap->pf); /* Calculate our initial PCI-E Memory Window Position and Offset into * that Window. @@ -625,6 +527,102 @@ int t4_memory_rw(struct adapter *adap, int win, int mtype, u32 addr, return 0; } +/* Return the specified PCI-E Configuration Space register from our Physical + * Function. We try first via a Firmware LDST Command since we prefer to let + * the firmware own all of these registers, but if that fails we go for it + * directly ourselves. + */ +u32 t4_read_pcie_cfg4(struct adapter *adap, int reg) +{ + u32 val, ldst_addrspace; + + /* If fw_attach != 0, construct and send the Firmware LDST Command to + * retrieve the specified PCI-E Configuration Space register. + */ + struct fw_ldst_cmd ldst_cmd; + int ret; + + memset(&ldst_cmd, 0, sizeof(ldst_cmd)); + ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_FUNC_PCIE); + ldst_cmd.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_READ_F | + ldst_addrspace); + ldst_cmd.cycles_to_len16 = cpu_to_be32(FW_LEN16(ldst_cmd)); + ldst_cmd.u.pcie.select_naccess = FW_LDST_CMD_NACCESS_V(1); + ldst_cmd.u.pcie.ctrl_to_fn = + (FW_LDST_CMD_LC_F | FW_LDST_CMD_FN_V(adap->pf)); + ldst_cmd.u.pcie.r = reg; + + /* If the LDST Command succeeds, return the result, otherwise + * fall through to reading it directly ourselves ... + */ + ret = t4_wr_mbox(adap, adap->mbox, &ldst_cmd, sizeof(ldst_cmd), + &ldst_cmd); + if (ret == 0) + val = be32_to_cpu(ldst_cmd.u.pcie.data[0]); + else + /* Read the desired Configuration Space register via the PCI-E + * Backdoor mechanism. + */ + t4_hw_pci_read_cfg4(adap, reg, &val); + return val; +} + +/* Get the window based on base passed to it. + * Window aperture is currently unhandled, but there is no use case for it + * right now + */ +static u32 t4_get_window(struct adapter *adap, u32 pci_base, u64 pci_mask, + u32 memwin_base) +{ + u32 ret; + + if (is_t4(adap->params.chip)) { + u32 bar0; + + /* Truncation intentional: we only read the bottom 32-bits of + * the 64-bit BAR0/BAR1 ... We use the hardware backdoor + * mechanism to read BAR0 instead of using + * pci_resource_start() because we could be operating from + * within a Virtual Machine which is trapping our accesses to + * our Configuration Space and we need to set up the PCI-E + * Memory Window decoders with the actual addresses which will + * be coming across the PCI-E link. + */ + bar0 = t4_read_pcie_cfg4(adap, pci_base); + bar0 &= pci_mask; + adap->t4_bar0 = bar0; + + ret = bar0 + memwin_base; + } else { + /* For T5, only relative offset inside the PCIe BAR is passed */ + ret = memwin_base; + } + return ret; +} + +/* Get the default utility window (win0) used by everyone */ +u32 t4_get_util_window(struct adapter *adap) +{ + return t4_get_window(adap, PCI_BASE_ADDRESS_0, + PCI_BASE_ADDRESS_MEM_MASK, MEMWIN0_BASE); +} + +/* Set up memory window for accessing adapter memory ranges. (Read + * back MA register to ensure that changes propagate before we attempt + * to use the new values.) + */ +void t4_setup_memwin(struct adapter *adap, u32 memwin_base, u32 window) +{ + t4_write_reg(adap, + PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, window), + memwin_base | BIR_V(0) | + WINDOW_V(ilog2(MEMWIN0_APERTURE) - WINDOW_SHIFT_X)); + t4_read_reg(adap, + PCIE_MEM_ACCESS_REG(PCIE_MEM_ACCESS_BASE_WIN_A, window)); +} + /** * t4_get_regs_len - return the size of the chips register set * @adapter: the adapter @@ -640,6 +638,7 @@ unsigned int t4_get_regs_len(struct adapter *adapter) return T4_REGMAP_SIZE; case CHELSIO_T5: + case CHELSIO_T6: return T5_REGMAP_SIZE; } @@ -666,7 +665,8 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x11fc, 0x123c, 0x1300, 0x173c, 0x1800, 0x18fc, - 0x3000, 0x30d8, + 0x3000, 0x305c, + 0x3068, 0x30d8, 0x30e0, 0x5924, 0x5960, 0x59d4, 0x5a00, 0x5af8, @@ -729,7 +729,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x19238, 0x1924c, 0x193f8, 0x19474, 0x19490, 0x194f8, - 0x19800, 0x19f30, + 0x19800, 0x19f4c, 0x1a000, 0x1a06c, 0x1a0b0, 0x1a120, 0x1a128, 0x1a138, @@ -878,7 +878,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x27780, 0x2778c, 0x27800, 0x27c38, 0x27c80, 0x27d7c, - 0x27e00, 0x27e04 + 0x27e00, 0x27e04, }; static const unsigned int t5_reg_ranges[] = { @@ -888,7 +888,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x1280, 0x173c, 0x1800, 0x18fc, 0x3000, 0x3028, - 0x3060, 0x30d8, + 0x3068, 0x30d8, 0x30e0, 0x30fc, 0x3140, 0x357c, 0x35a8, 0x35cc, @@ -900,7 +900,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x5940, 0x59dc, 0x59fc, 0x5a18, 0x5a60, 0x5a9c, - 0x5b9c, 0x5bfc, + 0x5b94, 0x5bfc, 0x6000, 0x6040, 0x6058, 0x614c, 0x7700, 0x7798, @@ -1014,27 +1014,30 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x30800, 0x30834, 0x308c0, 0x30908, 0x30910, 0x309ac, - 0x30a00, 0x30a04, - 0x30a0c, 0x30a2c, + 0x30a00, 0x30a2c, 0x30a44, 0x30a50, 0x30a74, 0x30c24, + 0x30d00, 0x30d00, 0x30d08, 0x30d14, 0x30d1c, 0x30d20, 0x30d3c, 0x30d50, 0x31200, 0x3120c, 0x31220, 0x31220, 0x31240, 0x31240, - 0x31600, 0x31600, - 0x31608, 0x3160c, + 0x31600, 0x3160c, 0x31a00, 0x31a1c, - 0x31e04, 0x31e20, + 0x31e00, 0x31e20, 0x31e38, 0x31e3c, 0x31e80, 0x31e80, 0x31e88, 0x31ea8, 0x31eb0, 0x31eb4, 0x31ec8, 0x31ed4, 0x31fb8, 0x32004, - 0x32208, 0x3223c, + 0x32200, 0x32200, + 0x32208, 0x32240, + 0x32248, 0x32280, + 0x32288, 0x322c0, + 0x322c8, 0x322fc, 0x32600, 0x32630, 0x32a00, 0x32abc, 0x32b00, 0x32b70, @@ -1074,27 +1077,30 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x34800, 0x34834, 0x348c0, 0x34908, 0x34910, 0x349ac, - 0x34a00, 0x34a04, - 0x34a0c, 0x34a2c, + 0x34a00, 0x34a2c, 0x34a44, 0x34a50, 0x34a74, 0x34c24, + 0x34d00, 0x34d00, 0x34d08, 0x34d14, 0x34d1c, 0x34d20, 0x34d3c, 0x34d50, 0x35200, 0x3520c, 0x35220, 0x35220, 0x35240, 0x35240, - 0x35600, 0x35600, - 0x35608, 0x3560c, + 0x35600, 0x3560c, 0x35a00, 0x35a1c, - 0x35e04, 0x35e20, + 0x35e00, 0x35e20, 0x35e38, 0x35e3c, 0x35e80, 0x35e80, 0x35e88, 0x35ea8, 0x35eb0, 0x35eb4, 0x35ec8, 0x35ed4, 0x35fb8, 0x36004, - 0x36208, 0x3623c, + 0x36200, 0x36200, + 0x36208, 0x36240, + 0x36248, 0x36280, + 0x36288, 0x362c0, + 0x362c8, 0x362fc, 0x36600, 0x36630, 0x36a00, 0x36abc, 0x36b00, 0x36b70, @@ -1134,27 +1140,30 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x38800, 0x38834, 0x388c0, 0x38908, 0x38910, 0x389ac, - 0x38a00, 0x38a04, - 0x38a0c, 0x38a2c, + 0x38a00, 0x38a2c, 0x38a44, 0x38a50, 0x38a74, 0x38c24, + 0x38d00, 0x38d00, 0x38d08, 0x38d14, 0x38d1c, 0x38d20, 0x38d3c, 0x38d50, 0x39200, 0x3920c, 0x39220, 0x39220, 0x39240, 0x39240, - 0x39600, 0x39600, - 0x39608, 0x3960c, + 0x39600, 0x3960c, 0x39a00, 0x39a1c, - 0x39e04, 0x39e20, + 0x39e00, 0x39e20, 0x39e38, 0x39e3c, 0x39e80, 0x39e80, 0x39e88, 0x39ea8, 0x39eb0, 0x39eb4, 0x39ec8, 0x39ed4, 0x39fb8, 0x3a004, - 0x3a208, 0x3a23c, + 0x3a200, 0x3a200, + 0x3a208, 0x3a240, + 0x3a248, 0x3a280, + 0x3a288, 0x3a2c0, + 0x3a2c8, 0x3a2fc, 0x3a600, 0x3a630, 0x3aa00, 0x3aabc, 0x3ab00, 0x3ab70, @@ -1194,27 +1203,30 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x3c800, 0x3c834, 0x3c8c0, 0x3c908, 0x3c910, 0x3c9ac, - 0x3ca00, 0x3ca04, - 0x3ca0c, 0x3ca2c, + 0x3ca00, 0x3ca2c, 0x3ca44, 0x3ca50, 0x3ca74, 0x3cc24, + 0x3cd00, 0x3cd00, 0x3cd08, 0x3cd14, 0x3cd1c, 0x3cd20, 0x3cd3c, 0x3cd50, 0x3d200, 0x3d20c, 0x3d220, 0x3d220, 0x3d240, 0x3d240, - 0x3d600, 0x3d600, - 0x3d608, 0x3d60c, + 0x3d600, 0x3d60c, 0x3da00, 0x3da1c, - 0x3de04, 0x3de20, + 0x3de00, 0x3de20, 0x3de38, 0x3de3c, 0x3de80, 0x3de80, 0x3de88, 0x3dea8, 0x3deb0, 0x3deb4, 0x3dec8, 0x3ded4, 0x3dfb8, 0x3e004, - 0x3e208, 0x3e23c, + 0x3e200, 0x3e200, + 0x3e208, 0x3e240, + 0x3e248, 0x3e280, + 0x3e288, 0x3e2c0, + 0x3e2c8, 0x3e2fc, 0x3e600, 0x3e630, 0x3ea00, 0x3eabc, 0x3eb00, 0x3eb70, @@ -1247,7 +1259,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x3fcf0, 0x3fcfc, 0x40000, 0x4000c, 0x40040, 0x40068, - 0x40080, 0x40144, + 0x4007c, 0x40144, 0x40180, 0x4018c, 0x40200, 0x40298, 0x402ac, 0x4033c, @@ -1275,7 +1287,7 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x47800, 0x47814, 0x48000, 0x4800c, 0x48040, 0x48068, - 0x48080, 0x48144, + 0x4807c, 0x48144, 0x48180, 0x4818c, 0x48200, 0x48298, 0x482ac, 0x4833c, @@ -1309,6 +1321,344 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) 0x51300, 0x51308, }; + static const unsigned int t6_reg_ranges[] = { + 0x1008, 0x114c, + 0x1180, 0x11b4, + 0x11fc, 0x1250, + 0x1280, 0x133c, + 0x1800, 0x18fc, + 0x3000, 0x302c, + 0x3060, 0x30d8, + 0x30e0, 0x30fc, + 0x3140, 0x357c, + 0x35a8, 0x35cc, + 0x35ec, 0x35ec, + 0x3600, 0x5624, + 0x56cc, 0x575c, + 0x580c, 0x5814, + 0x5890, 0x58bc, + 0x5940, 0x595c, + 0x5980, 0x598c, + 0x59b0, 0x59dc, + 0x59fc, 0x5a18, + 0x5a60, 0x5a6c, + 0x5a80, 0x5a9c, + 0x5b94, 0x5bfc, + 0x5c10, 0x5ec0, + 0x5ec8, 0x5ec8, + 0x6000, 0x6040, + 0x6058, 0x6154, + 0x7700, 0x7798, + 0x77c0, 0x7880, + 0x78cc, 0x78fc, + 0x7b00, 0x7c54, + 0x7d00, 0x7efc, + 0x8dc0, 0x8de0, + 0x8df8, 0x8e84, + 0x8ea0, 0x8f88, + 0x8fb8, 0x911c, + 0x9400, 0x9470, + 0x9600, 0x971c, + 0x9800, 0x9808, + 0x9820, 0x983c, + 0x9850, 0x9864, + 0x9c00, 0x9c6c, + 0x9c80, 0x9cec, + 0x9d00, 0x9d6c, + 0x9d80, 0x9dec, + 0x9e00, 0x9e6c, + 0x9e80, 0x9eec, + 0x9f00, 0x9f6c, + 0x9f80, 0xa020, + 0xd004, 0xd03c, + 0xdfc0, 0xdfe0, + 0xe000, 0xf008, + 0x11000, 0x11014, + 0x11048, 0x11110, + 0x11118, 0x1117c, + 0x11190, 0x11260, + 0x11300, 0x1130c, + 0x12000, 0x1205c, + 0x19040, 0x1906c, + 0x19078, 0x19080, + 0x1908c, 0x19124, + 0x19150, 0x191b0, + 0x191d0, 0x191e8, + 0x19238, 0x192b8, + 0x193f8, 0x19474, + 0x19490, 0x194cc, + 0x194f0, 0x194f8, + 0x19c00, 0x19c80, + 0x19c94, 0x19cbc, + 0x19ce4, 0x19d28, + 0x19d50, 0x19d78, + 0x19d94, 0x19dc8, + 0x19df0, 0x19e10, + 0x19e50, 0x19e6c, + 0x19ea0, 0x19f34, + 0x19f40, 0x19f50, + 0x19f90, 0x19fac, + 0x19fc4, 0x19fe4, + 0x1a000, 0x1a06c, + 0x1a0b0, 0x1a120, + 0x1a128, 0x1a138, + 0x1a190, 0x1a1c4, + 0x1a1fc, 0x1a1fc, + 0x1e008, 0x1e00c, + 0x1e040, 0x1e04c, + 0x1e284, 0x1e290, + 0x1e2c0, 0x1e2c0, + 0x1e2e0, 0x1e2e0, + 0x1e300, 0x1e384, + 0x1e3c0, 0x1e3c8, + 0x1e408, 0x1e40c, + 0x1e440, 0x1e44c, + 0x1e684, 0x1e690, + 0x1e6c0, 0x1e6c0, + 0x1e6e0, 0x1e6e0, + 0x1e700, 0x1e784, + 0x1e7c0, 0x1e7c8, + 0x1e808, 0x1e80c, + 0x1e840, 0x1e84c, + 0x1ea84, 0x1ea90, + 0x1eac0, 0x1eac0, + 0x1eae0, 0x1eae0, + 0x1eb00, 0x1eb84, + 0x1ebc0, 0x1ebc8, + 0x1ec08, 0x1ec0c, + 0x1ec40, 0x1ec4c, + 0x1ee84, 0x1ee90, + 0x1eec0, 0x1eec0, + 0x1eee0, 0x1eee0, + 0x1ef00, 0x1ef84, + 0x1efc0, 0x1efc8, + 0x1f008, 0x1f00c, + 0x1f040, 0x1f04c, + 0x1f284, 0x1f290, + 0x1f2c0, 0x1f2c0, + 0x1f2e0, 0x1f2e0, + 0x1f300, 0x1f384, + 0x1f3c0, 0x1f3c8, + 0x1f408, 0x1f40c, + 0x1f440, 0x1f44c, + 0x1f684, 0x1f690, + 0x1f6c0, 0x1f6c0, + 0x1f6e0, 0x1f6e0, + 0x1f700, 0x1f784, + 0x1f7c0, 0x1f7c8, + 0x1f808, 0x1f80c, + 0x1f840, 0x1f84c, + 0x1fa84, 0x1fa90, + 0x1fac0, 0x1fac0, + 0x1fae0, 0x1fae0, + 0x1fb00, 0x1fb84, + 0x1fbc0, 0x1fbc8, + 0x1fc08, 0x1fc0c, + 0x1fc40, 0x1fc4c, + 0x1fe84, 0x1fe90, + 0x1fec0, 0x1fec0, + 0x1fee0, 0x1fee0, + 0x1ff00, 0x1ff84, + 0x1ffc0, 0x1ffc8, + 0x30000, 0x30070, + 0x30100, 0x3015c, + 0x30190, 0x301d0, + 0x30200, 0x30318, + 0x30400, 0x3052c, + 0x30540, 0x3061c, + 0x30800, 0x3088c, + 0x308c0, 0x30908, + 0x30910, 0x309b8, + 0x30a00, 0x30a04, + 0x30a0c, 0x30a2c, + 0x30a44, 0x30a50, + 0x30a74, 0x30c24, + 0x30d00, 0x30d3c, + 0x30d44, 0x30d7c, + 0x30de0, 0x30de0, + 0x30e00, 0x30ed4, + 0x30f00, 0x30fa4, + 0x30fc0, 0x30fc4, + 0x31000, 0x31004, + 0x31080, 0x310fc, + 0x31208, 0x31220, + 0x3123c, 0x31254, + 0x31300, 0x31300, + 0x31308, 0x3131c, + 0x31338, 0x3133c, + 0x31380, 0x31380, + 0x31388, 0x313a8, + 0x313b4, 0x313b4, + 0x31400, 0x31420, + 0x31438, 0x3143c, + 0x31480, 0x31480, + 0x314a8, 0x314a8, + 0x314b0, 0x314b4, + 0x314c8, 0x314d4, + 0x31a40, 0x31a4c, + 0x31af0, 0x31b20, + 0x31b38, 0x31b3c, + 0x31b80, 0x31b80, + 0x31ba8, 0x31ba8, + 0x31bb0, 0x31bb4, + 0x31bc8, 0x31bd4, + 0x32140, 0x3218c, + 0x321f0, 0x32200, + 0x32218, 0x32218, + 0x32400, 0x32400, + 0x32408, 0x3241c, + 0x32618, 0x32620, + 0x32664, 0x32664, + 0x326a8, 0x326a8, + 0x326ec, 0x326ec, + 0x32a00, 0x32abc, + 0x32b00, 0x32b78, + 0x32c00, 0x32c00, + 0x32c08, 0x32c3c, + 0x32e00, 0x32e2c, + 0x32f00, 0x32f2c, + 0x33000, 0x330ac, + 0x330c0, 0x331ac, + 0x331c0, 0x332c4, + 0x332e4, 0x333c4, + 0x333e4, 0x334ac, + 0x334c0, 0x335ac, + 0x335c0, 0x336c4, + 0x336e4, 0x337c4, + 0x337e4, 0x337fc, + 0x33814, 0x33814, + 0x33854, 0x33868, + 0x33880, 0x3388c, + 0x338c0, 0x338d0, + 0x338e8, 0x338ec, + 0x33900, 0x339ac, + 0x339c0, 0x33ac4, + 0x33ae4, 0x33b10, + 0x33b24, 0x33b50, + 0x33bf0, 0x33c10, + 0x33c24, 0x33c50, + 0x33cf0, 0x33cfc, + 0x34000, 0x34070, + 0x34100, 0x3415c, + 0x34190, 0x341d0, + 0x34200, 0x34318, + 0x34400, 0x3452c, + 0x34540, 0x3461c, + 0x34800, 0x3488c, + 0x348c0, 0x34908, + 0x34910, 0x349b8, + 0x34a00, 0x34a04, + 0x34a0c, 0x34a2c, + 0x34a44, 0x34a50, + 0x34a74, 0x34c24, + 0x34d00, 0x34d3c, + 0x34d44, 0x34d7c, + 0x34de0, 0x34de0, + 0x34e00, 0x34ed4, + 0x34f00, 0x34fa4, + 0x34fc0, 0x34fc4, + 0x35000, 0x35004, + 0x35080, 0x350fc, + 0x35208, 0x35220, + 0x3523c, 0x35254, + 0x35300, 0x35300, + 0x35308, 0x3531c, + 0x35338, 0x3533c, + 0x35380, 0x35380, + 0x35388, 0x353a8, + 0x353b4, 0x353b4, + 0x35400, 0x35420, + 0x35438, 0x3543c, + 0x35480, 0x35480, + 0x354a8, 0x354a8, + 0x354b0, 0x354b4, + 0x354c8, 0x354d4, + 0x35a40, 0x35a4c, + 0x35af0, 0x35b20, + 0x35b38, 0x35b3c, + 0x35b80, 0x35b80, + 0x35ba8, 0x35ba8, + 0x35bb0, 0x35bb4, + 0x35bc8, 0x35bd4, + 0x36140, 0x3618c, + 0x361f0, 0x36200, + 0x36218, 0x36218, + 0x36400, 0x36400, + 0x36408, 0x3641c, + 0x36618, 0x36620, + 0x36664, 0x36664, + 0x366a8, 0x366a8, + 0x366ec, 0x366ec, + 0x36a00, 0x36abc, + 0x36b00, 0x36b78, + 0x36c00, 0x36c00, + 0x36c08, 0x36c3c, + 0x36e00, 0x36e2c, + 0x36f00, 0x36f2c, + 0x37000, 0x370ac, + 0x370c0, 0x371ac, + 0x371c0, 0x372c4, + 0x372e4, 0x373c4, + 0x373e4, 0x374ac, + 0x374c0, 0x375ac, + 0x375c0, 0x376c4, + 0x376e4, 0x377c4, + 0x377e4, 0x377fc, + 0x37814, 0x37814, + 0x37854, 0x37868, + 0x37880, 0x3788c, + 0x378c0, 0x378d0, + 0x378e8, 0x378ec, + 0x37900, 0x379ac, + 0x379c0, 0x37ac4, + 0x37ae4, 0x37b10, + 0x37b24, 0x37b50, + 0x37bf0, 0x37c10, + 0x37c24, 0x37c50, + 0x37cf0, 0x37cfc, + 0x40040, 0x40040, + 0x40080, 0x40084, + 0x40100, 0x40100, + 0x40140, 0x401bc, + 0x40200, 0x40214, + 0x40228, 0x40228, + 0x40240, 0x40258, + 0x40280, 0x40280, + 0x40304, 0x40304, + 0x40330, 0x4033c, + 0x41304, 0x413dc, + 0x41400, 0x4141c, + 0x41480, 0x414d0, + 0x44000, 0x4407c, + 0x440c0, 0x4427c, + 0x442c0, 0x4447c, + 0x444c0, 0x4467c, + 0x446c0, 0x4487c, + 0x448c0, 0x44a7c, + 0x44ac0, 0x44c7c, + 0x44cc0, 0x44e7c, + 0x44ec0, 0x4507c, + 0x450c0, 0x451fc, + 0x45800, 0x45868, + 0x45880, 0x45884, + 0x458a0, 0x458b0, + 0x45a00, 0x45a68, + 0x45a80, 0x45a84, + 0x45aa0, 0x45ab0, + 0x460c0, 0x460e4, + 0x47000, 0x4708c, + 0x47200, 0x47250, + 0x47400, 0x47420, + 0x47600, 0x47618, + 0x47800, 0x4782c, + 0x50000, 0x500cc, + 0x50400, 0x50400, + 0x50800, 0x508cc, + 0x50c00, 0x50c00, + 0x51000, 0x510b0, + 0x51300, 0x51324, + }; + u32 *buf_end = (u32 *)((char *)buf + buf_size); const unsigned int *reg_ranges; int reg_ranges_size, range; @@ -1328,6 +1678,11 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size) reg_ranges_size = ARRAY_SIZE(t5_reg_ranges); break; + case CHELSIO_T6: + reg_ranges = t6_reg_ranges; + reg_ranges_size = ARRAY_SIZE(t6_reg_ranges); + break; + default: dev_err(adap->pdev_dev, "Unsupported chip version %d\n", chip_version); @@ -1618,7 +1973,7 @@ int t4_read_flash(struct adapter *adapter, unsigned int addr, if (ret) return ret; if (byte_oriented) - *data = (__force __u32) (htonl(*data)); + *data = (__force __u32)(cpu_to_be32(*data)); } return 0; } @@ -1941,7 +2296,8 @@ static bool t4_fw_matches_chip(const struct adapter *adap, * which will keep us "honest" in the future ... */ if ((is_t4(adap->params.chip) && hdr->chip == FW_HDR_CHIP_T4) || - (is_t5(adap->params.chip) && hdr->chip == FW_HDR_CHIP_T5)) + (is_t5(adap->params.chip) && hdr->chip == FW_HDR_CHIP_T5) || + (is_t6(adap->params.chip) && hdr->chip == FW_HDR_CHIP_T6)) return true; dev_err(adap->pdev_dev, @@ -1979,7 +2335,7 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size) "FW image size not multiple of 512 bytes\n"); return -EINVAL; } - if (ntohs(hdr->len512) * 512 != size) { + if ((unsigned int)be16_to_cpu(hdr->len512) * 512 != size) { dev_err(adap->pdev_dev, "FW image size differs from size in FW header\n"); return -EINVAL; @@ -1993,7 +2349,7 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size) return -EINVAL; for (csum = 0, i = 0; i < size / sizeof(csum); i++) - csum += ntohl(p[i]); + csum += be32_to_cpu(p[i]); if (csum != 0xffffffff) { dev_err(adap->pdev_dev, @@ -2012,7 +2368,7 @@ int t4_load_fw(struct adapter *adap, const u8 *fw_data, unsigned int size) * first page with a bad version. */ memcpy(first_page, fw_data, SF_PAGE_SIZE); - ((struct fw_hdr *)first_page)->fw_ver = htonl(0xffffffff); + ((struct fw_hdr *)first_page)->fw_ver = cpu_to_be32(0xffffffff); ret = t4_write_flash(adap, fw_img_start, SF_PAGE_SIZE, first_page); if (ret) goto out; @@ -2039,6 +2395,147 @@ out: } /** + * t4_phy_fw_ver - return current PHY firmware version + * @adap: the adapter + * @phy_fw_ver: return value buffer for PHY firmware version + * + * Returns the current version of external PHY firmware on the + * adapter. + */ +int t4_phy_fw_ver(struct adapter *adap, int *phy_fw_ver) +{ + u32 param, val; + int ret; + + param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PHYFW) | + FW_PARAMS_PARAM_Y_V(adap->params.portvec) | + FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_VERSION)); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, + ¶m, &val); + if (ret < 0) + return ret; + *phy_fw_ver = val; + return 0; +} + +/** + * t4_load_phy_fw - download port PHY firmware + * @adap: the adapter + * @win: the PCI-E Memory Window index to use for t4_memory_rw() + * @win_lock: the lock to use to guard the memory copy + * @phy_fw_version: function to check PHY firmware versions + * @phy_fw_data: the PHY firmware image to write + * @phy_fw_size: image size + * + * Transfer the specified PHY firmware to the adapter. If a non-NULL + * @phy_fw_version is supplied, then it will be used to determine if + * it's necessary to perform the transfer by comparing the version + * of any existing adapter PHY firmware with that of the passed in + * PHY firmware image. If @win_lock is non-NULL then it will be used + * around the call to t4_memory_rw() which transfers the PHY firmware + * to the adapter. + * + * A negative error number will be returned if an error occurs. If + * version number support is available and there's no need to upgrade + * the firmware, 0 will be returned. If firmware is successfully + * transferred to the adapter, 1 will be retured. + * + * NOTE: some adapters only have local RAM to store the PHY firmware. As + * a result, a RESET of the adapter would cause that RAM to lose its + * contents. Thus, loading PHY firmware on such adapters must happen + * after any FW_RESET_CMDs ... + */ +int t4_load_phy_fw(struct adapter *adap, + int win, spinlock_t *win_lock, + int (*phy_fw_version)(const u8 *, size_t), + const u8 *phy_fw_data, size_t phy_fw_size) +{ + unsigned long mtype = 0, maddr = 0; + u32 param, val; + int cur_phy_fw_ver = 0, new_phy_fw_vers = 0; + int ret; + + /* If we have version number support, then check to see if the adapter + * already has up-to-date PHY firmware loaded. + */ + if (phy_fw_version) { + new_phy_fw_vers = phy_fw_version(phy_fw_data, phy_fw_size); + ret = t4_phy_fw_ver(adap, &cur_phy_fw_ver); + if (ret < 0) + return ret; + + if (cur_phy_fw_ver >= new_phy_fw_vers) { + CH_WARN(adap, "PHY Firmware already up-to-date, " + "version %#x\n", cur_phy_fw_ver); + return 0; + } + } + + /* Ask the firmware where it wants us to copy the PHY firmware image. + * The size of the file requires a special version of the READ coommand + * which will pass the file size via the values field in PARAMS_CMD and + * retrieve the return value from firmware and place it in the same + * buffer values + */ + param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PHYFW) | + FW_PARAMS_PARAM_Y_V(adap->params.portvec) | + FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD)); + val = phy_fw_size; + ret = t4_query_params_rw(adap, adap->mbox, adap->pf, 0, 1, + ¶m, &val, 1); + if (ret < 0) + return ret; + mtype = val >> 8; + maddr = (val & 0xff) << 16; + + /* Copy the supplied PHY Firmware image to the adapter memory location + * allocated by the adapter firmware. + */ + if (win_lock) + spin_lock_bh(win_lock); + ret = t4_memory_rw(adap, win, mtype, maddr, + phy_fw_size, (__be32 *)phy_fw_data, + T4_MEMORY_WRITE); + if (win_lock) + spin_unlock_bh(win_lock); + if (ret) + return ret; + + /* Tell the firmware that the PHY firmware image has been written to + * RAM and it can now start copying it over to the PHYs. The chip + * firmware will RESET the affected PHYs as part of this operation + * leaving them running the new PHY firmware image. + */ + param = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PHYFW) | + FW_PARAMS_PARAM_Y_V(adap->params.portvec) | + FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD)); + ret = t4_set_params_timeout(adap, adap->mbox, adap->pf, 0, 1, + ¶m, &val, 30000); + + /* If we have version number support, then check to see that the new + * firmware got loaded properly. + */ + if (phy_fw_version) { + ret = t4_phy_fw_ver(adap, &cur_phy_fw_ver); + if (ret < 0) + return ret; + + if (cur_phy_fw_ver != new_phy_fw_vers) { + CH_WARN(adap, "PHY Firmware did not update: " + "version on adapter %#x, " + "version flashed %#x\n", + cur_phy_fw_ver, new_phy_fw_vers); + return -ENXIO; + } + } + + return 1; +} + +/** * t4_fwcache - firmware cache operation * @adap: the adapter * @op : the operation (flush or flush and invalidate) @@ -2051,7 +2548,7 @@ int t4_fwcache(struct adapter *adap, enum fw_params_param_dev_fwcache op) c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PARAMS_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | - FW_PARAMS_CMD_PFN_V(adap->fn) | + FW_PARAMS_CMD_PFN_V(adap->pf) | FW_PARAMS_CMD_VFN_V(0)); c.retval_len16 = cpu_to_be32(FW_LEN16(c)); c.param[0].mnem = @@ -2107,19 +2604,22 @@ int t4_link_start(struct adapter *adap, unsigned int mbox, unsigned int port, fc |= FW_PORT_CAP_FC_TX; memset(&c, 0, sizeof(c)); - c.op_to_portid = htonl(FW_CMD_OP_V(FW_PORT_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_PORT_CMD_PORTID_V(port)); - c.action_to_len16 = htonl(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) | - FW_LEN16(c)); + c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_PORT_CMD_PORTID_V(port)); + c.action_to_len16 = + cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) | + FW_LEN16(c)); if (!(lc->supported & FW_PORT_CAP_ANEG)) { - c.u.l1cfg.rcap = htonl((lc->supported & ADVERT_MASK) | fc); + c.u.l1cfg.rcap = cpu_to_be32((lc->supported & ADVERT_MASK) | + fc); lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX); } else if (lc->autoneg == AUTONEG_DISABLE) { - c.u.l1cfg.rcap = htonl(lc->requested_speed | fc | mdi); + c.u.l1cfg.rcap = cpu_to_be32(lc->requested_speed | fc | mdi); lc->fc = lc->requested_fc & (PAUSE_RX | PAUSE_TX); } else - c.u.l1cfg.rcap = htonl(lc->advertising | fc | mdi); + c.u.l1cfg.rcap = cpu_to_be32(lc->advertising | fc | mdi); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -2137,11 +2637,13 @@ int t4_restart_aneg(struct adapter *adap, unsigned int mbox, unsigned int port) struct fw_port_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_portid = htonl(FW_CMD_OP_V(FW_PORT_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_PORT_CMD_PORTID_V(port)); - c.action_to_len16 = htonl(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) | - FW_LEN16(c)); - c.u.l1cfg.rcap = htonl(FW_PORT_CAP_ANEG); + c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_PORT_CMD_PORTID_V(port)); + c.action_to_len16 = + cpu_to_be32(FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_L1_CFG) | + FW_LEN16(c)); + c.u.l1cfg.rcap = cpu_to_be32(FW_PORT_CAP_ANEG); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -2335,6 +2837,7 @@ static void tp_intr_handler(struct adapter *adapter) static void sge_intr_handler(struct adapter *adapter) { u64 v; + u32 err; static const struct intr_info sge_intr_info[] = { { ERR_CPL_EXCEED_IQE_SIZE_F, @@ -2343,8 +2846,6 @@ static void sge_intr_handler(struct adapter *adapter) "SGE GTS CIDX increment too large", -1, 0 }, { ERR_CPL_OPCODE_0_F, "SGE received 0-length CPL", -1, 0 }, { DBFIFO_LP_INT_F, NULL, -1, 0, t4_db_full }, - { DBFIFO_HP_INT_F, NULL, -1, 0, t4_db_full }, - { ERR_DROPPED_DB_F, NULL, -1, 0, t4_db_dropped }, { ERR_DATA_CPL_ON_HIGH_QID1_F | ERR_DATA_CPL_ON_HIGH_QID0_F, "SGE IQID > 1023 received CPL for FL", -1, 0 }, { ERR_BAD_DB_PIDX3_F, "SGE DBP 3 pidx increment too large", -1, @@ -2357,13 +2858,19 @@ static void sge_intr_handler(struct adapter *adapter) 0 }, { ERR_ING_CTXT_PRIO_F, "SGE too many priority ingress contexts", -1, 0 }, - { ERR_EGR_CTXT_PRIO_F, - "SGE too many priority egress contexts", -1, 0 }, { INGRESS_SIZE_ERR_F, "SGE illegal ingress QID", -1, 0 }, { EGRESS_SIZE_ERR_F, "SGE illegal egress QID", -1, 0 }, { 0 } }; + static struct intr_info t4t5_sge_intr_info[] = { + { ERR_DROPPED_DB_F, NULL, -1, 0, t4_db_dropped }, + { DBFIFO_HP_INT_F, NULL, -1, 0, t4_db_full }, + { ERR_EGR_CTXT_PRIO_F, + "SGE too many priority egress contexts", -1, 0 }, + { 0 } + }; + v = (u64)t4_read_reg(adapter, SGE_INT_CAUSE1_A) | ((u64)t4_read_reg(adapter, SGE_INT_CAUSE2_A) << 32); if (v) { @@ -2373,8 +2880,23 @@ static void sge_intr_handler(struct adapter *adapter) t4_write_reg(adapter, SGE_INT_CAUSE2_A, v >> 32); } - if (t4_handle_intr_status(adapter, SGE_INT_CAUSE3_A, sge_intr_info) || - v != 0) + v |= t4_handle_intr_status(adapter, SGE_INT_CAUSE3_A, sge_intr_info); + if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) + v |= t4_handle_intr_status(adapter, SGE_INT_CAUSE3_A, + t4t5_sge_intr_info); + + err = t4_read_reg(adapter, SGE_ERROR_STATS_A); + if (err & ERROR_QID_VALID_F) { + dev_err(adapter->pdev_dev, "SGE error for queue %u\n", + ERROR_QID_G(err)); + if (err & UNCAPTURED_ERROR_F) + dev_err(adapter->pdev_dev, + "SGE UNCAPTURED_ERROR set (clearing)\n"); + t4_write_reg(adapter, SGE_ERROR_STATS_A, ERROR_QID_VALID_F | + UNCAPTURED_ERROR_F); + } + + if (v != 0) t4_fatal_err(adapter); } @@ -2547,6 +3069,7 @@ static void cplsw_intr_handler(struct adapter *adapter) */ static void le_intr_handler(struct adapter *adap) { + enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip); static const struct intr_info le_intr_info[] = { { LIPMISS_F, "LE LIP miss", -1, 0 }, { LIP0_F, "LE 0 LIP error", -1, 0 }, @@ -2556,7 +3079,18 @@ static void le_intr_handler(struct adapter *adap) { 0 } }; - if (t4_handle_intr_status(adap, LE_DB_INT_CAUSE_A, le_intr_info)) + static struct intr_info t6_le_intr_info[] = { + { T6_LIPMISS_F, "LE LIP miss", -1, 0 }, + { T6_LIP0_F, "LE 0 LIP error", -1, 0 }, + { TCAMINTPERR_F, "LE parity error", -1, 1 }, + { T6_UNKNOWNCMD_F, "LE unknown command", -1, 1 }, + { SSRAMINTPERR_F, "LE request queue parity error", -1, 1 }, + { 0 } + }; + + if (t4_handle_intr_status(adap, LE_DB_INT_CAUSE_A, + (chip <= CHELSIO_T5) ? + le_intr_info : t6_le_intr_info)) t4_fatal_err(adap); } @@ -2825,7 +3359,7 @@ int t4_slow_intr_handler(struct adapter *adapter) pcie_intr_handler(adapter); if (cause & MC_F) mem_intr_handler(adapter, MEM_MC); - if (!is_t4(adapter->params.chip) && (cause & MC1_S)) + if (is_t5(adapter->params.chip) && (cause & MC1_F)) mem_intr_handler(adapter, MEM_MC1); if (cause & EDC0_F) mem_intr_handler(adapter, MEM_EDC0); @@ -2871,17 +3405,18 @@ int t4_slow_intr_handler(struct adapter *adapter) */ void t4_intr_enable(struct adapter *adapter) { + u32 val = 0; u32 pf = SOURCEPF_G(t4_read_reg(adapter, PL_WHOAMI_A)); + if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) + val = ERR_DROPPED_DB_F | ERR_EGR_CTXT_PRIO_F | DBFIFO_HP_INT_F; t4_write_reg(adapter, SGE_INT_ENABLE3_A, ERR_CPL_EXCEED_IQE_SIZE_F | ERR_INVALID_CIDX_INC_F | ERR_CPL_OPCODE_0_F | - ERR_DROPPED_DB_F | ERR_DATA_CPL_ON_HIGH_QID1_F | + ERR_DATA_CPL_ON_HIGH_QID1_F | INGRESS_SIZE_ERR_F | ERR_DATA_CPL_ON_HIGH_QID0_F | ERR_BAD_DB_PIDX3_F | ERR_BAD_DB_PIDX2_F | ERR_BAD_DB_PIDX1_F | ERR_BAD_DB_PIDX0_F | ERR_ING_CTXT_PRIO_F | - ERR_EGR_CTXT_PRIO_F | INGRESS_SIZE_ERR_F | - DBFIFO_HP_INT_F | DBFIFO_LP_INT_F | - EGRESS_SIZE_ERR_F); + DBFIFO_LP_INT_F | EGRESS_SIZE_ERR_F | val); t4_write_reg(adapter, MYPF_REG(PL_PF_INT_ENABLE_A), PF_INTR_MASK); t4_set_reg_field(adapter, PL_INT_MAP0_A, 0, 1 << pf); } @@ -2945,18 +3480,18 @@ int t4_config_rss_range(struct adapter *adapter, int mbox, unsigned int viid, struct fw_rss_ind_tbl_cmd cmd; memset(&cmd, 0, sizeof(cmd)); - cmd.op_to_viid = htonl(FW_CMD_OP_V(FW_RSS_IND_TBL_CMD) | + cmd.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_RSS_IND_TBL_CMD) | FW_CMD_REQUEST_F | FW_CMD_WRITE_F | FW_RSS_IND_TBL_CMD_VIID_V(viid)); - cmd.retval_len16 = htonl(FW_LEN16(cmd)); + cmd.retval_len16 = cpu_to_be32(FW_LEN16(cmd)); /* each fw_rss_ind_tbl_cmd takes up to 32 entries */ while (n > 0) { int nq = min(n, 32); __be32 *qp = &cmd.iq0_to_iq2; - cmd.niqid = htons(nq); - cmd.startidx = htons(start); + cmd.niqid = cpu_to_be16(nq); + cmd.startidx = cpu_to_be16(start); start += nq; n -= nq; @@ -2974,7 +3509,7 @@ int t4_config_rss_range(struct adapter *adapter, int mbox, unsigned int viid, if (++rsp >= rsp_end) rsp = rspq; - *qp++ = htonl(v); + *qp++ = cpu_to_be32(v); nq -= 3; } @@ -3000,20 +3535,46 @@ int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode, struct fw_rss_glb_config_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_write = htonl(FW_CMD_OP_V(FW_RSS_GLB_CONFIG_CMD) | - FW_CMD_REQUEST_F | FW_CMD_WRITE_F); - c.retval_len16 = htonl(FW_LEN16(c)); + c.op_to_write = cpu_to_be32(FW_CMD_OP_V(FW_RSS_GLB_CONFIG_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F); + c.retval_len16 = cpu_to_be32(FW_LEN16(c)); if (mode == FW_RSS_GLB_CONFIG_CMD_MODE_MANUAL) { - c.u.manual.mode_pkd = htonl(FW_RSS_GLB_CONFIG_CMD_MODE_V(mode)); + c.u.manual.mode_pkd = + cpu_to_be32(FW_RSS_GLB_CONFIG_CMD_MODE_V(mode)); } else if (mode == FW_RSS_GLB_CONFIG_CMD_MODE_BASICVIRTUAL) { c.u.basicvirtual.mode_pkd = - htonl(FW_RSS_GLB_CONFIG_CMD_MODE_V(mode)); - c.u.basicvirtual.synmapen_to_hashtoeplitz = htonl(flags); + cpu_to_be32(FW_RSS_GLB_CONFIG_CMD_MODE_V(mode)); + c.u.basicvirtual.synmapen_to_hashtoeplitz = cpu_to_be32(flags); } else return -EINVAL; return t4_wr_mbox(adapter, mbox, &c, sizeof(c), NULL); } +/** + * t4_config_vi_rss - configure per VI RSS settings + * @adapter: the adapter + * @mbox: mbox to use for the FW command + * @viid: the VI id + * @flags: RSS flags + * @defq: id of the default RSS queue for the VI. + * + * Configures VI-specific RSS properties. + */ +int t4_config_vi_rss(struct adapter *adapter, int mbox, unsigned int viid, + unsigned int flags, unsigned int defq) +{ + struct fw_rss_vi_config_cmd c; + + memset(&c, 0, sizeof(c)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_RSS_VI_CONFIG_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + FW_RSS_VI_CONFIG_CMD_VIID_V(viid)); + c.retval_len16 = cpu_to_be32(FW_LEN16(c)); + c.u.basicvirtual.defaultq_to_udpen = cpu_to_be32(flags | + FW_RSS_VI_CONFIG_CMD_DEFAULTQ_V(defq)); + return t4_wr_mbox(adapter, mbox, &c, sizeof(c), NULL); +} + /* Read an RSS table row */ static int rd_rss_row(struct adapter *adap, int row, u32 *val) { @@ -3069,11 +3630,29 @@ void t4_read_rss_key(struct adapter *adap, u32 *key) */ void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx) { + u8 rss_key_addr_cnt = 16; + u32 vrt = t4_read_reg(adap, TP_RSS_CONFIG_VRT_A); + + /* T6 and later: for KeyMode 3 (per-vf and per-vf scramble), + * allows access to key addresses 16-63 by using KeyWrAddrX + * as index[5:4](upper 2) into key table + */ + if ((CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5) && + (vrt & KEYEXTEND_F) && (KEYMODE_G(vrt) == 3)) + rss_key_addr_cnt = 32; + t4_write_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, key, 10, TP_RSS_SECRET_KEY0_A); - if (idx >= 0 && idx < 16) - t4_write_reg(adap, TP_RSS_CONFIG_VRT_A, - KEYWRADDR_V(idx) | KEYWREN_F); + + if (idx >= 0 && idx < rss_key_addr_cnt) { + if (rss_key_addr_cnt > 16) + t4_write_reg(adap, TP_RSS_CONFIG_VRT_A, + KEYWRADDRX_V(idx >> 4) | + T6_VFWRADDR_V(idx) | KEYWREN_F); + else + t4_write_reg(adap, TP_RSS_CONFIG_VRT_A, + KEYWRADDR_V(idx) | KEYWREN_F); + } } /** @@ -3107,8 +3686,13 @@ void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index, { u32 vrt, mask, data; - mask = VFWRADDR_V(VFWRADDR_M); - data = VFWRADDR_V(index); + if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) { + mask = VFWRADDR_V(VFWRADDR_M); + data = VFWRADDR_V(index); + } else { + mask = T6_VFWRADDR_V(T6_VFWRADDR_M); + data = T6_VFWRADDR_V(index); + } /* Request that the index'th VF Table values be read into VFL/VFH. */ @@ -3401,7 +3985,7 @@ void t4_pmrx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]) } /** - * get_mps_bg_map - return the buffer groups associated with a port + * t4_get_mps_bg_map - return the buffer groups associated with a port * @adap: the adapter * @idx: the port index * @@ -3409,7 +3993,7 @@ void t4_pmrx_get_stats(struct adapter *adap, u32 cnt[], u64 cycles[]) * with the given port. Bit i is set if buffer group i is used by the * port. */ -static unsigned int get_mps_bg_map(struct adapter *adap, int idx) +unsigned int t4_get_mps_bg_map(struct adapter *adap, int idx) { u32 n = NUMPORTS_G(t4_read_reg(adap, MPS_CMN_CTL_A)); @@ -3460,7 +4044,7 @@ const char *t4_get_port_type_description(enum fw_port_type port_type) */ void t4_get_port_stats(struct adapter *adap, int idx, struct port_stats *p) { - u32 bgmap = get_mps_bg_map(adap, idx); + u32 bgmap = t4_get_mps_bg_map(adap, idx); #define GET_STAT(name) \ t4_read_reg64(adap, \ @@ -3644,33 +4228,38 @@ int t4_wol_pat_enable(struct adapter *adap, unsigned int port, unsigned int map, void t4_mk_filtdelwr(unsigned int ftid, struct fw_filter_wr *wr, int qid) { memset(wr, 0, sizeof(*wr)); - wr->op_pkd = htonl(FW_WR_OP_V(FW_FILTER_WR)); - wr->len16_pkd = htonl(FW_WR_LEN16_V(sizeof(*wr) / 16)); - wr->tid_to_iq = htonl(FW_FILTER_WR_TID_V(ftid) | - FW_FILTER_WR_NOREPLY_V(qid < 0)); - wr->del_filter_to_l2tix = htonl(FW_FILTER_WR_DEL_FILTER_F); + wr->op_pkd = cpu_to_be32(FW_WR_OP_V(FW_FILTER_WR)); + wr->len16_pkd = cpu_to_be32(FW_WR_LEN16_V(sizeof(*wr) / 16)); + wr->tid_to_iq = cpu_to_be32(FW_FILTER_WR_TID_V(ftid) | + FW_FILTER_WR_NOREPLY_V(qid < 0)); + wr->del_filter_to_l2tix = cpu_to_be32(FW_FILTER_WR_DEL_FILTER_F); if (qid >= 0) - wr->rx_chan_rx_rpl_iq = htons(FW_FILTER_WR_RX_RPL_IQ_V(qid)); + wr->rx_chan_rx_rpl_iq = + cpu_to_be16(FW_FILTER_WR_RX_RPL_IQ_V(qid)); } #define INIT_CMD(var, cmd, rd_wr) do { \ - (var).op_to_write = htonl(FW_CMD_OP_V(FW_##cmd##_CMD) | \ - FW_CMD_REQUEST_F | FW_CMD_##rd_wr##_F); \ - (var).retval_len16 = htonl(FW_LEN16(var)); \ + (var).op_to_write = cpu_to_be32(FW_CMD_OP_V(FW_##cmd##_CMD) | \ + FW_CMD_REQUEST_F | \ + FW_CMD_##rd_wr##_F); \ + (var).retval_len16 = cpu_to_be32(FW_LEN16(var)); \ } while (0) int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox, u32 addr, u32 val) { + u32 ldst_addrspace; struct fw_ldst_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_addrspace = htonl(FW_CMD_OP_V(FW_LDST_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | - FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_FIRMWARE)); - c.cycles_to_len16 = htonl(FW_LEN16(c)); - c.u.addrval.addr = htonl(addr); - c.u.addrval.val = htonl(val); + ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_FIRMWARE); + c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_WRITE_F | + ldst_addrspace); + c.cycles_to_len16 = cpu_to_be32(FW_LEN16(c)); + c.u.addrval.addr = cpu_to_be32(addr); + c.u.addrval.val = cpu_to_be32(val); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -3690,19 +4279,22 @@ int t4_mdio_rd(struct adapter *adap, unsigned int mbox, unsigned int phy_addr, unsigned int mmd, unsigned int reg, u16 *valp) { int ret; + u32 ldst_addrspace; struct fw_ldst_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_addrspace = htonl(FW_CMD_OP_V(FW_LDST_CMD) | FW_CMD_REQUEST_F | - FW_CMD_READ_F | FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MDIO)); - c.cycles_to_len16 = htonl(FW_LEN16(c)); - c.u.mdio.paddr_mmd = htons(FW_LDST_CMD_PADDR_V(phy_addr) | - FW_LDST_CMD_MMD_V(mmd)); - c.u.mdio.raddr = htons(reg); + ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MDIO); + c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F | + ldst_addrspace); + c.cycles_to_len16 = cpu_to_be32(FW_LEN16(c)); + c.u.mdio.paddr_mmd = cpu_to_be16(FW_LDST_CMD_PADDR_V(phy_addr) | + FW_LDST_CMD_MMD_V(mmd)); + c.u.mdio.raddr = cpu_to_be16(reg); ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c); if (ret == 0) - *valp = ntohs(c.u.mdio.rval); + *valp = be16_to_cpu(c.u.mdio.rval); return ret; } @@ -3720,16 +4312,19 @@ int t4_mdio_rd(struct adapter *adap, unsigned int mbox, unsigned int phy_addr, int t4_mdio_wr(struct adapter *adap, unsigned int mbox, unsigned int phy_addr, unsigned int mmd, unsigned int reg, u16 val) { + u32 ldst_addrspace; struct fw_ldst_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_addrspace = htonl(FW_CMD_OP_V(FW_LDST_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MDIO)); - c.cycles_to_len16 = htonl(FW_LEN16(c)); - c.u.mdio.paddr_mmd = htons(FW_LDST_CMD_PADDR_V(phy_addr) | - FW_LDST_CMD_MMD_V(mmd)); - c.u.mdio.raddr = htons(reg); - c.u.mdio.rval = htons(val); + ldst_addrspace = FW_LDST_CMD_ADDRSPACE_V(FW_LDST_ADDRSPC_MDIO); + c.op_to_addrspace = cpu_to_be32(FW_CMD_OP_V(FW_LDST_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + ldst_addrspace); + c.cycles_to_len16 = cpu_to_be32(FW_LEN16(c)); + c.u.mdio.paddr_mmd = cpu_to_be16(FW_LDST_CMD_PADDR_V(phy_addr) | + FW_LDST_CMD_MMD_V(mmd)); + c.u.mdio.raddr = cpu_to_be16(reg); + c.u.mdio.rval = cpu_to_be16(val); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -3863,11 +4458,11 @@ int t4_fw_hello(struct adapter *adap, unsigned int mbox, unsigned int evt_mbox, retry: memset(&c, 0, sizeof(c)); INIT_CMD(c, HELLO, WRITE); - c.err_to_clearinit = htonl( + c.err_to_clearinit = cpu_to_be32( FW_HELLO_CMD_MASTERDIS_V(master == MASTER_CANT) | FW_HELLO_CMD_MASTERFORCE_V(master == MASTER_MUST) | - FW_HELLO_CMD_MBMASTER_V(master == MASTER_MUST ? mbox : - FW_HELLO_CMD_MBMASTER_M) | + FW_HELLO_CMD_MBMASTER_V(master == MASTER_MUST ? + mbox : FW_HELLO_CMD_MBMASTER_M) | FW_HELLO_CMD_MBASYNCNOT_V(evt_mbox) | FW_HELLO_CMD_STAGE_V(fw_hello_cmd_stage_os) | FW_HELLO_CMD_CLEARINIT_F); @@ -3888,7 +4483,7 @@ retry: return ret; } - v = ntohl(c.err_to_clearinit); + v = be32_to_cpu(c.err_to_clearinit); master_mbox = FW_HELLO_CMD_MBMASTER_G(v); if (state) { if (v & FW_HELLO_CMD_ERR_F) @@ -4017,7 +4612,7 @@ int t4_fw_reset(struct adapter *adap, unsigned int mbox, int reset) memset(&c, 0, sizeof(c)); INIT_CMD(c, RESET, WRITE); - c.val = htonl(reset); + c.val = cpu_to_be32(reset); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4050,8 +4645,8 @@ static int t4_fw_halt(struct adapter *adap, unsigned int mbox, int force) memset(&c, 0, sizeof(c)); INIT_CMD(c, RESET, WRITE); - c.val = htonl(PIORST_F | PIORSTMODE_F); - c.halt_pkd = htonl(FW_RESET_CMD_HALT_F); + c.val = cpu_to_be32(PIORST_F | PIORSTMODE_F); + c.halt_pkd = cpu_to_be32(FW_RESET_CMD_HALT_F); ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4190,7 +4785,7 @@ int t4_fw_upgrade(struct adapter *adap, unsigned int mbox, * the newly loaded firmware will handle this right by checking * its header flags to see if it advertises the capability. */ - reset = ((ntohl(fw_hdr->flags) & FW_HDR_FLAGS_RESET_HALT) == 0); + reset = ((be32_to_cpu(fw_hdr->flags) & FW_HDR_FLAGS_RESET_HALT) == 0); return t4_fw_restart(adap, mbox, reset); } @@ -4321,7 +4916,7 @@ int t4_fw_initialize(struct adapter *adap, unsigned int mbox) } /** - * t4_query_params - query FW or device parameters + * t4_query_params_rw - query FW or device parameters * @adap: the adapter * @mbox: mailbox to use for the FW command * @pf: the PF @@ -4329,13 +4924,14 @@ int t4_fw_initialize(struct adapter *adap, unsigned int mbox) * @nparams: the number of parameters * @params: the parameter names * @val: the parameter values + * @rw: Write and read flag * * Reads the value of FW or device parameters. Up to 7 parameters can be * queried at once. */ -int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf, - unsigned int vf, unsigned int nparams, const u32 *params, - u32 *val) +int t4_query_params_rw(struct adapter *adap, unsigned int mbox, unsigned int pf, + unsigned int vf, unsigned int nparams, const u32 *params, + u32 *val, int rw) { int i, ret; struct fw_params_cmd c; @@ -4345,22 +4941,35 @@ int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf, return -EINVAL; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_PARAMS_CMD) | FW_CMD_REQUEST_F | - FW_CMD_READ_F | FW_PARAMS_CMD_PFN_V(pf) | - FW_PARAMS_CMD_VFN_V(vf)); - c.retval_len16 = htonl(FW_LEN16(c)); - for (i = 0; i < nparams; i++, p += 2) - *p = htonl(*params++); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PARAMS_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F | + FW_PARAMS_CMD_PFN_V(pf) | + FW_PARAMS_CMD_VFN_V(vf)); + c.retval_len16 = cpu_to_be32(FW_LEN16(c)); + + for (i = 0; i < nparams; i++) { + *p++ = cpu_to_be32(*params++); + if (rw) + *p = cpu_to_be32(*(val + i)); + p++; + } ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c); if (ret == 0) for (i = 0, p = &c.param[0].val; i < nparams; i++, p += 2) - *val++ = ntohl(*p); + *val++ = be32_to_cpu(*p); return ret; } +int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf, + unsigned int vf, unsigned int nparams, const u32 *params, + u32 *val) +{ + return t4_query_params_rw(adap, mbox, pf, vf, nparams, params, val, 0); +} + /** - * t4_set_params_nosleep - sets FW or device parameters + * t4_set_params_timeout - sets FW or device parameters * @adap: the adapter * @mbox: mailbox to use for the FW command * @pf: the PF @@ -4368,15 +4977,15 @@ int t4_query_params(struct adapter *adap, unsigned int mbox, unsigned int pf, * @nparams: the number of parameters * @params: the parameter names * @val: the parameter values + * @timeout: the timeout time * - * Does not ever sleep * Sets the value of FW or device parameters. Up to 7 parameters can be * specified at once. */ -int t4_set_params_nosleep(struct adapter *adap, unsigned int mbox, +int t4_set_params_timeout(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int nparams, const u32 *params, - const u32 *val) + const u32 *val, int timeout) { struct fw_params_cmd c; __be32 *p = &c.param[0].mnem; @@ -4386,9 +4995,9 @@ int t4_set_params_nosleep(struct adapter *adap, unsigned int mbox, memset(&c, 0, sizeof(c)); c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PARAMS_CMD) | - FW_CMD_REQUEST_F | FW_CMD_WRITE_F | - FW_PARAMS_CMD_PFN_V(pf) | - FW_PARAMS_CMD_VFN_V(vf)); + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + FW_PARAMS_CMD_PFN_V(pf) | + FW_PARAMS_CMD_VFN_V(vf)); c.retval_len16 = cpu_to_be32(FW_LEN16(c)); while (nparams--) { @@ -4396,7 +5005,7 @@ int t4_set_params_nosleep(struct adapter *adap, unsigned int mbox, *p++ = cpu_to_be32(*val++); } - return t4_wr_mbox_ns(adap, mbox, &c, sizeof(c), NULL); + return t4_wr_mbox_timeout(adap, mbox, &c, sizeof(c), NULL, timeout); } /** @@ -4416,23 +5025,8 @@ int t4_set_params(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int nparams, const u32 *params, const u32 *val) { - struct fw_params_cmd c; - __be32 *p = &c.param[0].mnem; - - if (nparams > 7) - return -EINVAL; - - memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_PARAMS_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_PARAMS_CMD_PFN_V(pf) | - FW_PARAMS_CMD_VFN_V(vf)); - c.retval_len16 = htonl(FW_LEN16(c)); - while (nparams--) { - *p++ = htonl(*params++); - *p++ = htonl(*val++); - } - - return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); + return t4_set_params_timeout(adap, mbox, pf, vf, nparams, params, val, + FW_CMD_MAX_TIMEOUT); } /** @@ -4465,20 +5059,21 @@ int t4_cfg_pfvf(struct adapter *adap, unsigned int mbox, unsigned int pf, struct fw_pfvf_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_PFVF_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_PFVF_CMD_PFN_V(pf) | - FW_PFVF_CMD_VFN_V(vf)); - c.retval_len16 = htonl(FW_LEN16(c)); - c.niqflint_niq = htonl(FW_PFVF_CMD_NIQFLINT_V(rxqi) | - FW_PFVF_CMD_NIQ_V(rxq)); - c.type_to_neq = htonl(FW_PFVF_CMD_CMASK_V(cmask) | - FW_PFVF_CMD_PMASK_V(pmask) | - FW_PFVF_CMD_NEQ_V(txq)); - c.tc_to_nexactf = htonl(FW_PFVF_CMD_TC_V(tc) | FW_PFVF_CMD_NVI_V(vi) | - FW_PFVF_CMD_NEXACTF_V(nexact)); - c.r_caps_to_nethctrl = htonl(FW_PFVF_CMD_R_CAPS_V(rcaps) | - FW_PFVF_CMD_WX_CAPS_V(wxcaps) | - FW_PFVF_CMD_NETHCTRL_V(txq_eth_ctrl)); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PFVF_CMD) | FW_CMD_REQUEST_F | + FW_CMD_WRITE_F | FW_PFVF_CMD_PFN_V(pf) | + FW_PFVF_CMD_VFN_V(vf)); + c.retval_len16 = cpu_to_be32(FW_LEN16(c)); + c.niqflint_niq = cpu_to_be32(FW_PFVF_CMD_NIQFLINT_V(rxqi) | + FW_PFVF_CMD_NIQ_V(rxq)); + c.type_to_neq = cpu_to_be32(FW_PFVF_CMD_CMASK_V(cmask) | + FW_PFVF_CMD_PMASK_V(pmask) | + FW_PFVF_CMD_NEQ_V(txq)); + c.tc_to_nexactf = cpu_to_be32(FW_PFVF_CMD_TC_V(tc) | + FW_PFVF_CMD_NVI_V(vi) | + FW_PFVF_CMD_NEXACTF_V(nexact)); + c.r_caps_to_nethctrl = cpu_to_be32(FW_PFVF_CMD_R_CAPS_V(rcaps) | + FW_PFVF_CMD_WX_CAPS_V(wxcaps) | + FW_PFVF_CMD_NETHCTRL_V(txq_eth_ctrl)); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4507,10 +5102,10 @@ int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port, struct fw_vi_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_VI_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_CMD_EXEC_F | - FW_VI_CMD_PFN_V(pf) | FW_VI_CMD_VFN_V(vf)); - c.alloc_to_len16 = htonl(FW_VI_CMD_ALLOC_F | FW_LEN16(c)); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_VI_CMD) | FW_CMD_REQUEST_F | + FW_CMD_WRITE_F | FW_CMD_EXEC_F | + FW_VI_CMD_PFN_V(pf) | FW_VI_CMD_VFN_V(vf)); + c.alloc_to_len16 = cpu_to_be32(FW_VI_CMD_ALLOC_F | FW_LEN16(c)); c.portid_pkd = FW_VI_CMD_PORTID_V(port); c.nmac = nmac - 1; @@ -4532,8 +5127,8 @@ int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port, } } if (rss_size) - *rss_size = FW_VI_CMD_RSSSIZE_G(ntohs(c.rsssize_pkd)); - return FW_VI_CMD_VIID_G(ntohs(c.type_viid)); + *rss_size = FW_VI_CMD_RSSSIZE_G(be16_to_cpu(c.rsssize_pkd)); + return FW_VI_CMD_VIID_G(be16_to_cpu(c.type_viid)); } /** @@ -4569,14 +5164,16 @@ int t4_set_rxmode(struct adapter *adap, unsigned int mbox, unsigned int viid, vlanex = FW_VI_RXMODE_CMD_VLANEXEN_M; memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_RXMODE_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_VI_RXMODE_CMD_VIID_V(viid)); - c.retval_len16 = htonl(FW_LEN16(c)); - c.mtu_to_vlanexen = htonl(FW_VI_RXMODE_CMD_MTU_V(mtu) | - FW_VI_RXMODE_CMD_PROMISCEN_V(promisc) | - FW_VI_RXMODE_CMD_ALLMULTIEN_V(all_multi) | - FW_VI_RXMODE_CMD_BROADCASTEN_V(bcast) | - FW_VI_RXMODE_CMD_VLANEXEN_V(vlanex)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_RXMODE_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + FW_VI_RXMODE_CMD_VIID_V(viid)); + c.retval_len16 = cpu_to_be32(FW_LEN16(c)); + c.mtu_to_vlanexen = + cpu_to_be32(FW_VI_RXMODE_CMD_MTU_V(mtu) | + FW_VI_RXMODE_CMD_PROMISCEN_V(promisc) | + FW_VI_RXMODE_CMD_ALLMULTIEN_V(all_multi) | + FW_VI_RXMODE_CMD_BROADCASTEN_V(bcast) | + FW_VI_RXMODE_CMD_VLANEXEN_V(vlanex)); return t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), NULL, sleep_ok); } @@ -4606,43 +5203,71 @@ int t4_alloc_mac_filt(struct adapter *adap, unsigned int mbox, unsigned int viid, bool free, unsigned int naddr, const u8 **addr, u16 *idx, u64 *hash, bool sleep_ok) { - int i, ret; + int offset, ret = 0; struct fw_vi_mac_cmd c; - struct fw_vi_mac_exact *p; - unsigned int max_naddr = is_t4(adap->params.chip) ? - NUM_MPS_CLS_SRAM_L_INSTANCES : - NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + unsigned int nfilters = 0; + unsigned int max_naddr = adap->params.arch.mps_tcam_size; + unsigned int rem = naddr; - if (naddr > 7) + if (naddr > max_naddr) return -EINVAL; - memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_MAC_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | (free ? FW_CMD_EXEC_F : 0) | - FW_VI_MAC_CMD_VIID_V(viid)); - c.freemacs_to_len16 = htonl(FW_VI_MAC_CMD_FREEMACS_V(free) | - FW_CMD_LEN16_V((naddr + 2) / 2)); - - for (i = 0, p = c.u.exact; i < naddr; i++, p++) { - p->valid_to_idx = htons(FW_VI_MAC_CMD_VALID_F | - FW_VI_MAC_CMD_IDX_V(FW_VI_MAC_ADD_MAC)); - memcpy(p->macaddr, addr[i], sizeof(p->macaddr)); - } + for (offset = 0; offset < naddr ; /**/) { + unsigned int fw_naddr = (rem < ARRAY_SIZE(c.u.exact) ? + rem : ARRAY_SIZE(c.u.exact)); + size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd, + u.exact[fw_naddr]), 16); + struct fw_vi_mac_exact *p; + int i; - ret = t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), &c, sleep_ok); - if (ret) - return ret; + memset(&c, 0, sizeof(c)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_WRITE_F | + FW_CMD_EXEC_V(free) | + FW_VI_MAC_CMD_VIID_V(viid)); + c.freemacs_to_len16 = + cpu_to_be32(FW_VI_MAC_CMD_FREEMACS_V(free) | + FW_CMD_LEN16_V(len16)); + + for (i = 0, p = c.u.exact; i < fw_naddr; i++, p++) { + p->valid_to_idx = + cpu_to_be16(FW_VI_MAC_CMD_VALID_F | + FW_VI_MAC_CMD_IDX_V( + FW_VI_MAC_ADD_MAC)); + memcpy(p->macaddr, addr[offset + i], + sizeof(p->macaddr)); + } + + /* It's okay if we run out of space in our MAC address arena. + * Some of the addresses we submit may get stored so we need + * to run through the reply to see what the results were ... + */ + ret = t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), &c, sleep_ok); + if (ret && ret != -FW_ENOMEM) + break; - for (i = 0, p = c.u.exact; i < naddr; i++, p++) { - u16 index = FW_VI_MAC_CMD_IDX_G(ntohs(p->valid_to_idx)); + for (i = 0, p = c.u.exact; i < fw_naddr; i++, p++) { + u16 index = FW_VI_MAC_CMD_IDX_G( + be16_to_cpu(p->valid_to_idx)); + + if (idx) + idx[offset + i] = (index >= max_naddr ? + 0xffff : index); + if (index < max_naddr) + nfilters++; + else if (hash) + *hash |= (1ULL << + hash_mac_addr(addr[offset + i])); + } - if (idx) - idx[i] = index >= max_naddr ? 0xffff : index; - if (index < max_naddr) - ret++; - else if (hash) - *hash |= (1ULL << hash_mac_addr(addr[i])); + free = false; + offset += fw_naddr; + rem -= fw_naddr; } + + if (ret == 0 || ret == -FW_ENOMEM) + ret = nfilters; return ret; } @@ -4671,26 +5296,25 @@ int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid, int ret, mode; struct fw_vi_mac_cmd c; struct fw_vi_mac_exact *p = c.u.exact; - unsigned int max_mac_addr = is_t4(adap->params.chip) ? - NUM_MPS_CLS_SRAM_L_INSTANCES : - NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + unsigned int max_mac_addr = adap->params.arch.mps_tcam_size; if (idx < 0) /* new allocation */ idx = persist ? FW_VI_MAC_ADD_PERSIST_MAC : FW_VI_MAC_ADD_MAC; mode = add_smt ? FW_VI_MAC_SMT_AND_MPSTCAM : FW_VI_MAC_MPS_TCAM_ENTRY; memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_MAC_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_VI_MAC_CMD_VIID_V(viid)); - c.freemacs_to_len16 = htonl(FW_CMD_LEN16_V(1)); - p->valid_to_idx = htons(FW_VI_MAC_CMD_VALID_F | - FW_VI_MAC_CMD_SMAC_RESULT_V(mode) | - FW_VI_MAC_CMD_IDX_V(idx)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + FW_VI_MAC_CMD_VIID_V(viid)); + c.freemacs_to_len16 = cpu_to_be32(FW_CMD_LEN16_V(1)); + p->valid_to_idx = cpu_to_be16(FW_VI_MAC_CMD_VALID_F | + FW_VI_MAC_CMD_SMAC_RESULT_V(mode) | + FW_VI_MAC_CMD_IDX_V(idx)); memcpy(p->macaddr, addr, sizeof(p->macaddr)); ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c); if (ret == 0) { - ret = FW_VI_MAC_CMD_IDX_G(ntohs(p->valid_to_idx)); + ret = FW_VI_MAC_CMD_IDX_G(be16_to_cpu(p->valid_to_idx)); if (ret >= max_mac_addr) ret = -ENOMEM; } @@ -4714,11 +5338,12 @@ int t4_set_addr_hash(struct adapter *adap, unsigned int mbox, unsigned int viid, struct fw_vi_mac_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_MAC_CMD) | FW_CMD_REQUEST_F | - FW_CMD_WRITE_F | FW_VI_ENABLE_CMD_VIID_V(viid)); - c.freemacs_to_len16 = htonl(FW_VI_MAC_CMD_HASHVECEN_F | - FW_VI_MAC_CMD_HASHUNIEN_V(ucast) | - FW_CMD_LEN16_V(1)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) | + FW_CMD_REQUEST_F | FW_CMD_WRITE_F | + FW_VI_ENABLE_CMD_VIID_V(viid)); + c.freemacs_to_len16 = cpu_to_be32(FW_VI_MAC_CMD_HASHVECEN_F | + FW_VI_MAC_CMD_HASHUNIEN_V(ucast) | + FW_CMD_LEN16_V(1)); c.u.hash.hashvec = cpu_to_be64(vec); return t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), NULL, sleep_ok); } @@ -4741,12 +5366,13 @@ int t4_enable_vi_params(struct adapter *adap, unsigned int mbox, struct fw_vi_enable_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_ENABLE_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_VI_ENABLE_CMD_VIID_V(viid)); - - c.ien_to_len16 = htonl(FW_VI_ENABLE_CMD_IEN_V(rx_en) | - FW_VI_ENABLE_CMD_EEN_V(tx_en) | FW_LEN16(c) | - FW_VI_ENABLE_CMD_DCB_INFO_V(dcb_en)); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_ENABLE_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_VI_ENABLE_CMD_VIID_V(viid)); + c.ien_to_len16 = cpu_to_be32(FW_VI_ENABLE_CMD_IEN_V(rx_en) | + FW_VI_ENABLE_CMD_EEN_V(tx_en) | + FW_VI_ENABLE_CMD_DCB_INFO_V(dcb_en) | + FW_LEN16(c)); return t4_wr_mbox_ns(adap, mbox, &c, sizeof(c), NULL); } @@ -4781,10 +5407,11 @@ int t4_identify_port(struct adapter *adap, unsigned int mbox, unsigned int viid, struct fw_vi_enable_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_viid = htonl(FW_CMD_OP_V(FW_VI_ENABLE_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_VI_ENABLE_CMD_VIID_V(viid)); - c.ien_to_len16 = htonl(FW_VI_ENABLE_CMD_LED_F | FW_LEN16(c)); - c.blinkdur = htons(nblinks); + c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_ENABLE_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_VI_ENABLE_CMD_VIID_V(viid)); + c.ien_to_len16 = cpu_to_be32(FW_VI_ENABLE_CMD_LED_F | FW_LEN16(c)); + c.blinkdur = cpu_to_be16(nblinks); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4808,14 +5435,14 @@ int t4_iq_free(struct adapter *adap, unsigned int mbox, unsigned int pf, struct fw_iq_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_IQ_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_IQ_CMD_PFN_V(pf) | - FW_IQ_CMD_VFN_V(vf)); - c.alloc_to_len16 = htonl(FW_IQ_CMD_FREE_F | FW_LEN16(c)); - c.type_to_iqandstindex = htonl(FW_IQ_CMD_TYPE_V(iqtype)); - c.iqid = htons(iqid); - c.fl0id = htons(fl0id); - c.fl1id = htons(fl1id); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_IQ_CMD) | FW_CMD_REQUEST_F | + FW_CMD_EXEC_F | FW_IQ_CMD_PFN_V(pf) | + FW_IQ_CMD_VFN_V(vf)); + c.alloc_to_len16 = cpu_to_be32(FW_IQ_CMD_FREE_F | FW_LEN16(c)); + c.type_to_iqandstindex = cpu_to_be32(FW_IQ_CMD_TYPE_V(iqtype)); + c.iqid = cpu_to_be16(iqid); + c.fl0id = cpu_to_be16(fl0id); + c.fl1id = cpu_to_be16(fl1id); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4835,11 +5462,12 @@ int t4_eth_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf, struct fw_eq_eth_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_ETH_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_EQ_ETH_CMD_PFN_V(pf) | - FW_EQ_ETH_CMD_VFN_V(vf)); - c.alloc_to_len16 = htonl(FW_EQ_ETH_CMD_FREE_F | FW_LEN16(c)); - c.eqid_pkd = htonl(FW_EQ_ETH_CMD_EQID_V(eqid)); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_EQ_ETH_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_EQ_ETH_CMD_PFN_V(pf) | + FW_EQ_ETH_CMD_VFN_V(vf)); + c.alloc_to_len16 = cpu_to_be32(FW_EQ_ETH_CMD_FREE_F | FW_LEN16(c)); + c.eqid_pkd = cpu_to_be32(FW_EQ_ETH_CMD_EQID_V(eqid)); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4859,11 +5487,12 @@ int t4_ctrl_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf, struct fw_eq_ctrl_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_CTRL_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_EQ_CTRL_CMD_PFN_V(pf) | - FW_EQ_CTRL_CMD_VFN_V(vf)); - c.alloc_to_len16 = htonl(FW_EQ_CTRL_CMD_FREE_F | FW_LEN16(c)); - c.cmpliqid_eqid = htonl(FW_EQ_CTRL_CMD_EQID_V(eqid)); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_EQ_CTRL_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_EQ_CTRL_CMD_PFN_V(pf) | + FW_EQ_CTRL_CMD_VFN_V(vf)); + c.alloc_to_len16 = cpu_to_be32(FW_EQ_CTRL_CMD_FREE_F | FW_LEN16(c)); + c.cmpliqid_eqid = cpu_to_be32(FW_EQ_CTRL_CMD_EQID_V(eqid)); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4883,11 +5512,12 @@ int t4_ofld_eq_free(struct adapter *adap, unsigned int mbox, unsigned int pf, struct fw_eq_ofld_cmd c; memset(&c, 0, sizeof(c)); - c.op_to_vfn = htonl(FW_CMD_OP_V(FW_EQ_OFLD_CMD) | FW_CMD_REQUEST_F | - FW_CMD_EXEC_F | FW_EQ_OFLD_CMD_PFN_V(pf) | - FW_EQ_OFLD_CMD_VFN_V(vf)); - c.alloc_to_len16 = htonl(FW_EQ_OFLD_CMD_FREE_F | FW_LEN16(c)); - c.eqid_pkd = htonl(FW_EQ_OFLD_CMD_EQID_V(eqid)); + c.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_EQ_OFLD_CMD) | + FW_CMD_REQUEST_F | FW_CMD_EXEC_F | + FW_EQ_OFLD_CMD_PFN_V(pf) | + FW_EQ_OFLD_CMD_VFN_V(vf)); + c.alloc_to_len16 = cpu_to_be32(FW_EQ_OFLD_CMD_FREE_F | FW_LEN16(c)); + c.eqid_pkd = cpu_to_be32(FW_EQ_OFLD_CMD_EQID_V(eqid)); return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } @@ -4905,11 +5535,11 @@ int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl) if (opcode == FW_PORT_CMD) { /* link/module state change message */ int speed = 0, fc = 0; const struct fw_port_cmd *p = (void *)rpl; - int chan = FW_PORT_CMD_PORTID_G(ntohl(p->op_to_portid)); + int chan = FW_PORT_CMD_PORTID_G(be32_to_cpu(p->op_to_portid)); int port = adap->chan_map[chan]; struct port_info *pi = adap2pinfo(adap, port); struct link_config *lc = &pi->link_cfg; - u32 stat = ntohl(p->u.info.lstatus_to_modtype); + u32 stat = be32_to_cpu(p->u.info.lstatus_to_modtype); int link_ok = (stat & FW_PORT_CMD_LSTATUS_F) != 0; u32 mod = FW_PORT_CMD_MODTYPE_G(stat); @@ -5075,9 +5705,30 @@ int t4_prep_adapter(struct adapter *adapter) switch (ver) { case CHELSIO_T4: adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T4, pl_rev); + adapter->params.arch.sge_fl_db = DBPRIO_F; + adapter->params.arch.mps_tcam_size = + NUM_MPS_CLS_SRAM_L_INSTANCES; + adapter->params.arch.mps_rplc_size = 128; + adapter->params.arch.nchan = NCHAN; + adapter->params.arch.vfcount = 128; break; case CHELSIO_T5: adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T5, pl_rev); + adapter->params.arch.sge_fl_db = DBPRIO_F | DBTYPE_F; + adapter->params.arch.mps_tcam_size = + NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + adapter->params.arch.mps_rplc_size = 128; + adapter->params.arch.nchan = NCHAN; + adapter->params.arch.vfcount = 128; + break; + case CHELSIO_T6: + adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T6, pl_rev); + adapter->params.arch.sge_fl_db = 0; + adapter->params.arch.mps_tcam_size = + NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + adapter->params.arch.mps_rplc_size = 256; + adapter->params.arch.nchan = 2; + adapter->params.arch.vfcount = 256; break; default: dev_err(adapter->pdev_dev, "Device %d is not supported\n", @@ -5098,7 +5749,7 @@ int t4_prep_adapter(struct adapter *adapter) } /** - * cxgb4_t4_bar2_sge_qregs - return BAR2 SGE Queue register information + * t4_bar2_sge_qregs - return BAR2 SGE Queue register information * @adapter: the adapter * @qid: the Queue ID * @qtype: the Ingress or Egress type for @qid @@ -5122,7 +5773,7 @@ int t4_prep_adapter(struct adapter *adapter) * Write Combining Doorbell Buffer. If the BAR2 Queue ID is not 0, * then these "Inferred Queue ID" register may not be used. */ -int cxgb4_t4_bar2_sge_qregs(struct adapter *adapter, +int t4_bar2_sge_qregs(struct adapter *adapter, unsigned int qid, enum t4_bar2_qtype qtype, u64 *pbar2_qoffset, @@ -5223,18 +5874,19 @@ int t4_init_devlog_params(struct adapter *adap) /* Otherwise, ask the firmware for it's Device Log Parameters. */ memset(&devlog_cmd, 0, sizeof(devlog_cmd)); - devlog_cmd.op_to_write = htonl(FW_CMD_OP_V(FW_DEVLOG_CMD) | - FW_CMD_REQUEST_F | FW_CMD_READ_F); - devlog_cmd.retval_len16 = htonl(FW_LEN16(devlog_cmd)); + devlog_cmd.op_to_write = cpu_to_be32(FW_CMD_OP_V(FW_DEVLOG_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F); + devlog_cmd.retval_len16 = cpu_to_be32(FW_LEN16(devlog_cmd)); ret = t4_wr_mbox(adap, adap->mbox, &devlog_cmd, sizeof(devlog_cmd), &devlog_cmd); if (ret) return ret; - devlog_meminfo = ntohl(devlog_cmd.memtype_devlog_memaddr16_devlog); + devlog_meminfo = + be32_to_cpu(devlog_cmd.memtype_devlog_memaddr16_devlog); dparams->memtype = FW_DEVLOG_CMD_MEMTYPE_DEVLOG_G(devlog_meminfo); dparams->start = FW_DEVLOG_CMD_MEMADDR16_DEVLOG_G(devlog_meminfo) << 4; - dparams->size = ntohl(devlog_cmd.memsize_devlog); + dparams->size = be32_to_cpu(devlog_cmd.memsize_devlog); return 0; } @@ -5255,13 +5907,13 @@ int t4_init_sge_params(struct adapter *adapter) */ hps = t4_read_reg(adapter, SGE_HOST_PAGE_SIZE_A); s_hps = (HOSTPAGESIZEPF0_S + - (HOSTPAGESIZEPF1_S - HOSTPAGESIZEPF0_S) * adapter->fn); + (HOSTPAGESIZEPF1_S - HOSTPAGESIZEPF0_S) * adapter->pf); sge_params->hps = ((hps >> s_hps) & HOSTPAGESIZEPF0_M); /* Extract the SGE Egress and Ingess Queues Per Page for our PF. */ s_qpp = (QUEUESPERPAGEPF0_S + - (QUEUESPERPAGEPF1_S - QUEUESPERPAGEPF0_S) * adapter->fn); + (QUEUESPERPAGEPF1_S - QUEUESPERPAGEPF0_S) * adapter->pf); qpp = t4_read_reg(adapter, SGE_EGRESS_QUEUES_PER_PAGE_PF_A); sge_params->eq_qpp = ((qpp >> s_qpp) & QUEUESPERPAGEPF0_M); qpp = t4_read_reg(adapter, SGE_INGRESS_QUEUES_PER_PAGE_PF_A); @@ -5373,6 +6025,29 @@ int t4_filter_field_shift(const struct adapter *adap, int filter_sel) return field_shift; } +int t4_init_rss_mode(struct adapter *adap, int mbox) +{ + int i, ret; + struct fw_rss_vi_config_cmd rvc; + + memset(&rvc, 0, sizeof(rvc)); + + for_each_port(adap, i) { + struct port_info *p = adap2pinfo(adap, i); + + rvc.op_to_viid = + cpu_to_be32(FW_CMD_OP_V(FW_RSS_VI_CONFIG_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F | + FW_RSS_VI_CONFIG_CMD_VIID_V(p->viid)); + rvc.retval_len16 = cpu_to_be32(FW_LEN16(rvc)); + ret = t4_wr_mbox(adap, mbox, &rvc, sizeof(rvc), &rvc); + if (ret) + return ret; + p->rss_mode = be32_to_cpu(rvc.u.basicvirtual.defaultq_to_udpen); + } + return 0; +} + int t4_port_init(struct adapter *adap, int mbox, int pf, int vf) { u8 addr[6]; @@ -5390,10 +6065,10 @@ int t4_port_init(struct adapter *adap, int mbox, int pf, int vf) while ((adap->params.portvec & (1 << j)) == 0) j++; - c.op_to_portid = htonl(FW_CMD_OP_V(FW_PORT_CMD) | - FW_CMD_REQUEST_F | FW_CMD_READ_F | - FW_PORT_CMD_PORTID_V(j)); - c.action_to_len16 = htonl( + c.op_to_portid = cpu_to_be32(FW_CMD_OP_V(FW_PORT_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F | + FW_PORT_CMD_PORTID_V(j)); + c.action_to_len16 = cpu_to_be32( FW_PORT_CMD_ACTION_V(FW_PORT_ACTION_GET_PORT_INFO) | FW_LEN16(c)); ret = t4_wr_mbox(adap, mbox, &c, sizeof(c), &c); @@ -5411,22 +6086,23 @@ int t4_port_init(struct adapter *adap, int mbox, int pf, int vf) memcpy(adap->port[i]->dev_addr, addr, ETH_ALEN); adap->port[i]->dev_port = j; - ret = ntohl(c.u.info.lstatus_to_modtype); + ret = be32_to_cpu(c.u.info.lstatus_to_modtype); p->mdio_addr = (ret & FW_PORT_CMD_MDIOCAP_F) ? FW_PORT_CMD_MDIOADDR_G(ret) : -1; p->port_type = FW_PORT_CMD_PTYPE_G(ret); p->mod_type = FW_PORT_MOD_TYPE_NA; - rvc.op_to_viid = htonl(FW_CMD_OP_V(FW_RSS_VI_CONFIG_CMD) | - FW_CMD_REQUEST_F | FW_CMD_READ_F | - FW_RSS_VI_CONFIG_CMD_VIID(p->viid)); - rvc.retval_len16 = htonl(FW_LEN16(rvc)); + rvc.op_to_viid = + cpu_to_be32(FW_CMD_OP_V(FW_RSS_VI_CONFIG_CMD) | + FW_CMD_REQUEST_F | FW_CMD_READ_F | + FW_RSS_VI_CONFIG_CMD_VIID(p->viid)); + rvc.retval_len16 = cpu_to_be32(FW_LEN16(rvc)); ret = t4_wr_mbox(adap, mbox, &rvc, sizeof(rvc), &rvc); if (ret) return ret; - p->rss_mode = ntohl(rvc.u.basicvirtual.defaultq_to_udpen); + p->rss_mode = be32_to_cpu(rvc.u.basicvirtual.defaultq_to_udpen); - init_link_config(&p->link_cfg, ntohs(c.u.info.pcap)); + init_link_config(&p->link_cfg, be16_to_cpu(c.u.info.pcap)); j++; } return 0; @@ -5717,3 +6393,130 @@ void t4_tp_read_la(struct adapter *adap, u64 *la_buf, unsigned int *wrptr) t4_write_reg(adap, TP_DBG_LA_CONFIG_A, cfg | adap->params.tp.la_mask); } + +/* SGE Hung Ingress DMA Warning Threshold time and Warning Repeat Rate (in + * seconds). If we find one of the SGE Ingress DMA State Machines in the same + * state for more than the Warning Threshold then we'll issue a warning about + * a potential hang. We'll repeat the warning as the SGE Ingress DMA Channel + * appears to be hung every Warning Repeat second till the situation clears. + * If the situation clears, we'll note that as well. + */ +#define SGE_IDMA_WARN_THRESH 1 +#define SGE_IDMA_WARN_REPEAT 300 + +/** + * t4_idma_monitor_init - initialize SGE Ingress DMA Monitor + * @adapter: the adapter + * @idma: the adapter IDMA Monitor state + * + * Initialize the state of an SGE Ingress DMA Monitor. + */ +void t4_idma_monitor_init(struct adapter *adapter, + struct sge_idma_monitor_state *idma) +{ + /* Initialize the state variables for detecting an SGE Ingress DMA + * hang. The SGE has internal counters which count up on each clock + * tick whenever the SGE finds its Ingress DMA State Engines in the + * same state they were on the previous clock tick. The clock used is + * the Core Clock so we have a limit on the maximum "time" they can + * record; typically a very small number of seconds. For instance, + * with a 600MHz Core Clock, we can only count up to a bit more than + * 7s. So we'll synthesize a larger counter in order to not run the + * risk of having the "timers" overflow and give us the flexibility to + * maintain a Hung SGE State Machine of our own which operates across + * a longer time frame. + */ + idma->idma_1s_thresh = core_ticks_per_usec(adapter) * 1000000; /* 1s */ + idma->idma_stalled[0] = 0; + idma->idma_stalled[1] = 0; +} + +/** + * t4_idma_monitor - monitor SGE Ingress DMA state + * @adapter: the adapter + * @idma: the adapter IDMA Monitor state + * @hz: number of ticks/second + * @ticks: number of ticks since the last IDMA Monitor call + */ +void t4_idma_monitor(struct adapter *adapter, + struct sge_idma_monitor_state *idma, + int hz, int ticks) +{ + int i, idma_same_state_cnt[2]; + + /* Read the SGE Debug Ingress DMA Same State Count registers. These + * are counters inside the SGE which count up on each clock when the + * SGE finds its Ingress DMA State Engines in the same states they + * were in the previous clock. The counters will peg out at + * 0xffffffff without wrapping around so once they pass the 1s + * threshold they'll stay above that till the IDMA state changes. + */ + t4_write_reg(adapter, SGE_DEBUG_INDEX_A, 13); + idma_same_state_cnt[0] = t4_read_reg(adapter, SGE_DEBUG_DATA_HIGH_A); + idma_same_state_cnt[1] = t4_read_reg(adapter, SGE_DEBUG_DATA_LOW_A); + + for (i = 0; i < 2; i++) { + u32 debug0, debug11; + + /* If the Ingress DMA Same State Counter ("timer") is less + * than 1s, then we can reset our synthesized Stall Timer and + * continue. If we have previously emitted warnings about a + * potential stalled Ingress Queue, issue a note indicating + * that the Ingress Queue has resumed forward progress. + */ + if (idma_same_state_cnt[i] < idma->idma_1s_thresh) { + if (idma->idma_stalled[i] >= SGE_IDMA_WARN_THRESH * hz) + dev_warn(adapter->pdev_dev, "SGE idma%d, queue %u, " + "resumed after %d seconds\n", + i, idma->idma_qid[i], + idma->idma_stalled[i] / hz); + idma->idma_stalled[i] = 0; + continue; + } + + /* Synthesize an SGE Ingress DMA Same State Timer in the Hz + * domain. The first time we get here it'll be because we + * passed the 1s Threshold; each additional time it'll be + * because the RX Timer Callback is being fired on its regular + * schedule. + * + * If the stall is below our Potential Hung Ingress Queue + * Warning Threshold, continue. + */ + if (idma->idma_stalled[i] == 0) { + idma->idma_stalled[i] = hz; + idma->idma_warn[i] = 0; + } else { + idma->idma_stalled[i] += ticks; + idma->idma_warn[i] -= ticks; + } + + if (idma->idma_stalled[i] < SGE_IDMA_WARN_THRESH * hz) + continue; + + /* We'll issue a warning every SGE_IDMA_WARN_REPEAT seconds. + */ + if (idma->idma_warn[i] > 0) + continue; + idma->idma_warn[i] = SGE_IDMA_WARN_REPEAT * hz; + + /* Read and save the SGE IDMA State and Queue ID information. + * We do this every time in case it changes across time ... + * can't be too careful ... + */ + t4_write_reg(adapter, SGE_DEBUG_INDEX_A, 0); + debug0 = t4_read_reg(adapter, SGE_DEBUG_DATA_LOW_A); + idma->idma_state[i] = (debug0 >> (i * 9)) & 0x3f; + + t4_write_reg(adapter, SGE_DEBUG_INDEX_A, 11); + debug11 = t4_read_reg(adapter, SGE_DEBUG_DATA_LOW_A); + idma->idma_qid[i] = (debug11 >> (i * 16)) & 0xffff; + + dev_warn(adapter->pdev_dev, "SGE idma%u, queue %u, potentially stuck in " + "state %u for %d seconds (debug0=%#x, debug11=%#x)\n", + i, idma->idma_qid[i], idma->idma_state[i], + idma->idma_stalled[i] / hz, + debug0, debug11); + t4_sge_decode_idma_state(adapter, idma->idma_state[i]); + } +} diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h index 380b15c0417a..88067d90121c 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h @@ -152,17 +152,33 @@ struct rsp_ctrl { }; }; -#define RSPD_NEWBUF 0x80000000U -#define RSPD_LEN(x) (((x) >> 0) & 0x7fffffffU) -#define RSPD_QID(x) RSPD_LEN(x) +#define RSPD_NEWBUF_S 31 +#define RSPD_NEWBUF_V(x) ((x) << RSPD_NEWBUF_S) +#define RSPD_NEWBUF_F RSPD_NEWBUF_V(1U) -#define RSPD_GEN(x) ((x) >> 7) -#define RSPD_TYPE(x) (((x) >> 4) & 3) +#define RSPD_LEN_S 0 +#define RSPD_LEN_M 0x7fffffff +#define RSPD_LEN_G(x) (((x) >> RSPD_LEN_S) & RSPD_LEN_M) -#define V_QINTR_CNT_EN 0x0 -#define QINTR_CNT_EN 0x1 -#define QINTR_TIMER_IDX(x) ((x) << 1) -#define QINTR_TIMER_IDX_GET(x) (((x) >> 1) & 0x7) +#define RSPD_QID_S RSPD_LEN_S +#define RSPD_QID_M RSPD_LEN_M +#define RSPD_QID_G(x) RSPD_LEN_G(x) + +#define RSPD_GEN_S 7 + +#define RSPD_TYPE_S 4 +#define RSPD_TYPE_M 0x3 +#define RSPD_TYPE_G(x) (((x) >> RSPD_TYPE_S) & RSPD_TYPE_M) + +/* Rx queue interrupt deferral fields: counter enable and timer index */ +#define QINTR_CNT_EN_S 0 +#define QINTR_CNT_EN_V(x) ((x) << QINTR_CNT_EN_S) +#define QINTR_CNT_EN_F QINTR_CNT_EN_V(1U) + +#define QINTR_TIMER_IDX_S 1 +#define QINTR_TIMER_IDX_M 0x7 +#define QINTR_TIMER_IDX_V(x) ((x) << QINTR_TIMER_IDX_S) +#define QINTR_TIMER_IDX_G(x) (((x) >> QINTR_TIMER_IDX_S) & QINTR_TIMER_IDX_M) /* * Flash layout. diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h index 30a2f56e99c2..132cb8fc0bf7 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h @@ -634,26 +634,9 @@ struct cpl_tid_release { struct cpl_tx_pkt_core { __be32 ctrl0; -#define TXPKT_VF(x) ((x) << 0) -#define TXPKT_PF(x) ((x) << 8) -#define TXPKT_VF_VLD (1 << 11) -#define TXPKT_OVLAN_IDX(x) ((x) << 12) -#define TXPKT_INTF(x) ((x) << 16) -#define TXPKT_INS_OVLAN (1 << 21) -#define TXPKT_OPCODE(x) ((x) << 24) __be16 pack; __be16 len; __be64 ctrl1; -#define TXPKT_CSUM_END(x) ((x) << 12) -#define TXPKT_CSUM_START(x) ((x) << 20) -#define TXPKT_IPHDR_LEN(x) ((u64)(x) << 20) -#define TXPKT_CSUM_LOC(x) ((u64)(x) << 30) -#define TXPKT_ETHHDR_LEN(x) ((u64)(x) << 34) -#define TXPKT_CSUM_TYPE(x) ((u64)(x) << 40) -#define TXPKT_VLAN(x) ((u64)(x) << 44) -#define TXPKT_VLAN_VLD (1ULL << 60) -#define TXPKT_IPCSUM_DIS (1ULL << 62) -#define TXPKT_L4CSUM_DIS (1ULL << 63) }; struct cpl_tx_pkt { @@ -663,16 +646,69 @@ struct cpl_tx_pkt { #define cpl_tx_pkt_xt cpl_tx_pkt +/* cpl_tx_pkt_core.ctrl0 fields */ +#define TXPKT_VF_S 0 +#define TXPKT_VF_V(x) ((x) << TXPKT_VF_S) + +#define TXPKT_PF_S 8 +#define TXPKT_PF_V(x) ((x) << TXPKT_PF_S) + +#define TXPKT_VF_VLD_S 11 +#define TXPKT_VF_VLD_V(x) ((x) << TXPKT_VF_VLD_S) +#define TXPKT_VF_VLD_F TXPKT_VF_VLD_V(1U) + +#define TXPKT_OVLAN_IDX_S 12 +#define TXPKT_OVLAN_IDX_V(x) ((x) << TXPKT_OVLAN_IDX_S) + +#define TXPKT_INTF_S 16 +#define TXPKT_INTF_V(x) ((x) << TXPKT_INTF_S) + +#define TXPKT_INS_OVLAN_S 21 +#define TXPKT_INS_OVLAN_V(x) ((x) << TXPKT_INS_OVLAN_S) +#define TXPKT_INS_OVLAN_F TXPKT_INS_OVLAN_V(1U) + +#define TXPKT_OPCODE_S 24 +#define TXPKT_OPCODE_V(x) ((x) << TXPKT_OPCODE_S) + +/* cpl_tx_pkt_core.ctrl1 fields */ +#define TXPKT_CSUM_END_S 12 +#define TXPKT_CSUM_END_V(x) ((x) << TXPKT_CSUM_END_S) + +#define TXPKT_CSUM_START_S 20 +#define TXPKT_CSUM_START_V(x) ((x) << TXPKT_CSUM_START_S) + +#define TXPKT_IPHDR_LEN_S 20 +#define TXPKT_IPHDR_LEN_V(x) ((__u64)(x) << TXPKT_IPHDR_LEN_S) + +#define TXPKT_CSUM_LOC_S 30 +#define TXPKT_CSUM_LOC_V(x) ((__u64)(x) << TXPKT_CSUM_LOC_S) + +#define TXPKT_ETHHDR_LEN_S 34 +#define TXPKT_ETHHDR_LEN_V(x) ((__u64)(x) << TXPKT_ETHHDR_LEN_S) + +#define T6_TXPKT_ETHHDR_LEN_S 32 +#define T6_TXPKT_ETHHDR_LEN_V(x) ((__u64)(x) << T6_TXPKT_ETHHDR_LEN_S) + +#define TXPKT_CSUM_TYPE_S 40 +#define TXPKT_CSUM_TYPE_V(x) ((__u64)(x) << TXPKT_CSUM_TYPE_S) + +#define TXPKT_VLAN_S 44 +#define TXPKT_VLAN_V(x) ((__u64)(x) << TXPKT_VLAN_S) + +#define TXPKT_VLAN_VLD_S 60 +#define TXPKT_VLAN_VLD_V(x) ((__u64)(x) << TXPKT_VLAN_VLD_S) +#define TXPKT_VLAN_VLD_F TXPKT_VLAN_VLD_V(1ULL) + +#define TXPKT_IPCSUM_DIS_S 62 +#define TXPKT_IPCSUM_DIS_V(x) ((__u64)(x) << TXPKT_IPCSUM_DIS_S) +#define TXPKT_IPCSUM_DIS_F TXPKT_IPCSUM_DIS_V(1ULL) + +#define TXPKT_L4CSUM_DIS_S 63 +#define TXPKT_L4CSUM_DIS_V(x) ((__u64)(x) << TXPKT_L4CSUM_DIS_S) +#define TXPKT_L4CSUM_DIS_F TXPKT_L4CSUM_DIS_V(1ULL) + struct cpl_tx_pkt_lso_core { __be32 lso_ctrl; -#define LSO_TCPHDR_LEN(x) ((x) << 0) -#define LSO_IPHDR_LEN(x) ((x) << 4) -#define LSO_ETHHDR_LEN(x) ((x) << 16) -#define LSO_IPV6(x) ((x) << 20) -#define LSO_LAST_SLICE (1 << 22) -#define LSO_FIRST_SLICE (1 << 23) -#define LSO_OPCODE(x) ((x) << 24) -#define LSO_T5_XFER_SIZE(x) ((x) << 0) __be16 ipid_ofst; __be16 mss; __be32 seqno_offset; diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h index 326674b19983..c7fc3d3068f9 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h @@ -418,6 +418,20 @@ #define SGE_INGRESS_QUEUES_PER_PAGE_PF_A 0x10f4 #define SGE_INGRESS_QUEUES_PER_PAGE_VF_A 0x10f8 +#define SGE_ERROR_STATS_A 0x1100 + +#define UNCAPTURED_ERROR_S 18 +#define UNCAPTURED_ERROR_V(x) ((x) << UNCAPTURED_ERROR_S) +#define UNCAPTURED_ERROR_F UNCAPTURED_ERROR_V(1U) + +#define ERROR_QID_VALID_S 17 +#define ERROR_QID_VALID_V(x) ((x) << ERROR_QID_VALID_S) +#define ERROR_QID_VALID_F ERROR_QID_VALID_V(1U) + +#define ERROR_QID_S 0 +#define ERROR_QID_M 0x1ffffU +#define ERROR_QID_G(x) (((x) >> ERROR_QID_S) & ERROR_QID_M) + #define HP_INT_THRESH_S 28 #define HP_INT_THRESH_M 0xfU #define HP_INT_THRESH_V(x) ((x) << HP_INT_THRESH_S) @@ -705,6 +719,10 @@ #define REGISTER_S 0 #define REGISTER_V(x) ((x) << REGISTER_S) +#define T6_ENABLE_S 31 +#define T6_ENABLE_V(x) ((x) << T6_ENABLE_S) +#define T6_ENABLE_F T6_ENABLE_V(1U) + #define PFNUM_S 0 #define PFNUM_V(x) ((x) << PFNUM_S) @@ -2054,6 +2072,11 @@ #define VFLKPIDX_M 0xffU #define VFLKPIDX_G(x) (((x) >> VFLKPIDX_S) & VFLKPIDX_M) +#define T6_VFWRADDR_S 8 +#define T6_VFWRADDR_M 0xffU +#define T6_VFWRADDR_V(x) ((x) << T6_VFWRADDR_S) +#define T6_VFWRADDR_G(x) (((x) >> T6_VFWRADDR_S) & T6_VFWRADDR_M) + #define TP_RSS_CONFIG_CNG_A 0x7e04 #define TP_RSS_SECRET_KEY0_A 0x40 #define TP_RSS_PF0_CONFIG_A 0x30 @@ -2175,7 +2198,28 @@ #define MPS_RX_PERR_INT_CAUSE_A 0x11074 #define MPS_CLS_TCAM_Y_L_A 0xf000 +#define MPS_CLS_TCAM_DATA0_A 0xf000 +#define MPS_CLS_TCAM_DATA1_A 0xf004 + +#define DMACH_S 0 +#define DMACH_M 0xffffU +#define DMACH_G(x) (((x) >> DMACH_S) & DMACH_M) + #define MPS_CLS_TCAM_X_L_A 0xf008 +#define MPS_CLS_TCAM_DATA2_CTL_A 0xf008 + +#define CTLCMDTYPE_S 31 +#define CTLCMDTYPE_V(x) ((x) << CTLCMDTYPE_S) +#define CTLCMDTYPE_F CTLCMDTYPE_V(1U) + +#define CTLTCAMSEL_S 25 +#define CTLTCAMSEL_V(x) ((x) << CTLTCAMSEL_S) + +#define CTLTCAMINDEX_S 17 +#define CTLTCAMINDEX_V(x) ((x) << CTLTCAMINDEX_S) + +#define CTLXYBITSEL_S 16 +#define CTLXYBITSEL_V(x) ((x) << CTLXYBITSEL_S) #define MPS_CLS_TCAM_Y_L(idx) (MPS_CLS_TCAM_Y_L_A + (idx) * 16) #define NUM_MPS_CLS_TCAM_Y_L_INSTANCES 512 @@ -2184,6 +2228,45 @@ #define NUM_MPS_CLS_TCAM_X_L_INSTANCES 512 #define MPS_CLS_SRAM_L_A 0xe000 + +#define T6_MULTILISTEN0_S 26 + +#define T6_SRAM_PRIO3_S 23 +#define T6_SRAM_PRIO3_M 0x7U +#define T6_SRAM_PRIO3_G(x) (((x) >> T6_SRAM_PRIO3_S) & T6_SRAM_PRIO3_M) + +#define T6_SRAM_PRIO2_S 20 +#define T6_SRAM_PRIO2_M 0x7U +#define T6_SRAM_PRIO2_G(x) (((x) >> T6_SRAM_PRIO2_S) & T6_SRAM_PRIO2_M) + +#define T6_SRAM_PRIO1_S 17 +#define T6_SRAM_PRIO1_M 0x7U +#define T6_SRAM_PRIO1_G(x) (((x) >> T6_SRAM_PRIO1_S) & T6_SRAM_PRIO1_M) + +#define T6_SRAM_PRIO0_S 14 +#define T6_SRAM_PRIO0_M 0x7U +#define T6_SRAM_PRIO0_G(x) (((x) >> T6_SRAM_PRIO0_S) & T6_SRAM_PRIO0_M) + +#define T6_SRAM_VLD_S 13 +#define T6_SRAM_VLD_V(x) ((x) << T6_SRAM_VLD_S) +#define T6_SRAM_VLD_F T6_SRAM_VLD_V(1U) + +#define T6_REPLICATE_S 12 +#define T6_REPLICATE_V(x) ((x) << T6_REPLICATE_S) +#define T6_REPLICATE_F T6_REPLICATE_V(1U) + +#define T6_PF_S 9 +#define T6_PF_M 0x7U +#define T6_PF_G(x) (((x) >> T6_PF_S) & T6_PF_M) + +#define T6_VF_VALID_S 8 +#define T6_VF_VALID_V(x) ((x) << T6_VF_VALID_S) +#define T6_VF_VALID_F T6_VF_VALID_V(1U) + +#define T6_VF_S 0 +#define T6_VF_M 0xffU +#define T6_VF_G(x) (((x) >> T6_VF_S) & T6_VF_M) + #define MPS_CLS_SRAM_H_A 0xe004 #define MPS_CLS_SRAM_L(idx) (MPS_CLS_SRAM_L_A + (idx) * 8) @@ -2433,6 +2516,8 @@ #define CIM_F CIM_V(1U) #define MC1_S 31 +#define MC1_V(x) ((x) << MC1_S) +#define MC1_F MC1_V(1U) #define PL_INT_ENABLE_A 0x19410 #define PL_INT_MAP0_A 0x19414 @@ -2463,6 +2548,18 @@ #define REV_V(x) ((x) << REV_S) #define REV_G(x) (((x) >> REV_S) & REV_M) +#define T6_UNKNOWNCMD_S 3 +#define T6_UNKNOWNCMD_V(x) ((x) << T6_UNKNOWNCMD_S) +#define T6_UNKNOWNCMD_F T6_UNKNOWNCMD_V(1U) + +#define T6_LIP0_S 2 +#define T6_LIP0_V(x) ((x) << T6_LIP0_S) +#define T6_LIP0_F T6_LIP0_V(1U) + +#define T6_LIPMISS_S 1 +#define T6_LIPMISS_V(x) ((x) << T6_LIPMISS_S) +#define T6_LIPMISS_F T6_LIPMISS_V(1U) + #define LE_DB_INT_CAUSE_A 0x19c3c #define REQQPARERR_S 16 @@ -2485,6 +2582,14 @@ #define LIP0_V(x) ((x) << LIP0_S) #define LIP0_F LIP0_V(1U) +#define TCAMINTPERR_S 13 +#define TCAMINTPERR_V(x) ((x) << TCAMINTPERR_S) +#define TCAMINTPERR_F TCAMINTPERR_V(1U) + +#define SSRAMINTPERR_S 10 +#define SSRAMINTPERR_V(x) ((x) << SSRAMINTPERR_S) +#define SSRAMINTPERR_F SSRAMINTPERR_V(1U) + #define NCSI_INT_CAUSE_A 0x1a0d8 #define CIM_DM_PRTY_ERR_S 8 diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h index 19b2dcf6acde..7bdee3bf75ec 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h @@ -61,6 +61,30 @@ #define SGE_TIMERREGS 6 #define TIMERREG_COUNTER0_X 0 +#define FETCHBURSTMIN_64B_X 2 + +#define FETCHBURSTMAX_256B_X 2 +#define FETCHBURSTMAX_512B_X 3 + +#define HOSTFCMODE_STATUS_PAGE_X 2 + +#define CIDXFLUSHTHRESH_32_X 5 + +#define UPDATEDELIVERY_INTERRUPT_X 1 + +#define RSPD_TYPE_FLBUF_X 0 +#define RSPD_TYPE_CPL_X 1 +#define RSPD_TYPE_INTR_X 2 + +/* Congestion Manager Definitions. + */ +#define CONMCTXT_CNGTPMODE_S 19 +#define CONMCTXT_CNGTPMODE_V(x) ((x) << CONMCTXT_CNGTPMODE_S) +#define CONMCTXT_CNGCHMAP_S 0 +#define CONMCTXT_CNGCHMAP_V(x) ((x) << CONMCTXT_CNGCHMAP_S) +#define CONMCTXT_CNGTPMODE_CHANNEL_X 2 +#define CONMCTXT_CNGTPMODE_QUEUE_X 1 + /* T5 and later support a new BAR2-based doorbell mechanism for Egress Queues. * The User Doorbells are each 128 bytes in length with a Simple Doorbell at * offsets 8x and a Write Combining single 64-byte Egress Queue Unit diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index 03fbfd1fb3df..aceb1e8cacc8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -788,15 +788,27 @@ struct fw_ldst_cmd { __be16 vctl; __be16 rval; } mdio; - struct fw_ldst_mps { - __be16 fid_ctl; - __be16 rplcpf_pkd; - __be32 rplc127_96; - __be32 rplc95_64; - __be32 rplc63_32; - __be32 rplc31_0; - __be32 atrb; - __be16 vlan[16]; + union fw_ldst_mps { + struct fw_ldst_mps_rplc { + __be16 fid_idx; + __be16 rplcpf_pkd; + __be32 rplc255_224; + __be32 rplc223_192; + __be32 rplc191_160; + __be32 rplc159_128; + __be32 rplc127_96; + __be32 rplc95_64; + __be32 rplc63_32; + __be32 rplc31_0; + } rplc; + struct fw_ldst_mps_atrb { + __be16 fid_mpsid; + __be16 r2[3]; + __be32 r3[2]; + __be32 r4; + __be32 atrb; + __be16 vlan[16]; + } atrb; } mps; struct fw_ldst_func { u8 access_ctl; @@ -831,8 +843,8 @@ struct fw_ldst_cmd { #define FW_LDST_CMD_FID_S 15 #define FW_LDST_CMD_FID_V(x) ((x) << FW_LDST_CMD_FID_S) -#define FW_LDST_CMD_CTL_S 0 -#define FW_LDST_CMD_CTL_V(x) ((x) << FW_LDST_CMD_CTL_S) +#define FW_LDST_CMD_IDX_S 0 +#define FW_LDST_CMD_IDX_V(x) ((x) << FW_LDST_CMD_IDX_S) #define FW_LDST_CMD_RPLCPF_S 0 #define FW_LDST_CMD_RPLCPF_V(x) ((x) << FW_LDST_CMD_RPLCPF_S) @@ -1061,6 +1073,7 @@ enum fw_params_param_dev { FW_PARAMS_PARAM_DEV_FWREV = 0x0B, FW_PARAMS_PARAM_DEV_TPREV = 0x0C, FW_PARAMS_PARAM_DEV_CF = 0x0D, + FW_PARAMS_PARAM_DEV_PHYFW = 0x0F, FW_PARAMS_PARAM_DEV_DIAG = 0x11, FW_PARAMS_PARAM_DEV_MAXORDIRD_QP = 0x13, /* max supported QP IRD/ORD */ FW_PARAMS_PARAM_DEV_MAXIRD_ADAPTER = 0x14, /* max supported adap IRD */ @@ -1123,6 +1136,12 @@ enum fw_params_param_dmaq { FW_PARAMS_PARAM_DMAQ_EQ_CMPLIQID_CTRL = 0x11, FW_PARAMS_PARAM_DMAQ_EQ_SCHEDCLASS_ETH = 0x12, FW_PARAMS_PARAM_DMAQ_EQ_DCBPRIO_ETH = 0x13, + FW_PARAMS_PARAM_DMAQ_CONM_CTXT = 0x20, +}; + +enum fw_params_param_dev_phyfw { + FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD = 0x00, + FW_PARAMS_PARAM_DEV_PHYFW_VERSION = 0x01, }; enum fw_params_param_dev_diag { @@ -1377,6 +1396,7 @@ struct fw_iq_cmd { #define FW_IQ_CMD_IQFLINTCONGEN_S 27 #define FW_IQ_CMD_IQFLINTCONGEN_V(x) ((x) << FW_IQ_CMD_IQFLINTCONGEN_S) +#define FW_IQ_CMD_IQFLINTCONGEN_F FW_IQ_CMD_IQFLINTCONGEN_V(1U) #define FW_IQ_CMD_IQFLINTISCSIC_S 26 #define FW_IQ_CMD_IQFLINTISCSIC_V(x) ((x) << FW_IQ_CMD_IQFLINTISCSIC_S) @@ -1399,6 +1419,7 @@ struct fw_iq_cmd { #define FW_IQ_CMD_FL0CONGCIF_S 11 #define FW_IQ_CMD_FL0CONGCIF_V(x) ((x) << FW_IQ_CMD_FL0CONGCIF_S) +#define FW_IQ_CMD_FL0CONGCIF_F FW_IQ_CMD_FL0CONGCIF_V(1U) #define FW_IQ_CMD_FL0ONCHIP_S 10 #define FW_IQ_CMD_FL0ONCHIP_V(x) ((x) << FW_IQ_CMD_FL0ONCHIP_S) @@ -1589,6 +1610,7 @@ struct fw_eq_eth_cmd { #define FW_EQ_ETH_CMD_FETCHRO_S 22 #define FW_EQ_ETH_CMD_FETCHRO_V(x) ((x) << FW_EQ_ETH_CMD_FETCHRO_S) +#define FW_EQ_ETH_CMD_FETCHRO_F FW_EQ_ETH_CMD_FETCHRO_V(1U) #define FW_EQ_ETH_CMD_HOSTFCMODE_S 20 #define FW_EQ_ETH_CMD_HOSTFCMODE_V(x) ((x) << FW_EQ_ETH_CMD_HOSTFCMODE_S) @@ -2526,13 +2548,8 @@ enum fw_port_mod_sub_type { FW_PORT_MOD_SUB_TYPE_TWINAX_7 = 0xC, }; -/* port stats */ -#define FW_NUM_PORT_STATS 50 -#define FW_NUM_PORT_TX_STATS 23 -#define FW_NUM_PORT_RX_STATS 27 - enum fw_port_stats_tx_index { - FW_STAT_TX_PORT_BYTES_IX, + FW_STAT_TX_PORT_BYTES_IX = 0, FW_STAT_TX_PORT_FRAMES_IX, FW_STAT_TX_PORT_BCAST_IX, FW_STAT_TX_PORT_MCAST_IX, @@ -2554,11 +2571,12 @@ enum fw_port_stats_tx_index { FW_STAT_TX_PORT_PPP4_IX, FW_STAT_TX_PORT_PPP5_IX, FW_STAT_TX_PORT_PPP6_IX, - FW_STAT_TX_PORT_PPP7_IX + FW_STAT_TX_PORT_PPP7_IX, + FW_NUM_PORT_TX_STATS }; enum fw_port_stat_rx_index { - FW_STAT_RX_PORT_BYTES_IX, + FW_STAT_RX_PORT_BYTES_IX = 0, FW_STAT_RX_PORT_FRAMES_IX, FW_STAT_RX_PORT_BCAST_IX, FW_STAT_RX_PORT_MCAST_IX, @@ -2584,9 +2602,14 @@ enum fw_port_stat_rx_index { FW_STAT_RX_PORT_PPP5_IX, FW_STAT_RX_PORT_PPP6_IX, FW_STAT_RX_PORT_PPP7_IX, - FW_STAT_RX_PORT_LESS_64B_IX + FW_STAT_RX_PORT_LESS_64B_IX, + FW_STAT_RX_PORT_MAC_ERROR_IX, + FW_NUM_PORT_RX_STATS }; +/* port stats */ +#define FW_NUM_PORT_STATS (FW_NUM_PORT_TX_STATS + FW_NUM_PORT_RX_STATS) + struct fw_port_stats_cmd { __be32 op_to_portid; __be32 retval_len16; @@ -3015,7 +3038,8 @@ struct fw_hdr { enum fw_hdr_chip { FW_HDR_CHIP_T4, - FW_HDR_CHIP_T5 + FW_HDR_CHIP_T5, + FW_HDR_CHIP_T6 }; #define FW_HDR_FW_VER_MAJOR_S 24 diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h index b9d1cbac0eee..32b213559b02 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h @@ -45,4 +45,9 @@ #define T5FW_VERSION_MICRO 0x20 #define T5FW_VERSION_BUILD 0x00 +#define T6FW_VERSION_MAJOR 0x01 +#define T6FW_VERSION_MINOR 0x0D +#define T6FW_VERSION_MICRO 0x2D +#define T6FW_VERSION_BUILD 0x00 + #endif diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 1d893b0b7ddf..b2b5e5bbe04c 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -1021,7 +1021,7 @@ static int closest_thres(const struct sge *s, int thres) static unsigned int qtimer_val(const struct adapter *adapter, const struct sge_rspq *rspq) { - unsigned int timer_idx = QINTR_TIMER_IDX_GET(rspq->intr_params); + unsigned int timer_idx = QINTR_TIMER_IDX_G(rspq->intr_params); return timer_idx < SGE_NTIMERS ? adapter->sge.timer_val[timer_idx] @@ -1086,8 +1086,8 @@ static int set_rxq_intr_params(struct adapter *adapter, struct sge_rspq *rspq, * Update the response queue's interrupt coalescing parameters and * return success. */ - rspq->intr_params = (QINTR_TIMER_IDX(timer_idx) | - (cnt > 0 ? QINTR_CNT_EN : 0)); + rspq->intr_params = (QINTR_TIMER_IDX_V(timer_idx) | + QINTR_CNT_EN_V(cnt > 0)); return 0; } @@ -1439,7 +1439,7 @@ static int cxgb4vf_get_coalesce(struct net_device *dev, coalesce->rx_coalesce_usecs = qtimer_val(adapter, rspq); coalesce->rx_max_coalesced_frames = - ((rspq->intr_params & QINTR_CNT_EN) + ((rspq->intr_params & QINTR_CNT_EN_F) ? adapter->sge.counter_val[rspq->pktcnt_idx] : 0); return 0; @@ -2393,8 +2393,9 @@ static inline void init_rspq(struct sge_rspq *rspq, u8 timer_idx, u8 pkt_cnt_idx, unsigned int size, unsigned int iqe_size) { - rspq->intr_params = (QINTR_TIMER_IDX(timer_idx) | - (pkt_cnt_idx < SGE_NCOUNTERS ? QINTR_CNT_EN : 0)); + rspq->intr_params = (QINTR_TIMER_IDX_V(timer_idx) | + (pkt_cnt_idx < SGE_NCOUNTERS ? + QINTR_CNT_EN_F : 0)); rspq->pktcnt_idx = (pkt_cnt_idx < SGE_NCOUNTERS ? pkt_cnt_idx : 0); diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c index 482f6de6817d..ad53e5ad2acd 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c @@ -524,7 +524,7 @@ static void unmap_rx_buf(struct adapter *adapter, struct sge_fl *fl) */ static inline void ring_fl_db(struct adapter *adapter, struct sge_fl *fl) { - u32 val; + u32 val = adapter->params.arch.sge_fl_db; /* The SGE keeps track of its Producer and Consumer Indices in terms * of Egress Queue Units so we can only tell it about integral numbers @@ -532,11 +532,9 @@ static inline void ring_fl_db(struct adapter *adapter, struct sge_fl *fl) */ if (fl->pend_cred >= FL_PER_EQ_UNIT) { if (is_t4(adapter->params.chip)) - val = PIDX_V(fl->pend_cred / FL_PER_EQ_UNIT); + val |= PIDX_V(fl->pend_cred / FL_PER_EQ_UNIT); else - val = PIDX_T5_V(fl->pend_cred / FL_PER_EQ_UNIT) | - DBTYPE_F; - val |= DBPRIO_F; + val |= PIDX_T5_V(fl->pend_cred / FL_PER_EQ_UNIT); /* Make sure all memory writes to the Free List queue are * committed before we tell the hardware about them. @@ -1084,7 +1082,7 @@ static void inline_tx_skb(const struct sk_buff *skb, const struct sge_txq *tq, * Figure out what HW csum a packet wants and return the appropriate control * bits. */ -static u64 hwcsum(const struct sk_buff *skb) +static u64 hwcsum(enum chip_type chip, const struct sk_buff *skb) { int csum_type; const struct iphdr *iph = ip_hdr(skb); @@ -1100,7 +1098,7 @@ nocsum: * unknown protocol, disable HW csum * and hope a bad packet is detected */ - return TXPKT_L4CSUM_DIS; + return TXPKT_L4CSUM_DIS_F; } } else { /* @@ -1116,16 +1114,21 @@ nocsum: goto nocsum; } - if (likely(csum_type >= TX_CSUM_TCPIP)) - return TXPKT_CSUM_TYPE(csum_type) | - TXPKT_IPHDR_LEN(skb_network_header_len(skb)) | - TXPKT_ETHHDR_LEN(skb_network_offset(skb) - ETH_HLEN); - else { + if (likely(csum_type >= TX_CSUM_TCPIP)) { + u64 hdr_len = TXPKT_IPHDR_LEN_V(skb_network_header_len(skb)); + int eth_hdr_len = skb_network_offset(skb) - ETH_HLEN; + + if (chip <= CHELSIO_T5) + hdr_len |= TXPKT_ETHHDR_LEN_V(eth_hdr_len); + else + hdr_len |= T6_TXPKT_ETHHDR_LEN_V(eth_hdr_len); + return TXPKT_CSUM_TYPE_V(csum_type) | hdr_len; + } else { int start = skb_transport_offset(skb); - return TXPKT_CSUM_TYPE(csum_type) | - TXPKT_CSUM_START(start) | - TXPKT_CSUM_LOC(start + skb->csum_offset); + return TXPKT_CSUM_TYPE_V(csum_type) | + TXPKT_CSUM_START_V(start) | + TXPKT_CSUM_LOC_V(start + skb->csum_offset); } } @@ -1160,7 +1163,7 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) { u32 wr_mid; u64 cntrl, *end; - int qidx, credits; + int qidx, credits, max_pkt_len; unsigned int flits, ndesc; struct adapter *adapter; struct sge_eth_txq *txq; @@ -1183,6 +1186,13 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) if (unlikely(skb->len < fw_hdr_copy_len)) goto out_free; + /* Discard the packet if the length is greater than mtu */ + max_pkt_len = ETH_HLEN + dev->mtu; + if (skb_vlan_tag_present(skb)) + max_pkt_len += VLAN_HLEN; + if (!skb_shinfo(skb)->gso_size && (unlikely(skb->len > max_pkt_len))) + goto out_free; + /* * Figure out which TX Queue we're going to use. */ @@ -1281,29 +1291,35 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) * Fill in the LSO CPL message. */ lso->lso_ctrl = - cpu_to_be32(LSO_OPCODE(CPL_TX_PKT_LSO) | - LSO_FIRST_SLICE | - LSO_LAST_SLICE | - LSO_IPV6(v6) | - LSO_ETHHDR_LEN(eth_xtra_len/4) | - LSO_IPHDR_LEN(l3hdr_len/4) | - LSO_TCPHDR_LEN(tcp_hdr(skb)->doff)); + cpu_to_be32(LSO_OPCODE_V(CPL_TX_PKT_LSO) | + LSO_FIRST_SLICE_F | + LSO_LAST_SLICE_F | + LSO_IPV6_V(v6) | + LSO_ETHHDR_LEN_V(eth_xtra_len / 4) | + LSO_IPHDR_LEN_V(l3hdr_len / 4) | + LSO_TCPHDR_LEN_V(tcp_hdr(skb)->doff)); lso->ipid_ofst = cpu_to_be16(0); lso->mss = cpu_to_be16(ssi->gso_size); lso->seqno_offset = cpu_to_be32(0); if (is_t4(adapter->params.chip)) lso->len = cpu_to_be32(skb->len); else - lso->len = cpu_to_be32(LSO_T5_XFER_SIZE(skb->len)); + lso->len = cpu_to_be32(LSO_T5_XFER_SIZE_V(skb->len)); /* * Set up TX Packet CPL pointer, control word and perform * accounting. */ cpl = (void *)(lso + 1); - cntrl = (TXPKT_CSUM_TYPE(v6 ? TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | - TXPKT_IPHDR_LEN(l3hdr_len) | - TXPKT_ETHHDR_LEN(eth_xtra_len)); + + if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) + cntrl = TXPKT_ETHHDR_LEN_V(eth_xtra_len); + else + cntrl = T6_TXPKT_ETHHDR_LEN_V(eth_xtra_len); + + cntrl |= TXPKT_CSUM_TYPE_V(v6 ? + TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | + TXPKT_IPHDR_LEN_V(l3hdr_len); txq->tso++; txq->tx_cso += ssi->gso_segs; } else { @@ -1320,10 +1336,11 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) */ cpl = (void *)(wr + 1); if (skb->ip_summed == CHECKSUM_PARTIAL) { - cntrl = hwcsum(skb) | TXPKT_IPCSUM_DIS; + cntrl = hwcsum(adapter->params.chip, skb) | + TXPKT_IPCSUM_DIS_F; txq->tx_cso++; } else - cntrl = TXPKT_L4CSUM_DIS | TXPKT_IPCSUM_DIS; + cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F; } /* @@ -1332,15 +1349,15 @@ int t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) */ if (skb_vlan_tag_present(skb)) { txq->vlan_ins++; - cntrl |= TXPKT_VLAN_VLD | TXPKT_VLAN(skb_vlan_tag_get(skb)); + cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb)); } /* * Fill in the TX Packet CPL message header. */ - cpl->ctrl0 = cpu_to_be32(TXPKT_OPCODE(CPL_TX_PKT_XT) | - TXPKT_INTF(pi->port_id) | - TXPKT_PF(0)); + cpl->ctrl0 = cpu_to_be32(TXPKT_OPCODE_V(CPL_TX_PKT_XT) | + TXPKT_INTF_V(pi->port_id) | + TXPKT_PF_V(0)); cpl->pack = cpu_to_be16(0); cpl->len = cpu_to_be16(skb->len); cpl->ctrl1 = cpu_to_be64(cntrl); @@ -1663,7 +1680,7 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp, static inline bool is_new_response(const struct rsp_ctrl *rc, const struct sge_rspq *rspq) { - return RSPD_GEN(rc->type_gen) == rspq->gen; + return ((rc->type_gen >> RSPD_GEN_S) & 0x1) == rspq->gen; } /** @@ -1752,8 +1769,8 @@ static int process_responses(struct sge_rspq *rspq, int budget) * SGE. */ dma_rmb(); - rsp_type = RSPD_TYPE(rc->type_gen); - if (likely(rsp_type == RSP_TYPE_FLBUF)) { + rsp_type = RSPD_TYPE_G(rc->type_gen); + if (likely(rsp_type == RSPD_TYPE_FLBUF_X)) { struct page_frag *fp; struct pkt_gl gl; const struct rx_sw_desc *sdesc; @@ -1764,7 +1781,7 @@ static int process_responses(struct sge_rspq *rspq, int budget) * If we get a "new buffer" message from the SGE we * need to move on to the next Free List buffer. */ - if (len & RSPD_NEWBUF) { + if (len & RSPD_NEWBUF_F) { /* * We get one "new buffer" message when we * first start up a queue so we need to ignore @@ -1775,7 +1792,7 @@ static int process_responses(struct sge_rspq *rspq, int budget) 1); rspq->offset = 0; } - len = RSPD_LEN(len); + len = RSPD_LEN_G(len); } gl.tot_len = len; @@ -1818,10 +1835,10 @@ static int process_responses(struct sge_rspq *rspq, int budget) rspq->offset += ALIGN(fp->size, s->fl_align); else restore_rx_bufs(&gl, &rxq->fl, frag); - } else if (likely(rsp_type == RSP_TYPE_CPL)) { + } else if (likely(rsp_type == RSPD_TYPE_CPL_X)) { ret = rspq->handler(rspq, rspq->cur_desc, NULL); } else { - WARN_ON(rsp_type > RSP_TYPE_CPL); + WARN_ON(rsp_type > RSPD_TYPE_CPL_X); ret = 0; } @@ -1833,7 +1850,7 @@ static int process_responses(struct sge_rspq *rspq, int budget) */ const int NOMEM_TIMER_IDX = SGE_NTIMERS-1; rspq->next_intr_params = - QINTR_TIMER_IDX(NOMEM_TIMER_IDX); + QINTR_TIMER_IDX_V(NOMEM_TIMER_IDX); break; } @@ -1875,7 +1892,7 @@ static int napi_rx_handler(struct napi_struct *napi, int budget) intr_params = rspq->next_intr_params; rspq->next_intr_params = rspq->intr_params; } else - intr_params = QINTR_TIMER_IDX(SGE_TIMER_UPD_CIDX); + intr_params = QINTR_TIMER_IDX_V(SGE_TIMER_UPD_CIDX); if (unlikely(work_done == 0)) rspq->unhandled_irqs++; @@ -1936,10 +1953,10 @@ static unsigned int process_intrq(struct adapter *adapter) * never happen ... */ dma_rmb(); - if (unlikely(RSPD_TYPE(rc->type_gen) != RSP_TYPE_INTR)) { + if (unlikely(RSPD_TYPE_G(rc->type_gen) != RSPD_TYPE_INTR_X)) { dev_err(adapter->pdev_dev, "Unexpected INTRQ response type %d\n", - RSPD_TYPE(rc->type_gen)); + RSPD_TYPE_G(rc->type_gen)); continue; } @@ -1951,7 +1968,7 @@ static unsigned int process_intrq(struct adapter *adapter) * want to either make them fatal and/or conditionalized under * DEBUG. */ - qid = RSPD_QID(be32_to_cpu(rc->pldbuflen_qid)); + qid = RSPD_QID_G(be32_to_cpu(rc->pldbuflen_qid)); iq_idx = IQ_IDX(s, qid); if (unlikely(iq_idx >= MAX_INGQ)) { dev_err(adapter->pdev_dev, @@ -2154,8 +2171,8 @@ static void __iomem *bar2_address(struct adapter *adapter, u64 bar2_qoffset; int ret; - ret = t4_bar2_sge_qregs(adapter, qid, qtype, - &bar2_qoffset, pbar2_qid); + ret = t4vf_bar2_sge_qregs(adapter, qid, qtype, + &bar2_qoffset, pbar2_qid); if (ret) return NULL; @@ -2239,12 +2256,18 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq, cmd.iqaddr = cpu_to_be64(rspq->phys_addr); if (fl) { + enum chip_type chip = + CHELSIO_CHIP_VERSION(adapter->params.chip); /* * Allocate the ring for the hardware free list (with space * for its status page) along with the associated software * descriptor ring. The free list size needs to be a multiple - * of the Egress Queue Unit. + * of the Egress Queue Unit and at least 2 Egress Units larger + * than the SGE's Egress Congrestion Threshold + * (fl_starve_thres - 1). */ + if (fl->size < s->fl_starve_thres - 1 + 2 * FL_PER_EQ_UNIT) + fl->size = s->fl_starve_thres - 1 + 2 * FL_PER_EQ_UNIT; fl->size = roundup(fl->size, FL_PER_EQ_UNIT); fl->desc = alloc_ring(adapter->pdev_dev, fl->size, sizeof(__be64), sizeof(struct rx_sw_desc), @@ -2274,7 +2297,9 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq, cmd.fl0dcaen_to_fl0cidxfthresh = cpu_to_be16( FW_IQ_CMD_FL0FBMIN_V(SGE_FETCHBURSTMIN_64B) | - FW_IQ_CMD_FL0FBMAX_V(SGE_FETCHBURSTMAX_512B)); + FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ? + FETCHBURSTMAX_512B_X : + FETCHBURSTMAX_256B_X)); cmd.fl0size = cpu_to_be16(flsz); cmd.fl0addr = cpu_to_be64(fl->addr); } diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h index b9debb4f29a3..88b8981b4751 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h @@ -51,6 +51,7 @@ */ #define CHELSIO_T4 0x4 #define CHELSIO_T5 0x5 +#define CHELSIO_T6 0x6 enum chip_type { T4_A1 = CHELSIO_CHIP_CODE(CHELSIO_T4, 1), @@ -156,6 +157,12 @@ struct vpd_params { u32 cclk; /* Core Clock (KHz) */ }; +/* Stores chip specific parameters */ +struct arch_specific_params { + u32 sge_fl_db; + u16 mps_tcam_size; +}; + /* * Global Receive Side Scaling (RSS) parameters in host-native format. */ @@ -215,6 +222,7 @@ struct adapter_params { struct vpd_params vpd; /* Vital Product Data */ struct rss_params rss; /* Receive Side Scaling */ struct vf_resources vfres; /* Virtual Function Resource limits */ + struct arch_specific_params arch; /* chip specific params */ enum chip_type chip; /* chip code */ u8 nports; /* # of Ethernet "ports" */ }; @@ -284,11 +292,11 @@ int t4vf_fw_reset(struct adapter *); int t4vf_set_params(struct adapter *, unsigned int, const u32 *, const u32 *); enum t4_bar2_qtype { T4_BAR2_QTYPE_EGRESS, T4_BAR2_QTYPE_INGRESS }; -int t4_bar2_sge_qregs(struct adapter *adapter, - unsigned int qid, - enum t4_bar2_qtype qtype, - u64 *pbar2_qoffset, - unsigned int *pbar2_qid); +int t4vf_bar2_sge_qregs(struct adapter *adapter, + unsigned int qid, + enum t4_bar2_qtype qtype, + u64 *pbar2_qoffset, + unsigned int *pbar2_qid); int t4vf_get_sge_params(struct adapter *); int t4vf_get_vpd_params(struct adapter *); diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c index 966ee900ed00..0db6dc9e9ed2 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c @@ -428,7 +428,7 @@ int t4vf_set_params(struct adapter *adapter, unsigned int nparams, } /** - * t4_bar2_sge_qregs - return BAR2 SGE Queue register information + * t4vf_bar2_sge_qregs - return BAR2 SGE Queue register information * @adapter: the adapter * @qid: the Queue ID * @qtype: the Ingress or Egress type for @qid @@ -452,11 +452,11 @@ int t4vf_set_params(struct adapter *adapter, unsigned int nparams, * Write Combining Doorbell Buffer. If the BAR2 Queue ID is not 0, * then these "Inferred Queue ID" register may not be used. */ -int t4_bar2_sge_qregs(struct adapter *adapter, - unsigned int qid, - enum t4_bar2_qtype qtype, - u64 *pbar2_qoffset, - unsigned int *pbar2_qid) +int t4vf_bar2_sge_qregs(struct adapter *adapter, + unsigned int qid, + enum t4_bar2_qtype qtype, + u64 *pbar2_qoffset, + unsigned int *pbar2_qid) { unsigned int page_shift, page_size, qpp_shift, qpp_mask; u64 bar2_page_offset, bar2_qoffset; @@ -1191,9 +1191,7 @@ int t4vf_alloc_mac_filt(struct adapter *adapter, unsigned int viid, bool free, unsigned nfilters = 0; unsigned int rem = naddr; struct fw_vi_mac_cmd cmd, rpl; - unsigned int max_naddr = is_t4(adapter->params.chip) ? - NUM_MPS_CLS_SRAM_L_INSTANCES : - NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + unsigned int max_naddr = adapter->params.arch.mps_tcam_size; if (naddr > max_naddr) return -EINVAL; @@ -1285,9 +1283,7 @@ int t4vf_change_mac(struct adapter *adapter, unsigned int viid, struct fw_vi_mac_exact *p = &cmd.u.exact[0]; size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd, u.exact[1]), 16); - unsigned int max_naddr = is_t4(adapter->params.chip) ? - NUM_MPS_CLS_SRAM_L_INSTANCES : - NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + unsigned int max_mac_addr = adapter->params.arch.mps_tcam_size; /* * If this is a new allocation, determine whether it should be @@ -1310,7 +1306,7 @@ int t4vf_change_mac(struct adapter *adapter, unsigned int viid, if (ret == 0) { p = &rpl.u.exact[0]; ret = FW_VI_MAC_CMD_IDX_G(be16_to_cpu(p->valid_to_idx)); - if (ret >= max_naddr) + if (ret >= max_mac_addr) ret = -ENOMEM; } return ret; @@ -1590,11 +1586,25 @@ int t4vf_prep_adapter(struct adapter *adapter) switch (CHELSIO_PCI_ID_VER(adapter->pdev->device)) { case CHELSIO_T4: adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T4, 0); + adapter->params.arch.sge_fl_db = DBPRIO_F; + adapter->params.arch.mps_tcam_size = + NUM_MPS_CLS_SRAM_L_INSTANCES; break; case CHELSIO_T5: chipid = REV_G(t4_read_reg(adapter, PL_VF_REV_A)); adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T5, chipid); + adapter->params.arch.sge_fl_db = DBPRIO_F | DBTYPE_F; + adapter->params.arch.mps_tcam_size = + NUM_MPS_T5_CLS_SRAM_L_INSTANCES; + break; + + case CHELSIO_T6: + chipid = REV_G(t4_read_reg(adapter, PL_VF_REV_A)); + adapter->params.chip |= CHELSIO_CHIP_CODE(CHELSIO_T6, chipid); + adapter->params.arch.sge_fl_db = 0; + adapter->params.arch.mps_tcam_size = + NUM_MPS_T5_CLS_SRAM_L_INSTANCES; break; } diff --git a/drivers/net/ethernet/cisco/enic/enic_clsf.c b/drivers/net/ethernet/cisco/enic/enic_clsf.c index 0be6850be8a2..a31b57adcb37 100644 --- a/drivers/net/ethernet/cisco/enic/enic_clsf.c +++ b/drivers/net/ethernet/cisco/enic/enic_clsf.c @@ -5,7 +5,7 @@ #include <linux/in.h> #include <linux/types.h> #include <linux/skbuff.h> -#include <net/flow_keys.h> +#include <net/flow_dissector.h> #include "enic_res.h" #include "enic_clsf.h" @@ -15,14 +15,14 @@ * @rq: rq number to steer to * * This function returns filter_id(hardware_id) of the filter - * added. In case of error it returns an negative number. + * added. In case of error it returns a negative number. */ int enic_addfltr_5t(struct enic *enic, struct flow_keys *keys, u16 rq) { int res; struct filter data; - switch (keys->ip_proto) { + switch (keys->basic.ip_proto) { case IPPROTO_TCP: data.u.ipv4.protocol = PROTO_TCP; break; @@ -33,10 +33,10 @@ int enic_addfltr_5t(struct enic *enic, struct flow_keys *keys, u16 rq) return -EPROTONOSUPPORT; }; data.type = FILTER_IPV4_5TUPLE; - data.u.ipv4.src_addr = ntohl(keys->src); - data.u.ipv4.dst_addr = ntohl(keys->dst); - data.u.ipv4.src_port = ntohs(keys->port16[0]); - data.u.ipv4.dst_port = ntohs(keys->port16[1]); + data.u.ipv4.src_addr = ntohl(keys->addrs.src); + data.u.ipv4.dst_addr = ntohl(keys->addrs.dst); + data.u.ipv4.src_port = ntohs(keys->ports.src); + data.u.ipv4.dst_port = ntohs(keys->ports.dst); data.u.ipv4.flags = FILTER_FIELDS_IPV4_5TUPLE; spin_lock_bh(&enic->devcmd_lock); @@ -158,11 +158,11 @@ static struct enic_rfs_fltr_node *htbl_key_search(struct hlist_head *h, struct enic_rfs_fltr_node *tpos; hlist_for_each_entry(tpos, h, node) - if (tpos->keys.src == k->src && - tpos->keys.dst == k->dst && - tpos->keys.ports == k->ports && - tpos->keys.ip_proto == k->ip_proto && - tpos->keys.n_proto == k->n_proto) + if (tpos->keys.addrs.src == k->addrs.src && + tpos->keys.addrs.dst == k->addrs.dst && + tpos->keys.ports.ports == k->ports.ports && + tpos->keys.basic.ip_proto == k->basic.ip_proto && + tpos->keys.basic.n_proto == k->basic.n_proto) return tpos; return NULL; } @@ -177,9 +177,10 @@ int enic_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, int res, i; enic = netdev_priv(dev); - res = skb_flow_dissect(skb, &keys); - if (!res || keys.n_proto != htons(ETH_P_IP) || - (keys.ip_proto != IPPROTO_TCP && keys.ip_proto != IPPROTO_UDP)) + res = skb_flow_dissect_flow_keys(skb, &keys); + if (!res || keys.basic.n_proto != htons(ETH_P_IP) || + (keys.basic.ip_proto != IPPROTO_TCP && + keys.basic.ip_proto != IPPROTO_UDP)) return -EPROTONOSUPPORT; tbl_idx = skb_get_hash_raw(skb) & ENIC_RFS_FLW_MASK; diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c index 28d9ca675a27..117c0968dd0b 100644 --- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c +++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c @@ -334,7 +334,7 @@ static int enic_grxclsrule(struct enic *enic, struct ethtool_rxnfc *cmd) n = htbl_fltr_search(enic, (u16)fsp->location); if (!n) return -EINVAL; - switch (n->keys.ip_proto) { + switch (n->keys.basic.ip_proto) { case IPPROTO_TCP: fsp->flow_type = TCP_V4_FLOW; break; @@ -346,16 +346,16 @@ static int enic_grxclsrule(struct enic *enic, struct ethtool_rxnfc *cmd) break; } - fsp->h_u.tcp_ip4_spec.ip4src = n->keys.src; + fsp->h_u.tcp_ip4_spec.ip4src = n->keys.addrs.src; fsp->m_u.tcp_ip4_spec.ip4src = (__u32)~0; - fsp->h_u.tcp_ip4_spec.ip4dst = n->keys.dst; + fsp->h_u.tcp_ip4_spec.ip4dst = n->keys.addrs.dst; fsp->m_u.tcp_ip4_spec.ip4dst = (__u32)~0; - fsp->h_u.tcp_ip4_spec.psrc = n->keys.port16[0]; + fsp->h_u.tcp_ip4_spec.psrc = n->keys.ports.src; fsp->m_u.tcp_ip4_spec.psrc = (__u16)~0; - fsp->h_u.tcp_ip4_spec.pdst = n->keys.port16[1]; + fsp->h_u.tcp_ip4_spec.pdst = n->keys.ports.dst; fsp->m_u.tcp_ip4_spec.pdst = (__u16)~0; fsp->ring_cookie = n->rq_id; diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index 1274b6fdac8a..cf0a5fcdaaaf 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -463,10 +463,8 @@ rio_open (struct net_device *dev) dw32(MACCtrl, dr32(MACCtrl) | AutoVLANuntagging); } - init_timer (&np->timer); + setup_timer(&np->timer, rio_timer, (unsigned long)dev); np->timer.expires = jiffies + 1*HZ; - np->timer.data = (unsigned long) dev; - np->timer.function = rio_timer; add_timer (&np->timer); /* Start Tx/Rx */ diff --git a/drivers/net/ethernet/emulex/benet/Kconfig b/drivers/net/ethernet/emulex/benet/Kconfig index ea94a8eb6b35..7108563260ae 100644 --- a/drivers/net/ethernet/emulex/benet/Kconfig +++ b/drivers/net/ethernet/emulex/benet/Kconfig @@ -5,6 +5,15 @@ config BE2NET This driver implements the NIC functionality for ServerEngines' 10Gbps network adapter - BladeEngine. +config BE2NET_HWMON + bool "HWMON support for be2net driver" + depends on BE2NET && HWMON + depends on !(BE2NET=y && HWMON=m) + default y + ---help--- + Say Y here if you want to expose thermal sensor data on + be2net network adapter. + config BE2NET_VXLAN bool "VXLAN offload support on be2net driver" default y diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 1bf1cdce74ac..8d12b41b3b19 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -31,11 +31,13 @@ #include <linux/slab.h> #include <linux/u64_stats_sync.h> #include <linux/cpumask.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> #include "be_hw.h" #include "be_roce.h" -#define DRV_VER "10.6.0.1" +#define DRV_VER "10.6.0.2" #define DRV_NAME "be2net" #define BE_NAME "Emulex BladeEngine2" #define BE3_NAME "Emulex BladeEngine3" @@ -314,7 +316,6 @@ struct be_rx_obj { } ____cacheline_aligned_in_smp; struct be_drv_stats { - u32 be_on_die_temperature; u32 eth_red_drops; u32 dma_map_errors; u32 rx_drops_no_pbuf; @@ -366,6 +367,7 @@ struct be_vf_cfg { u32 tx_rate; u32 plink_tracking; u32 privileges; + bool spoofchk; }; enum vf_state { @@ -382,6 +384,7 @@ enum vf_state { #define BE_FLAGS_SETUP_DONE BIT(9) #define BE_FLAGS_EVT_INCOMPATIBLE_SFP BIT(10) #define BE_FLAGS_ERR_DETECTION_SCHEDULED BIT(11) +#define BE_FLAGS_OS2BMC BIT(12) #define BE_UC_PMAC_COUNT 30 #define BE_VF_UC_PMAC_COUNT 2 @@ -426,6 +429,8 @@ struct be_resources { u32 vf_if_cap_flags; /* VF if capability flags */ }; +#define be_is_os2bmc_enabled(adapter) (adapter->flags & BE_FLAGS_OS2BMC) + struct rss_info { u64 rss_flags; u8 rsstable[RSS_INDIR_TABLE_LEN]; @@ -433,6 +438,12 @@ struct rss_info { u8 rss_hkey[RSS_HASH_KEY_LEN]; }; +#define BE_INVALID_DIE_TEMP 0xFF +struct be_hwmon { + struct device *hwmon_dev; + u8 be_on_die_temp; /* Unit: millidegree Celsius */ +}; + /* Macros to read/write the 'features' word of be_wrb_params structure. */ #define BE_WRB_F_BIT(name) BE_WRB_F_##name##_BIT @@ -453,7 +464,8 @@ enum { BE_WRB_F_LSO_BIT, /* LSO */ BE_WRB_F_LSO6_BIT, /* LSO6 */ BE_WRB_F_VLAN_BIT, /* VLAN */ - BE_WRB_F_VLAN_SKIP_HW_BIT /* Skip VLAN tag (workaround) */ + BE_WRB_F_VLAN_SKIP_HW_BIT, /* Skip VLAN tag (workaround) */ + BE_WRB_F_OS2BMC_BIT /* Send packet to the management ring */ }; /* The structure below provides a HW-agnostic abstraction of WRB params @@ -514,6 +526,7 @@ struct be_adapter { u16 work_counter; struct delayed_work be_err_detection_work; + u8 err_flags; u32 flags; u32 cmd_privileges; /* Ethtool knobs and info */ @@ -572,8 +585,11 @@ struct be_adapter { u16 qnq_vid; u32 msg_enable; int be_get_temp_freq; + struct be_hwmon hwmon_info; u8 pf_number; struct rss_info rss_info; + /* Filters for packets that need to be sent to BMC */ + u32 bmc_filt_mask; }; #define be_physfn(adapter) (!adapter->virtfn) @@ -772,26 +788,36 @@ static inline bool is_ipv4_pkt(struct sk_buff *skb) return skb->protocol == htons(ETH_P_IP) && ip_hdr(skb)->version == 4; } -static inline bool be_multi_rxq(const struct be_adapter *adapter) +#define BE_ERROR_EEH 1 +#define BE_ERROR_UE BIT(1) +#define BE_ERROR_FW BIT(2) +#define BE_ERROR_HW (BE_ERROR_EEH | BE_ERROR_UE) +#define BE_ERROR_ANY (BE_ERROR_EEH | BE_ERROR_UE | BE_ERROR_FW) +#define BE_CLEAR_ALL 0xFF + +static inline u8 be_check_error(struct be_adapter *adapter, u32 err_type) { - return adapter->num_rx_qs > 1; + return (adapter->err_flags & err_type); } -static inline bool be_error(struct be_adapter *adapter) +static inline void be_set_error(struct be_adapter *adapter, int err_type) { - return adapter->eeh_error || adapter->hw_error || adapter->fw_timeout; + struct net_device *netdev = adapter->netdev; + + adapter->err_flags |= err_type; + netif_carrier_off(netdev); + + dev_info(&adapter->pdev->dev, "%s: Link down\n", netdev->name); } -static inline bool be_hw_error(struct be_adapter *adapter) +static inline void be_clear_error(struct be_adapter *adapter, int err_type) { - return adapter->eeh_error || adapter->hw_error; + adapter->err_flags &= ~err_type; } -static inline void be_clear_all_error(struct be_adapter *adapter) +static inline bool be_multi_rxq(const struct be_adapter *adapter) { - adapter->eeh_error = false; - adapter->hw_error = false; - adapter->fw_timeout = false; + return adapter->num_rx_qs > 1; } void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm, @@ -804,6 +830,7 @@ bool be_pause_supported(struct be_adapter *adapter); u32 be_get_fw_log_level(struct be_adapter *adapter); int be_update_queues(struct be_adapter *adapter); int be_poll(struct napi_struct *napi, int budget); +void be_eqd_update(struct be_adapter *adapter, bool force_update); /* * internal function to initialize-cleanup roce device. diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index fb140faeafb1..41150543906a 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -93,7 +93,7 @@ static void be_mcc_notify(struct be_adapter *adapter) struct be_queue_info *mccq = &adapter->mcc_obj.q; u32 val = 0; - if (be_error(adapter)) + if (be_check_error(adapter, BE_ERROR_ANY)) return; val |= mccq->id & DB_MCCQ_RING_ID_MASK; @@ -140,6 +140,7 @@ static bool be_skip_err_log(u8 opcode, u16 base_status, u16 addl_status) if (base_status == MCC_STATUS_NOT_SUPPORTED || base_status == MCC_STATUS_ILLEGAL_REQUEST || addl_status == MCC_ADDL_STATUS_TOO_MANY_INTERFACES || + addl_status == MCC_ADDL_STATUS_INSUFFICIENT_VLANS || (opcode == OPCODE_COMMON_WRITE_FLASHROM && (base_status == MCC_STATUS_ILLEGAL_FIELD || addl_status == MCC_ADDL_STATUS_FLASH_IMAGE_CRC_MISMATCH))) @@ -191,10 +192,12 @@ static void be_async_cmd_process(struct be_adapter *adapter, if (base_status == MCC_STATUS_SUCCESS) { struct be_cmd_resp_get_cntl_addnl_attribs *resp = (void *)resp_hdr; - adapter->drv_stats.be_on_die_temperature = + adapter->hwmon_info.be_on_die_temp = resp->on_die_temperature; } else { adapter->be_get_temp_freq = 0; + adapter->hwmon_info.be_on_die_temp = + BE_INVALID_DIE_TEMP; } return; } @@ -330,6 +333,21 @@ static void be_async_grp5_pvid_state_process(struct be_adapter *adapter, } } +#define MGMT_ENABLE_MASK 0x4 +static void be_async_grp5_fw_control_process(struct be_adapter *adapter, + struct be_mcc_compl *compl) +{ + struct be_async_fw_control *evt = (struct be_async_fw_control *)compl; + u32 evt_dw1 = le32_to_cpu(evt->event_data_word1); + + if (evt_dw1 & MGMT_ENABLE_MASK) { + adapter->flags |= BE_FLAGS_OS2BMC; + adapter->bmc_filt_mask = le32_to_cpu(evt->event_data_word2); + } else { + adapter->flags &= ~BE_FLAGS_OS2BMC; + } +} + static void be_async_grp5_evt_process(struct be_adapter *adapter, struct be_mcc_compl *compl) { @@ -346,6 +364,10 @@ static void be_async_grp5_evt_process(struct be_adapter *adapter, case ASYNC_EVENT_PVID_STATE: be_async_grp5_pvid_state_process(adapter, compl); break; + /* Async event to disable/enable os2bmc and/or mac-learning */ + case ASYNC_EVENT_FW_CONTROL: + be_async_grp5_fw_control_process(adapter, compl); + break; default: break; } @@ -486,7 +508,7 @@ static int be_mcc_wait_compl(struct be_adapter *adapter) struct be_mcc_obj *mcc_obj = &adapter->mcc_obj; for (i = 0; i < mcc_timeout; i++) { - if (be_error(adapter)) + if (be_check_error(adapter, BE_ERROR_ANY)) return -EIO; local_bh_disable(); @@ -499,7 +521,7 @@ static int be_mcc_wait_compl(struct be_adapter *adapter) } if (i == mcc_timeout) { dev_err(&adapter->pdev->dev, "FW not responding\n"); - adapter->fw_timeout = true; + be_set_error(adapter, BE_ERROR_FW); return -EIO; } return status; @@ -538,7 +560,7 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) u32 ready; do { - if (be_error(adapter)) + if (be_check_error(adapter, BE_ERROR_ANY)) return -EIO; ready = ioread32(db); @@ -551,7 +573,7 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) if (msecs > 4000) { dev_err(&adapter->pdev->dev, "FW not responding\n"); - adapter->fw_timeout = true; + be_set_error(adapter, BE_ERROR_FW); be_detect_error(adapter); return -1; } @@ -1457,7 +1479,7 @@ int be_cmd_if_create(struct be_adapter *adapter, u32 cap_flags, u32 en_flags, *if_handle = le32_to_cpu(resp->interface_id); /* Hack to retrieve VF's pmac-id on BE3 */ - if (BE3_chip(adapter) && !be_physfn(adapter)) + if (BE3_chip(adapter) && be_virtfn(adapter)) adapter->pmac_id[0] = le32_to_cpu(resp->pmac_id); } return status; @@ -3153,7 +3175,7 @@ int be_cmd_set_mac(struct be_adapter *adapter, u8 *mac, int if_id, u32 dom) } int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, - u32 domain, u16 intf_id, u16 hsw_mode) + u32 domain, u16 intf_id, u16 hsw_mode, u8 spoofchk) { struct be_mcc_wrb *wrb; struct be_cmd_req_set_hsw_config *req; @@ -3189,6 +3211,14 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, ctxt, hsw_mode); } + /* Enable/disable both mac and vlan spoof checking */ + if (!BEx_chip(adapter) && spoofchk) { + AMAP_SET_BITS(struct amap_set_hsw_context, mac_spoofchk, + ctxt, spoofchk); + AMAP_SET_BITS(struct amap_set_hsw_context, vlan_spoofchk, + ctxt, spoofchk); + } + be_dws_cpu_to_le(req->context, sizeof(req->context)); status = be_mcc_notify_wait(adapter); @@ -3199,7 +3229,7 @@ err: /* Get Hyper switch config */ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid, - u32 domain, u16 intf_id, u8 *mode) + u32 domain, u16 intf_id, u8 *mode, bool *spoofchk) { struct be_mcc_wrb *wrb; struct be_cmd_req_get_hsw_config *req; @@ -3247,6 +3277,10 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid, if (mode) *mode = AMAP_GET_BITS(struct amap_get_hsw_resp_context, port_fwd_type, &resp->context); + if (spoofchk) + *spoofchk = + AMAP_GET_BITS(struct amap_get_hsw_resp_context, + spoofchk, &resp->context); } err: @@ -3258,7 +3292,7 @@ static bool be_is_wol_excluded(struct be_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; - if (!be_physfn(adapter)) + if (be_virtfn(adapter)) return true; switch (pdev->subsystem_device) { diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index 1ec22300e254..2716e6f30d9a 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -65,7 +65,8 @@ enum mcc_base_status { enum mcc_addl_status { MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES = 0x16, MCC_ADDL_STATUS_FLASH_IMAGE_CRC_MISMATCH = 0x4d, - MCC_ADDL_STATUS_TOO_MANY_INTERFACES = 0x4a + MCC_ADDL_STATUS_TOO_MANY_INTERFACES = 0x4a, + MCC_ADDL_STATUS_INSUFFICIENT_VLANS = 0xab }; #define CQE_BASE_STATUS_MASK 0xFFFF @@ -104,6 +105,7 @@ struct be_mcc_compl { #define ASYNC_DEBUG_EVENT_TYPE_QNQ 1 #define ASYNC_EVENT_CODE_SLIPORT 0x11 #define ASYNC_EVENT_PORT_MISCONFIG 0x9 +#define ASYNC_EVENT_FW_CONTROL 0x5 enum { LINK_DOWN = 0x0, @@ -180,6 +182,22 @@ struct be_async_event_misconfig_port { u32 flags; } __packed; +#define BMC_FILT_BROADCAST_ARP BIT(0) +#define BMC_FILT_BROADCAST_DHCP_CLIENT BIT(1) +#define BMC_FILT_BROADCAST_DHCP_SERVER BIT(2) +#define BMC_FILT_BROADCAST_NET_BIOS BIT(3) +#define BMC_FILT_BROADCAST BIT(7) +#define BMC_FILT_MULTICAST_IPV6_NEIGH_ADVER BIT(8) +#define BMC_FILT_MULTICAST_IPV6_RA BIT(9) +#define BMC_FILT_MULTICAST_IPV6_RAS BIT(10) +#define BMC_FILT_MULTICAST BIT(15) +struct be_async_fw_control { + u32 event_data_word1; + u32 event_data_word2; + u32 evt_tag; + u32 event_data_word4; +} __packed; + struct be_mcc_mailbox { struct be_mcc_wrb wrb; struct be_mcc_compl compl; @@ -1109,10 +1127,6 @@ struct be_cmd_req_query_fw_cfg { u32 rsvd[31]; }; -/* ASIC revisions */ -#define ASIC_REV_B0 0x10 -#define ASIC_REV_P2 0x11 - struct be_cmd_resp_query_fw_cfg { struct be_cmd_resp_hdr hdr; u32 be_config_number; @@ -1745,18 +1759,24 @@ struct be_cmd_req_set_mac_list { #define PORT_FWD_TYPE_VEPA 0x3 #define PORT_FWD_TYPE_VEB 0x2 +#define ENABLE_MAC_SPOOFCHK 0x2 +#define DISABLE_MAC_SPOOFCHK 0x3 + struct amap_set_hsw_context { u8 interface_id[16]; - u8 rsvd0[14]; + u8 rsvd0[8]; + u8 mac_spoofchk[2]; + u8 rsvd1[4]; u8 pvid_valid; u8 pport; - u8 rsvd1[6]; + u8 rsvd2[6]; u8 port_fwd_type[3]; - u8 rsvd2[7]; + u8 rsvd3[5]; + u8 vlan_spoofchk[2]; u8 pvid[16]; - u8 rsvd3[32]; u8 rsvd4[32]; u8 rsvd5[32]; + u8 rsvd6[32]; } __packed; struct be_cmd_req_set_hsw_config { @@ -1774,11 +1794,13 @@ struct amap_get_hsw_req_context { struct amap_get_hsw_resp_context { u8 rsvd0[6]; u8 port_fwd_type[3]; - u8 rsvd1[7]; + u8 rsvd1[5]; + u8 spoofchk; + u8 rsvd2; u8 pvid[16]; - u8 rsvd2[32]; u8 rsvd3[32]; u8 rsvd4[32]; + u8 rsvd5[32]; } __packed; struct be_cmd_req_get_hsw_config { @@ -2334,9 +2356,9 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array, u8 mac_count, u32 domain); int be_cmd_set_mac(struct be_adapter *adapter, u8 *mac, int if_id, u32 dom); int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, u32 domain, - u16 intf_id, u16 hsw_mode); + u16 intf_id, u16 hsw_mode, u8 spoofchk); int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid, u32 domain, - u16 intf_id, u8 *mode); + u16 intf_id, u8 *mode, bool *spoofchk); int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter); int be_cmd_set_fw_log_level(struct be_adapter *adapter, u32 level); int be_cmd_get_fw_log_level(struct be_adapter *adapter); diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index b765c24625bf..675cbacef772 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -123,7 +123,6 @@ static const struct be_ethtool_stat et_stats[] = { {DRVSTAT_INFO(dma_map_errors)}, /* Number of packets dropped due to random early drop function */ {DRVSTAT_INFO(eth_red_drops)}, - {DRVSTAT_INFO(be_on_die_temperature)}, {DRVSTAT_INFO(rx_roce_bytes_lsd)}, {DRVSTAT_INFO(rx_roce_bytes_msd)}, {DRVSTAT_INFO(rx_roce_frames)}, @@ -368,6 +367,14 @@ static int be_set_coalesce(struct net_device *netdev, aic++; } + /* For Skyhawk, the EQD setting happens via EQ_DB when AIC is enabled. + * When AIC is disabled, persistently force set EQD value via the + * FW cmd, so that we don't have to calculate the delay multiplier + * encode value each time EQ_DB is rung + */ + if (!et->use_adaptive_rx_coalesce && skyhawk_chip(adapter)) + be_eqd_update(adapter, true); + return 0; } diff --git a/drivers/net/ethernet/emulex/benet/be_hw.h b/drivers/net/ethernet/emulex/benet/be_hw.h index 48840889db62..c684bb32b487 100644 --- a/drivers/net/ethernet/emulex/benet/be_hw.h +++ b/drivers/net/ethernet/emulex/benet/be_hw.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -132,6 +132,18 @@ #define DB_EQ_NUM_POPPED_SHIFT (16) /* bits 16 - 28 */ /* Rearm bit */ #define DB_EQ_REARM_SHIFT (29) /* bit 29 */ +/* Rearm to interrupt delay encoding */ +#define DB_EQ_R2I_DLY_SHIFT (30) /* bits 30 - 31 */ + +/* Rearm to interrupt (R2I) delay multiplier encoding represents 3 different + * values configured in CEV_REARM2IRPT_DLY_MULT_CSR register. This value is + * programmed by host driver while ringing an EQ doorbell(EQ_DB) if a delay + * between rearming the EQ and next interrupt on this EQ is desired. + */ +#define R2I_DLY_ENC_0 0 /* No delay */ +#define R2I_DLY_ENC_1 1 /* maps to 160us EQ delay */ +#define R2I_DLY_ENC_2 2 /* maps to 96us EQ delay */ +#define R2I_DLY_ENC_3 3 /* maps to 48us EQ delay */ /********* Compl Q door bell *************/ #define DB_CQ_OFFSET 0x120 diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 6f9ffb9026cd..1365a56f78df 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -179,7 +179,7 @@ static void be_intr_set(struct be_adapter *adapter, bool enable) if (lancer_chip(adapter)) return; - if (adapter->eeh_error) + if (be_check_error(adapter, BE_ERROR_EEH)) return; status = be_cmd_intr_set(adapter, enable); @@ -191,6 +191,9 @@ static void be_rxq_notify(struct be_adapter *adapter, u16 qid, u16 posted) { u32 val = 0; + if (be_check_error(adapter, BE_ERROR_HW)) + return; + val |= qid & DB_RQ_RING_ID_MASK; val |= posted << DB_RQ_NUM_POSTED_SHIFT; @@ -203,6 +206,9 @@ static void be_txq_notify(struct be_adapter *adapter, struct be_tx_obj *txo, { u32 val = 0; + if (be_check_error(adapter, BE_ERROR_HW)) + return; + val |= txo->q.id & DB_TXULP_RING_ID_MASK; val |= (posted & DB_TXULP_NUM_POSTED_MASK) << DB_TXULP_NUM_POSTED_SHIFT; @@ -211,14 +217,15 @@ static void be_txq_notify(struct be_adapter *adapter, struct be_tx_obj *txo, } static void be_eq_notify(struct be_adapter *adapter, u16 qid, - bool arm, bool clear_int, u16 num_popped) + bool arm, bool clear_int, u16 num_popped, + u32 eq_delay_mult_enc) { u32 val = 0; val |= qid & DB_EQ_RING_ID_MASK; val |= ((qid & DB_EQ_RING_ID_EXT_MASK) << DB_EQ_RING_ID_EXT_MASK_SHIFT); - if (adapter->eeh_error) + if (be_check_error(adapter, BE_ERROR_HW)) return; if (arm) @@ -227,6 +234,7 @@ static void be_eq_notify(struct be_adapter *adapter, u16 qid, val |= 1 << DB_EQ_CLR_SHIFT; val |= 1 << DB_EQ_EVNT_SHIFT; val |= num_popped << DB_EQ_NUM_POPPED_SHIFT; + val |= eq_delay_mult_enc << DB_EQ_R2I_DLY_SHIFT; iowrite32(val, adapter->db + DB_EQ_OFFSET); } @@ -238,7 +246,7 @@ void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm, u16 num_popped) val |= ((qid & DB_CQ_RING_ID_EXT_MASK) << DB_CQ_RING_ID_EXT_MASK_SHIFT); - if (adapter->eeh_error) + if (be_check_error(adapter, BE_ERROR_HW)) return; if (arm) @@ -662,6 +670,8 @@ void be_link_status_update(struct be_adapter *adapter, u8 link_status) netif_carrier_on(netdev); else netif_carrier_off(netdev); + + netdev_info(netdev, "Link is %s\n", link_status ? "Up" : "Down"); } static void be_tx_stats_update(struct be_tx_obj *txo, struct sk_buff *skb) @@ -810,6 +820,8 @@ static void wrb_fill_hdr(struct be_adapter *adapter, SET_TX_WRB_HDR_BITS(num_wrb, hdr, skb_wrb_cnt(skb)); SET_TX_WRB_HDR_BITS(len, hdr, skb->len); + SET_TX_WRB_HDR_BITS(mgmt, hdr, + BE_WRB_F_GET(wrb_params->features, OS2BMC)); } static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb, @@ -1146,6 +1158,130 @@ static void be_xmit_flush(struct be_adapter *adapter, struct be_tx_obj *txo) txo->pend_wrb_cnt = 0; } +/* OS2BMC related */ + +#define DHCP_CLIENT_PORT 68 +#define DHCP_SERVER_PORT 67 +#define NET_BIOS_PORT1 137 +#define NET_BIOS_PORT2 138 +#define DHCPV6_RAS_PORT 547 + +#define is_mc_allowed_on_bmc(adapter, eh) \ + (!is_multicast_filt_enabled(adapter) && \ + is_multicast_ether_addr(eh->h_dest) && \ + !is_broadcast_ether_addr(eh->h_dest)) + +#define is_bc_allowed_on_bmc(adapter, eh) \ + (!is_broadcast_filt_enabled(adapter) && \ + is_broadcast_ether_addr(eh->h_dest)) + +#define is_arp_allowed_on_bmc(adapter, skb) \ + (is_arp(skb) && is_arp_filt_enabled(adapter)) + +#define is_broadcast_packet(eh, adapter) \ + (is_multicast_ether_addr(eh->h_dest) && \ + !compare_ether_addr(eh->h_dest, adapter->netdev->broadcast)) + +#define is_arp(skb) (skb->protocol == htons(ETH_P_ARP)) + +#define is_arp_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & (BMC_FILT_BROADCAST_ARP)) + +#define is_dhcp_client_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_BROADCAST_DHCP_CLIENT) + +#define is_dhcp_srvr_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_BROADCAST_DHCP_SERVER) + +#define is_nbios_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_BROADCAST_NET_BIOS) + +#define is_ipv6_na_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & \ + BMC_FILT_MULTICAST_IPV6_NEIGH_ADVER) + +#define is_ipv6_ra_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_MULTICAST_IPV6_RA) + +#define is_ipv6_ras_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_MULTICAST_IPV6_RAS) + +#define is_broadcast_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_BROADCAST) + +#define is_multicast_filt_enabled(adapter) \ + (adapter->bmc_filt_mask & BMC_FILT_MULTICAST) + +static bool be_send_pkt_to_bmc(struct be_adapter *adapter, + struct sk_buff **skb) +{ + struct ethhdr *eh = (struct ethhdr *)(*skb)->data; + bool os2bmc = false; + + if (!be_is_os2bmc_enabled(adapter)) + goto done; + + if (!is_multicast_ether_addr(eh->h_dest)) + goto done; + + if (is_mc_allowed_on_bmc(adapter, eh) || + is_bc_allowed_on_bmc(adapter, eh) || + is_arp_allowed_on_bmc(adapter, (*skb))) { + os2bmc = true; + goto done; + } + + if ((*skb)->protocol == htons(ETH_P_IPV6)) { + struct ipv6hdr *hdr = ipv6_hdr((*skb)); + u8 nexthdr = hdr->nexthdr; + + if (nexthdr == IPPROTO_ICMPV6) { + struct icmp6hdr *icmp6 = icmp6_hdr((*skb)); + + switch (icmp6->icmp6_type) { + case NDISC_ROUTER_ADVERTISEMENT: + os2bmc = is_ipv6_ra_filt_enabled(adapter); + goto done; + case NDISC_NEIGHBOUR_ADVERTISEMENT: + os2bmc = is_ipv6_na_filt_enabled(adapter); + goto done; + default: + break; + } + } + } + + if (is_udp_pkt((*skb))) { + struct udphdr *udp = udp_hdr((*skb)); + + switch (udp->dest) { + case DHCP_CLIENT_PORT: + os2bmc = is_dhcp_client_filt_enabled(adapter); + goto done; + case DHCP_SERVER_PORT: + os2bmc = is_dhcp_srvr_filt_enabled(adapter); + goto done; + case NET_BIOS_PORT1: + case NET_BIOS_PORT2: + os2bmc = is_nbios_filt_enabled(adapter); + goto done; + case DHCPV6_RAS_PORT: + os2bmc = is_ipv6_ras_filt_enabled(adapter); + goto done; + default: + break; + } + } +done: + /* For packets over a vlan, which are destined + * to BMC, asic expects the vlan to be inline in the packet. + */ + if (os2bmc) + *skb = be_insert_vlan_in_pkt(adapter, *skb, NULL); + + return os2bmc; +} + static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev) { struct be_adapter *adapter = netdev_priv(netdev); @@ -1167,6 +1303,18 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev) goto drop; } + /* if os2bmc is enabled and if the pkt is destined to bmc, + * enqueue the pkt a 2nd time with mgmt bit set. + */ + if (be_send_pkt_to_bmc(adapter, &skb)) { + BE_WRB_F_SET(wrb_params.features, OS2BMC, 1); + wrb_cnt = be_xmit_enqueue(adapter, txo, skb, &wrb_params); + if (unlikely(!wrb_cnt)) + goto drop; + else + skb_get(skb); + } + if (be_is_txq_full(txo)) { netif_stop_subqueue(netdev, q_idx); tx_stats(txo)->tx_stops++; @@ -1265,7 +1413,8 @@ static int be_vid_config(struct be_adapter *adapter) if (status) { dev_err(dev, "Setting HW VLAN filtering failed\n"); /* Set to VLAN promisc mode as setting VLAN filter failed */ - if (addl_status(status) == + if (addl_status(status) == MCC_ADDL_STATUS_INSUFFICIENT_VLANS || + addl_status(status) == MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES) return be_set_vlan_promisc(adapter); } else if (adapter->if_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { @@ -1466,6 +1615,7 @@ static int be_get_vf_config(struct net_device *netdev, int vf, vi->qos = vf_cfg->vlan_tag >> VLAN_PRIO_SHIFT; memcpy(&vi->mac, vf_cfg->mac_addr, ETH_ALEN); vi->linkstate = adapter->vf_cfg[vf].plink_tracking; + vi->spoofchk = adapter->vf_cfg[vf].spoofchk; return 0; } @@ -1478,7 +1628,7 @@ static int be_set_vf_tvt(struct be_adapter *adapter, int vf, u16 vlan) int status; /* Enable Transparent VLAN Tagging */ - status = be_cmd_set_hsw_config(adapter, vlan, vf + 1, vf_if_id, 0); + status = be_cmd_set_hsw_config(adapter, vlan, vf + 1, vf_if_id, 0, 0); if (status) return status; @@ -1507,7 +1657,7 @@ static int be_clear_vf_tvt(struct be_adapter *adapter, int vf) /* Reset Transparent VLAN Tagging. */ status = be_cmd_set_hsw_config(adapter, BE_RESET_VLAN_TAG_ID, vf + 1, - vf_cfg->if_handle, 0); + vf_cfg->if_handle, 0, 0); if (status) return status; @@ -1642,6 +1792,39 @@ static int be_set_vf_link_state(struct net_device *netdev, int vf, return 0; } +static int be_set_vf_spoofchk(struct net_device *netdev, int vf, bool enable) +{ + struct be_adapter *adapter = netdev_priv(netdev); + struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf]; + u8 spoofchk; + int status; + + if (!sriov_enabled(adapter)) + return -EPERM; + + if (vf >= adapter->num_vfs) + return -EINVAL; + + if (BEx_chip(adapter)) + return -EOPNOTSUPP; + + if (enable == vf_cfg->spoofchk) + return 0; + + spoofchk = enable ? ENABLE_MAC_SPOOFCHK : DISABLE_MAC_SPOOFCHK; + + status = be_cmd_set_hsw_config(adapter, 0, vf + 1, vf_cfg->if_handle, + 0, spoofchk); + if (status) { + dev_err(&adapter->pdev->dev, + "Spoofchk change on VF %d failed: %#x\n", vf, status); + return be_cmd_status(status); + } + + vf_cfg->spoofchk = enable; + return 0; +} + static void be_aic_update(struct be_aic_obj *aic, u64 rx_pkts, u64 tx_pkts, ulong now) { @@ -1650,61 +1833,110 @@ static void be_aic_update(struct be_aic_obj *aic, u64 rx_pkts, u64 tx_pkts, aic->jiffies = now; } -static void be_eqd_update(struct be_adapter *adapter) +static int be_get_new_eqd(struct be_eq_obj *eqo) { - struct be_set_eqd set_eqd[MAX_EVT_QS]; - int eqd, i, num = 0, start; + struct be_adapter *adapter = eqo->adapter; + int eqd, start; struct be_aic_obj *aic; - struct be_eq_obj *eqo; struct be_rx_obj *rxo; struct be_tx_obj *txo; - u64 rx_pkts, tx_pkts; + u64 rx_pkts = 0, tx_pkts = 0; ulong now; u32 pps, delta; + int i; - for_all_evt_queues(adapter, eqo, i) { - aic = &adapter->aic_obj[eqo->idx]; - if (!aic->enable) { - if (aic->jiffies) - aic->jiffies = 0; - eqd = aic->et_eqd; - goto modify_eqd; - } + aic = &adapter->aic_obj[eqo->idx]; + if (!aic->enable) { + if (aic->jiffies) + aic->jiffies = 0; + eqd = aic->et_eqd; + return eqd; + } - rxo = &adapter->rx_obj[eqo->idx]; + for_all_rx_queues_on_eq(adapter, eqo, rxo, i) { do { start = u64_stats_fetch_begin_irq(&rxo->stats.sync); - rx_pkts = rxo->stats.rx_pkts; + rx_pkts += rxo->stats.rx_pkts; } while (u64_stats_fetch_retry_irq(&rxo->stats.sync, start)); + } - txo = &adapter->tx_obj[eqo->idx]; + for_all_tx_queues_on_eq(adapter, eqo, txo, i) { do { start = u64_stats_fetch_begin_irq(&txo->stats.sync); - tx_pkts = txo->stats.tx_reqs; + tx_pkts += txo->stats.tx_reqs; } while (u64_stats_fetch_retry_irq(&txo->stats.sync, start)); + } - /* Skip, if wrapped around or first calculation */ - now = jiffies; - if (!aic->jiffies || time_before(now, aic->jiffies) || - rx_pkts < aic->rx_pkts_prev || - tx_pkts < aic->tx_reqs_prev) { - be_aic_update(aic, rx_pkts, tx_pkts, now); - continue; - } + /* Skip, if wrapped around or first calculation */ + now = jiffies; + if (!aic->jiffies || time_before(now, aic->jiffies) || + rx_pkts < aic->rx_pkts_prev || + tx_pkts < aic->tx_reqs_prev) { + be_aic_update(aic, rx_pkts, tx_pkts, now); + return aic->prev_eqd; + } - delta = jiffies_to_msecs(now - aic->jiffies); - pps = (((u32)(rx_pkts - aic->rx_pkts_prev) * 1000) / delta) + - (((u32)(tx_pkts - aic->tx_reqs_prev) * 1000) / delta); - eqd = (pps / 15000) << 2; + delta = jiffies_to_msecs(now - aic->jiffies); + if (delta == 0) + return aic->prev_eqd; - if (eqd < 8) - eqd = 0; - eqd = min_t(u32, eqd, aic->max_eqd); - eqd = max_t(u32, eqd, aic->min_eqd); + pps = (((u32)(rx_pkts - aic->rx_pkts_prev) * 1000) / delta) + + (((u32)(tx_pkts - aic->tx_reqs_prev) * 1000) / delta); + eqd = (pps / 15000) << 2; - be_aic_update(aic, rx_pkts, tx_pkts, now); -modify_eqd: - if (eqd != aic->prev_eqd) { + if (eqd < 8) + eqd = 0; + eqd = min_t(u32, eqd, aic->max_eqd); + eqd = max_t(u32, eqd, aic->min_eqd); + + be_aic_update(aic, rx_pkts, tx_pkts, now); + + return eqd; +} + +/* For Skyhawk-R only */ +static u32 be_get_eq_delay_mult_enc(struct be_eq_obj *eqo) +{ + struct be_adapter *adapter = eqo->adapter; + struct be_aic_obj *aic = &adapter->aic_obj[eqo->idx]; + ulong now = jiffies; + int eqd; + u32 mult_enc; + + if (!aic->enable) + return 0; + + if (time_before_eq(now, aic->jiffies) || + jiffies_to_msecs(now - aic->jiffies) < 1) + eqd = aic->prev_eqd; + else + eqd = be_get_new_eqd(eqo); + + if (eqd > 100) + mult_enc = R2I_DLY_ENC_1; + else if (eqd > 60) + mult_enc = R2I_DLY_ENC_2; + else if (eqd > 20) + mult_enc = R2I_DLY_ENC_3; + else + mult_enc = R2I_DLY_ENC_0; + + aic->prev_eqd = eqd; + + return mult_enc; +} + +void be_eqd_update(struct be_adapter *adapter, bool force_update) +{ + struct be_set_eqd set_eqd[MAX_EVT_QS]; + struct be_aic_obj *aic; + struct be_eq_obj *eqo; + int i, num = 0, eqd; + + for_all_evt_queues(adapter, eqo, i) { + aic = &adapter->aic_obj[eqo->idx]; + eqd = be_get_new_eqd(eqo); + if (force_update || eqd != aic->prev_eqd) { set_eqd[num].delay_multiplier = (eqd * 65)/100; set_eqd[num].eq_id = eqo->q.id; aic->prev_eqd = eqd; @@ -2212,7 +2444,7 @@ static void be_eq_clean(struct be_eq_obj *eqo) { int num = events_get(eqo); - be_eq_notify(eqo->adapter, eqo->q.id, false, true, num); + be_eq_notify(eqo->adapter, eqo->q.id, false, true, num, 0); } static void be_rx_cq_clean(struct be_rx_obj *rxo) @@ -2236,7 +2468,9 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo) if (lancer_chip(adapter)) break; - if (flush_wait++ > 10 || be_hw_error(adapter)) { + if (flush_wait++ > 50 || + be_check_error(adapter, + BE_ERROR_HW)) { dev_warn(&adapter->pdev->dev, "did not receive flush compl\n"); break; @@ -2297,7 +2531,8 @@ static void be_tx_compl_clean(struct be_adapter *adapter) pending_txqs--; } - if (pending_txqs == 0 || ++timeo > 10 || be_hw_error(adapter)) + if (pending_txqs == 0 || ++timeo > 10 || + be_check_error(adapter, BE_ERROR_HW)) break; mdelay(1); @@ -2573,7 +2808,7 @@ static irqreturn_t be_intx(int irq, void *dev) if (num_evts) eqo->spurious_intr = 0; } - be_eq_notify(adapter, eqo->q.id, false, true, num_evts); + be_eq_notify(adapter, eqo->q.id, false, true, num_evts, 0); /* Return IRQ_HANDLED only for the the first spurious intr * after a valid intr to stop the kernel from branding @@ -2589,7 +2824,7 @@ static irqreturn_t be_msix(int irq, void *dev) { struct be_eq_obj *eqo = dev; - be_eq_notify(eqo->adapter, eqo->q.id, false, true, 0); + be_eq_notify(eqo->adapter, eqo->q.id, false, true, 0, 0); napi_schedule(&eqo->napi); return IRQ_HANDLED; } @@ -2838,6 +3073,7 @@ int be_poll(struct napi_struct *napi, int budget) int max_work = 0, work, i, num_evts; struct be_rx_obj *rxo; struct be_tx_obj *txo; + u32 mult_enc = 0; num_evts = events_get(eqo); @@ -2863,10 +3099,18 @@ int be_poll(struct napi_struct *napi, int budget) if (max_work < budget) { napi_complete(napi); - be_eq_notify(adapter, eqo->q.id, true, false, num_evts); + + /* Skyhawk EQ_DB has a provision to set the rearm to interrupt + * delay via a delay multiplier encoding value + */ + if (skyhawk_chip(adapter)) + mult_enc = be_get_eq_delay_mult_enc(eqo); + + be_eq_notify(adapter, eqo->q.id, true, false, num_evts, + mult_enc); } else { /* As we'll continue in polling mode, count and clear events */ - be_eq_notify(adapter, eqo->q.id, false, false, num_evts); + be_eq_notify(adapter, eqo->q.id, false, false, num_evts, 0); } return max_work; } @@ -2898,22 +3142,19 @@ void be_detect_error(struct be_adapter *adapter) u32 ue_lo = 0, ue_hi = 0, ue_lo_mask = 0, ue_hi_mask = 0; u32 sliport_status = 0, sliport_err1 = 0, sliport_err2 = 0; u32 i; - bool error_detected = false; struct device *dev = &adapter->pdev->dev; - struct net_device *netdev = adapter->netdev; - if (be_hw_error(adapter)) + if (be_check_error(adapter, BE_ERROR_HW)) return; if (lancer_chip(adapter)) { sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET); if (sliport_status & SLIPORT_STATUS_ERR_MASK) { + be_set_error(adapter, BE_ERROR_UE); sliport_err1 = ioread32(adapter->db + SLIPORT_ERROR1_OFFSET); sliport_err2 = ioread32(adapter->db + SLIPORT_ERROR2_OFFSET); - adapter->hw_error = true; - error_detected = true; /* Do not log error messages if its a FW reset */ if (sliport_err1 == SLIPORT_ERROR_FW_RESET1 && sliport_err2 == SLIPORT_ERROR_FW_RESET2) { @@ -2945,12 +3186,12 @@ void be_detect_error(struct be_adapter *adapter) */ if (ue_lo || ue_hi) { - error_detected = true; dev_err(dev, "Unrecoverable Error detected in the adapter"); dev_err(dev, "Please reboot server to recover"); if (skyhawk_chip(adapter)) - adapter->hw_error = true; + be_set_error(adapter, BE_ERROR_UE); + for (i = 0; ue_lo; ue_lo >>= 1, i++) { if (ue_lo & 1) dev_err(dev, "UE: %s bit set\n", @@ -2963,8 +3204,6 @@ void be_detect_error(struct be_adapter *adapter) } } } - if (error_detected) - netif_carrier_off(netdev); } static void be_msix_disable(struct be_adapter *adapter) @@ -3015,7 +3254,7 @@ fail: dev_warn(dev, "MSIx enable failed\n"); /* INTx is not supported in VFs, so fail probe if enable_msix fails */ - if (!be_physfn(adapter)) + if (be_virtfn(adapter)) return num_vec; return 0; } @@ -3062,7 +3301,7 @@ static int be_irq_register(struct be_adapter *adapter) if (status == 0) goto done; /* INTx is not supported for VF */ - if (!be_physfn(adapter)) + if (be_virtfn(adapter)) return status; } @@ -3229,9 +3468,12 @@ static int be_rx_qs_create(struct be_adapter *adapter) memcpy(rss->rss_hkey, rss_key, RSS_HASH_KEY_LEN); - /* First time posting */ + /* Post 1 less than RXQ-len to avoid head being equal to tail, + * which is a queue empty condition + */ for_all_rx_queues(adapter, rxo, i) - be_post_rx_frags(rxo, GFP_KERNEL, MAX_RX_POST); + be_post_rx_frags(rxo, GFP_KERNEL, RX_Q_LEN - 1); + return 0; } @@ -3263,7 +3505,7 @@ static int be_open(struct net_device *netdev) for_all_evt_queues(adapter, eqo, i) { napi_enable(&eqo->napi); be_enable_busy_poll(eqo); - be_eq_notify(adapter, eqo->q.id, true, true, 0); + be_eq_notify(adapter, eqo->q.id, true, true, 0, 0); } adapter->flags |= BE_FLAGS_NAPI_ENABLED; @@ -3563,7 +3805,7 @@ static int be_vfs_if_create(struct be_adapter *adapter) /* If a FW profile exists, then cap_flags are updated */ cap_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST | - BE_IF_FLAGS_MULTICAST; + BE_IF_FLAGS_MULTICAST | BE_IF_FLAGS_PASS_L3L4_ERRORS; for_all_vfs(adapter, vf_cfg, vf) { if (!BE3_chip(adapter)) { @@ -3610,6 +3852,7 @@ static int be_vf_setup(struct be_adapter *adapter) struct device *dev = &adapter->pdev->dev; struct be_vf_cfg *vf_cfg; int status, old_vfs, vf; + bool spoofchk; old_vfs = pci_num_vf(adapter->pdev); @@ -3657,6 +3900,12 @@ static int be_vf_setup(struct be_adapter *adapter) if (!old_vfs) be_cmd_config_qos(adapter, 0, 0, vf + 1); + status = be_cmd_get_hsw_config(adapter, NULL, vf + 1, + vf_cfg->if_handle, NULL, + &spoofchk); + if (!status) + vf_cfg->spoofchk = spoofchk; + if (!old_vfs) { be_cmd_enable_vf(adapter, vf + 1); be_cmd_set_logical_link_config(adapter, @@ -3733,8 +3982,9 @@ static void BEx_get_resources(struct be_adapter *adapter, * *only* if it is RSS-capable. */ if (BE2_chip(adapter) || use_sriov || (adapter->port_num > 1) || - !be_physfn(adapter) || (be_is_mc(adapter) && - !(adapter->function_caps & BE_FUNCTION_CAPS_RSS))) { + be_virtfn(adapter) || + (be_is_mc(adapter) && + !(adapter->function_caps & BE_FUNCTION_CAPS_RSS))) { res->max_tx_qs = 1; } else if (adapter->function_caps & BE_FUNCTION_CAPS_SUPER_NIC) { struct be_resources super_nic_res = {0}; @@ -4075,7 +4325,7 @@ static int be_func_init(struct be_adapter *adapter) msleep(100); /* We can clear all errors when function reset succeeds */ - be_clear_all_error(adapter); + be_clear_error(adapter, BE_CLEAR_ALL); } /* Tell FW we're ready to fire cmds */ @@ -4182,7 +4432,7 @@ static void be_netpoll(struct net_device *netdev) int i; for_all_evt_queues(adapter, eqo, i) { - be_eq_notify(eqo->adapter, eqo->q.id, false, true, 0); + be_eq_notify(eqo->adapter, eqo->q.id, false, true, 0, 0); napi_schedule(&eqo->napi); } } @@ -4666,14 +4916,11 @@ static int lancer_fw_download(struct be_adapter *adapter, return 0; } -#define BE2_UFI 2 -#define BE3_UFI 3 -#define BE3R_UFI 10 -#define SH_UFI 4 -#define SH_P2_UFI 11 - -static int be_get_ufi_type(struct be_adapter *adapter, - struct flash_file_hdr_g3 *fhdr) +/* Check if the flash image file is compatible with the adapter that + * is being flashed. + */ +static bool be_check_ufi_compatibility(struct be_adapter *adapter, + struct flash_file_hdr_g3 *fhdr) { if (!fhdr) { dev_err(&adapter->pdev->dev, "Invalid FW UFI file"); @@ -4685,43 +4932,22 @@ static int be_get_ufi_type(struct be_adapter *adapter, */ switch (fhdr->build[0]) { case BLD_STR_UFI_TYPE_SH: - return (fhdr->asic_type_rev == ASIC_REV_P2) ? SH_P2_UFI : - SH_UFI; + if (!skyhawk_chip(adapter)) + return false; + break; case BLD_STR_UFI_TYPE_BE3: - return (fhdr->asic_type_rev == ASIC_REV_B0) ? BE3R_UFI : - BE3_UFI; + if (!BE3_chip(adapter)) + return false; + break; case BLD_STR_UFI_TYPE_BE2: - return BE2_UFI; - default: - return -1; - } -} - -/* Check if the flash image file is compatible with the adapter that - * is being flashed. - * BE3 chips with asic-rev B0 must be flashed only with BE3R_UFI type. - * Skyhawk chips with asic-rev P2 must be flashed only with SH_P2_UFI type. - */ -static bool be_check_ufi_compatibility(struct be_adapter *adapter, - struct flash_file_hdr_g3 *fhdr) -{ - int ufi_type = be_get_ufi_type(adapter, fhdr); - - switch (ufi_type) { - case SH_P2_UFI: - return skyhawk_chip(adapter); - case SH_UFI: - return (skyhawk_chip(adapter) && - adapter->asic_rev < ASIC_REV_P2); - case BE3R_UFI: - return BE3_chip(adapter); - case BE3_UFI: - return (BE3_chip(adapter) && adapter->asic_rev < ASIC_REV_B0); - case BE2_UFI: - return BE2_chip(adapter); + if (!BE2_chip(adapter)) + return false; + break; default: return false; } + + return (fhdr->asic_type_rev >= adapter->asic_rev); } static int be_fw_download(struct be_adapter *adapter, const struct firmware* fw) @@ -4829,7 +5055,7 @@ static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, adapter->if_handle, mode == BRIDGE_MODE_VEPA ? PORT_FWD_TYPE_VEPA : - PORT_FWD_TYPE_VEB); + PORT_FWD_TYPE_VEB, 0); if (status) goto err; @@ -4861,7 +5087,8 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, hsw_mode = PORT_FWD_TYPE_VEB; } else { status = be_cmd_get_hsw_config(adapter, NULL, 0, - adapter->if_handle, &hsw_mode); + adapter->if_handle, &hsw_mode, + NULL); if (status) return 0; } @@ -5014,6 +5241,7 @@ static const struct net_device_ops be_netdev_ops = { .ndo_set_vf_rate = be_set_vf_tx_rate, .ndo_get_vf_config = be_get_vf_config, .ndo_set_vf_link_state = be_set_vf_link_state, + .ndo_set_vf_spoofchk = be_set_vf_spoofchk, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = be_netpoll, #endif @@ -5118,7 +5346,7 @@ static void be_err_detection_task(struct work_struct *work) be_detect_error(adapter); - if (adapter->hw_error) { + if (be_check_error(adapter, BE_ERROR_HW)) { be_cleanup(adapter); /* As of now error recovery support is in Lancer only */ @@ -5182,7 +5410,9 @@ static void be_worker(struct work_struct *work) be_post_rx_frags(rxo, GFP_KERNEL, MAX_RX_POST); } - be_eqd_update(adapter); + /* EQ-delay update for Skyhawk is done while notifying EQ */ + if (!skyhawk_chip(adapter)) + be_eqd_update(adapter, false); if (adapter->flags & BE_FLAGS_EVT_INCOMPATIBLE_SFP) be_log_sfp_info(adapter); @@ -5202,7 +5432,7 @@ static void be_unmap_pci_bars(struct be_adapter *adapter) static int db_bar(struct be_adapter *adapter) { - if (lancer_chip(adapter) || !be_physfn(adapter)) + if (lancer_chip(adapter) || be_virtfn(adapter)) return 0; else return 4; @@ -5382,6 +5612,30 @@ static void be_remove(struct pci_dev *pdev) free_netdev(adapter->netdev); } +static ssize_t be_hwmon_show_temp(struct device *dev, + struct device_attribute *dev_attr, + char *buf) +{ + struct be_adapter *adapter = dev_get_drvdata(dev); + + /* Unit: millidegree Celsius */ + if (adapter->hwmon_info.be_on_die_temp == BE_INVALID_DIE_TEMP) + return -EIO; + else + return sprintf(buf, "%u\n", + adapter->hwmon_info.be_on_die_temp * 1000); +} + +static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, + be_hwmon_show_temp, NULL, 1); + +static struct attribute *be_hwmon_attrs[] = { + &sensor_dev_attr_temp1_input.dev_attr.attr, + NULL +}; + +ATTRIBUTE_GROUPS(be_hwmon); + static char *mc_name(struct be_adapter *adapter) { char *str = ""; /* default */ @@ -5501,6 +5755,16 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id) be_schedule_err_detection(adapter); + /* On Die temperature not supported for VF. */ + if (be_physfn(adapter) && IS_ENABLED(CONFIG_BE2NET_HWMON)) { + adapter->hwmon_info.hwmon_dev = + devm_hwmon_device_register_with_groups(&pdev->dev, + DRV_NAME, + adapter, + be_hwmon_groups); + adapter->hwmon_info.be_on_die_temp = BE_INVALID_DIE_TEMP; + } + dev_info(&pdev->dev, "%s: %s %s port %c\n", nic_name(pdev), func_name(adapter), mc_name(adapter), adapter->port_name); @@ -5593,8 +5857,8 @@ static pci_ers_result_t be_eeh_err_detected(struct pci_dev *pdev, dev_err(&adapter->pdev->dev, "EEH error detected\n"); - if (!adapter->eeh_error) { - adapter->eeh_error = true; + if (!be_check_error(adapter, BE_ERROR_EEH)) { + be_set_error(adapter, BE_ERROR_EEH); be_cancel_err_detection(adapter); @@ -5641,7 +5905,7 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev) return PCI_ERS_RESULT_DISCONNECT; pci_cleanup_aer_uncorrect_error_status(pdev); - be_clear_all_error(adapter); + be_clear_error(adapter, BE_CLEAR_ALL); return PCI_ERS_RESULT_RECOVERED; } diff --git a/drivers/net/ethernet/emulex/benet/be_roce.c b/drivers/net/ethernet/emulex/benet/be_roce.c index 132866433a25..60368207bf58 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.c +++ b/drivers/net/ethernet/emulex/benet/be_roce.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_roce.h b/drivers/net/ethernet/emulex/benet/be_roce.h index e6f7eb1a7d87..cde6ef905ec4 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.h +++ b/drivers/net/ethernet/emulex/benet/be_roce.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2014 Emulex + * Copyright (C) 2005 - 2015 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 66d47e448e4d..bf4cf3fbb5f2 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2118,6 +2118,82 @@ static void fec_enet_get_drvinfo(struct net_device *ndev, strlcpy(info->bus_info, dev_name(&ndev->dev), sizeof(info->bus_info)); } +static int fec_enet_get_regs_len(struct net_device *ndev) +{ + struct fec_enet_private *fep = netdev_priv(ndev); + struct resource *r; + int s = 0; + + r = platform_get_resource(fep->pdev, IORESOURCE_MEM, 0); + if (r) + s = resource_size(r); + + return s; +} + +/* List of registers that can be safety be read to dump them with ethtool */ +#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ + defined(CONFIG_M520x) || defined(CONFIG_M532x) || \ + defined(CONFIG_ARCH_MXC) || defined(CONFIG_SOC_IMX28) +static u32 fec_enet_register_offset[] = { + FEC_IEVENT, FEC_IMASK, FEC_R_DES_ACTIVE_0, FEC_X_DES_ACTIVE_0, + FEC_ECNTRL, FEC_MII_DATA, FEC_MII_SPEED, FEC_MIB_CTRLSTAT, FEC_R_CNTRL, + FEC_X_CNTRL, FEC_ADDR_LOW, FEC_ADDR_HIGH, FEC_OPD, FEC_TXIC0, FEC_TXIC1, + FEC_TXIC2, FEC_RXIC0, FEC_RXIC1, FEC_RXIC2, FEC_HASH_TABLE_HIGH, + FEC_HASH_TABLE_LOW, FEC_GRP_HASH_TABLE_HIGH, FEC_GRP_HASH_TABLE_LOW, + FEC_X_WMRK, FEC_R_BOUND, FEC_R_FSTART, FEC_R_DES_START_1, + FEC_X_DES_START_1, FEC_R_BUFF_SIZE_1, FEC_R_DES_START_2, + FEC_X_DES_START_2, FEC_R_BUFF_SIZE_2, FEC_R_DES_START_0, + FEC_X_DES_START_0, FEC_R_BUFF_SIZE_0, FEC_R_FIFO_RSFL, FEC_R_FIFO_RSEM, + FEC_R_FIFO_RAEM, FEC_R_FIFO_RAFL, FEC_RACC, FEC_RCMR_1, FEC_RCMR_2, + FEC_DMA_CFG_1, FEC_DMA_CFG_2, FEC_R_DES_ACTIVE_1, FEC_X_DES_ACTIVE_1, + FEC_R_DES_ACTIVE_2, FEC_X_DES_ACTIVE_2, FEC_QOS_SCHEME, + RMON_T_DROP, RMON_T_PACKETS, RMON_T_BC_PKT, RMON_T_MC_PKT, + RMON_T_CRC_ALIGN, RMON_T_UNDERSIZE, RMON_T_OVERSIZE, RMON_T_FRAG, + RMON_T_JAB, RMON_T_COL, RMON_T_P64, RMON_T_P65TO127, RMON_T_P128TO255, + RMON_T_P256TO511, RMON_T_P512TO1023, RMON_T_P1024TO2047, + RMON_T_P_GTE2048, RMON_T_OCTETS, + IEEE_T_DROP, IEEE_T_FRAME_OK, IEEE_T_1COL, IEEE_T_MCOL, IEEE_T_DEF, + IEEE_T_LCOL, IEEE_T_EXCOL, IEEE_T_MACERR, IEEE_T_CSERR, IEEE_T_SQE, + IEEE_T_FDXFC, IEEE_T_OCTETS_OK, + RMON_R_PACKETS, RMON_R_BC_PKT, RMON_R_MC_PKT, RMON_R_CRC_ALIGN, + RMON_R_UNDERSIZE, RMON_R_OVERSIZE, RMON_R_FRAG, RMON_R_JAB, + RMON_R_RESVD_O, RMON_R_P64, RMON_R_P65TO127, RMON_R_P128TO255, + RMON_R_P256TO511, RMON_R_P512TO1023, RMON_R_P1024TO2047, + RMON_R_P_GTE2048, RMON_R_OCTETS, + IEEE_R_DROP, IEEE_R_FRAME_OK, IEEE_R_CRC, IEEE_R_ALIGN, IEEE_R_MACERR, + IEEE_R_FDXFC, IEEE_R_OCTETS_OK +}; +#else +static u32 fec_enet_register_offset[] = { + FEC_ECNTRL, FEC_IEVENT, FEC_IMASK, FEC_IVEC, FEC_R_DES_ACTIVE_0, + FEC_R_DES_ACTIVE_1, FEC_R_DES_ACTIVE_2, FEC_X_DES_ACTIVE_0, + FEC_X_DES_ACTIVE_1, FEC_X_DES_ACTIVE_2, FEC_MII_DATA, FEC_MII_SPEED, + FEC_R_BOUND, FEC_R_FSTART, FEC_X_WMRK, FEC_X_FSTART, FEC_R_CNTRL, + FEC_MAX_FRM_LEN, FEC_X_CNTRL, FEC_ADDR_LOW, FEC_ADDR_HIGH, + FEC_GRP_HASH_TABLE_HIGH, FEC_GRP_HASH_TABLE_LOW, FEC_R_DES_START_0, + FEC_R_DES_START_1, FEC_R_DES_START_2, FEC_X_DES_START_0, + FEC_X_DES_START_1, FEC_X_DES_START_2, FEC_R_BUFF_SIZE_0, + FEC_R_BUFF_SIZE_1, FEC_R_BUFF_SIZE_2 +}; +#endif + +static void fec_enet_get_regs(struct net_device *ndev, + struct ethtool_regs *regs, void *regbuf) +{ + struct fec_enet_private *fep = netdev_priv(ndev); + u32 __iomem *theregs = (u32 __iomem *)fep->hwp; + u32 *buf = (u32 *)regbuf; + u32 i, off; + + memset(buf, 0, regs->len); + + for (i = 0; i < ARRAY_SIZE(fec_enet_register_offset); i++) { + off = fec_enet_register_offset[i] / 4; + buf[off] = readl(&theregs[off]); + } +} + static int fec_enet_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info) { @@ -2515,6 +2591,8 @@ static const struct ethtool_ops fec_enet_ethtool_ops = { .get_settings = fec_enet_get_settings, .set_settings = fec_enet_set_settings, .get_drvinfo = fec_enet_get_drvinfo, + .get_regs_len = fec_enet_get_regs_len, + .get_regs = fec_enet_get_regs, .nway_reset = fec_enet_nway_reset, .get_link = ethtool_op_get_link, .get_coalesce = fec_enet_get_coalesce, diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 4ee080d49bc0..ff875028fdff 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -516,6 +516,15 @@ static struct net_device_stats *gfar_get_stats(struct net_device *dev) return &dev->stats; } +static int gfar_set_mac_addr(struct net_device *dev, void *p) +{ + eth_mac_addr(dev, p); + + gfar_set_mac_for_addr(dev, 0, dev->dev_addr); + + return 0; +} + static const struct net_device_ops gfar_netdev_ops = { .ndo_open = gfar_enet_open, .ndo_start_xmit = gfar_start_xmit, @@ -526,7 +535,7 @@ static const struct net_device_ops gfar_netdev_ops = { .ndo_tx_timeout = gfar_timeout, .ndo_do_ioctl = gfar_ioctl, .ndo_get_stats = gfar_get_stats, - .ndo_set_mac_address = eth_mac_addr, + .ndo_set_mac_address = gfar_set_mac_addr, .ndo_validate_addr = eth_validate_addr, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = gfar_netpoll, @@ -1411,6 +1420,8 @@ static int gfar_probe(struct platform_device *ofdev) dev->features |= NETIF_F_HW_VLAN_CTAG_RX; } + dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + gfar_init_addr_hash_table(priv); /* Insert receive time stamps into padding alignment bytes */ @@ -2254,7 +2265,6 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) int i, rq = 0; int do_tstamp, do_csum, do_vlan; u32 bufaddr; - unsigned long flags; unsigned int nr_frags, nr_txbds, bytes_sent, fcb_len = 0; rq = skb->queue_mapping; @@ -2434,19 +2444,6 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) netdev_tx_sent_queue(txq, bytes_sent); - /* We can work in parallel with gfar_clean_tx_ring(), except - * when modifying num_txbdfree. Note that we didn't grab the lock - * when we were reading the num_txbdfree and checking for available - * space, that's because outside of this function it can only grow, - * and once we've got needed space, it cannot suddenly disappear. - * - * The lock also protects us from gfar_error(), which can modify - * regs->tstat and thus retrigger the transfers, which is why we - * also must grab the lock before setting ready bit for the first - * to be transmitted BD. - */ - spin_lock_irqsave(&tx_queue->txlock, flags); - gfar_wmb(); txbdp_start->lstatus = cpu_to_be32(lstatus); @@ -2463,8 +2460,15 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) tx_queue->cur_tx = next_txbd(txbdp, base, tx_queue->tx_ring_size); + /* We can work in parallel with gfar_clean_tx_ring(), except + * when modifying num_txbdfree. Note that we didn't grab the lock + * when we were reading the num_txbdfree and checking for available + * space, that's because outside of this function it can only grow. + */ + spin_lock_bh(&tx_queue->txlock); /* reduce TxBD free count */ tx_queue->num_txbdfree -= (nr_txbds); + spin_unlock_bh(&tx_queue->txlock); /* If the next BD still needs to be cleaned up, then the bds * are full. We need to tell the kernel to stop sending us stuff. @@ -2478,9 +2482,6 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Tell the DMA to go go go */ gfar_write(®s->tstat, TSTAT_CLEAR_THALT >> tx_queue->qindex); - /* Unlock priv */ - spin_unlock_irqrestore(&tx_queue->txlock, flags); - return NETDEV_TX_OK; dma_map_err: @@ -2622,7 +2623,6 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) skb_dirtytx = tx_queue->skb_dirtytx; while ((skb = tx_queue->tx_skbuff[skb_dirtytx])) { - unsigned long flags; frags = skb_shinfo(skb)->nr_frags; @@ -2686,9 +2686,9 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) TX_RING_MOD_MASK(tx_ring_size); howmany++; - spin_lock_irqsave(&tx_queue->txlock, flags); + spin_lock(&tx_queue->txlock); tx_queue->num_txbdfree += nr_txbds; - spin_unlock_irqrestore(&tx_queue->txlock, flags); + spin_unlock(&tx_queue->txlock); } /* If we freed a buffer, we can restart transmission, if necessary */ @@ -3411,21 +3411,12 @@ static irqreturn_t gfar_error(int irq, void *grp_id) if (events & IEVENT_CRL) dev->stats.tx_aborted_errors++; if (events & IEVENT_XFUN) { - unsigned long flags; - netif_dbg(priv, tx_err, dev, "TX FIFO underrun, packet dropped\n"); dev->stats.tx_dropped++; atomic64_inc(&priv->extra_stats.tx_underrun); - local_irq_save(flags); - lock_tx_qs(priv); - - /* Reactivate the Tx Queues */ - gfar_write(®s->tstat, gfargrp->tstat); - - unlock_tx_qs(priv); - local_irq_restore(flags); + schedule_work(&priv->reset_task); } netif_dbg(priv, tx_err, dev, "Transmit Error\n"); } diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c index 3b39fdddeb57..d49bee38cd31 100644 --- a/drivers/net/ethernet/hisilicon/hip04_eth.c +++ b/drivers/net/ethernet/hisilicon/hip04_eth.c @@ -798,7 +798,7 @@ static void hip04_free_ring(struct net_device *ndev, struct device *d) for (i = 0; i < RX_DESC_NUM; i++) if (priv->rx_buf[i]) - put_page(virt_to_head_page(priv->rx_buf[i])); + skb_free_frag(priv->rx_buf[i]); for (i = 0; i < TX_DESC_NUM; i++) if (priv->tx_skb[i]) diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c index 0ffdcd381fdd..a5e077eac99a 100644 --- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c +++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c @@ -500,7 +500,6 @@ static int hix5hd2_rx(struct net_device *dev, int limit) napi_gro_receive(&priv->napi, skb); dev->stats.rx_packets++; dev->stats.rx_bytes += skb->len; - dev->last_rx = jiffies; next: pos = dma_ring_incr(pos, RX_DESC_NUM); } diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index 18134766a114..29bbb628d712 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -58,7 +58,7 @@ static struct kobj_type ktype_veth_pool; static const char ibmveth_driver_name[] = "ibmveth"; static const char ibmveth_driver_string[] = "IBM Power Virtual Ethernet Driver"; -#define ibmveth_driver_version "1.04" +#define ibmveth_driver_version "1.05" MODULE_AUTHOR("Santiago Leon <santil@linux.vnet.ibm.com>"); MODULE_DESCRIPTION("IBM Power Virtual Ethernet Driver"); @@ -100,6 +100,8 @@ struct ibmveth_stat ibmveth_stats[] = { { "tx_send_failed", IBMVETH_STAT_OFF(tx_send_failed) }, { "fw_enabled_ipv4_csum", IBMVETH_STAT_OFF(fw_ipv4_csum_support) }, { "fw_enabled_ipv6_csum", IBMVETH_STAT_OFF(fw_ipv6_csum_support) }, + { "tx_large_packets", IBMVETH_STAT_OFF(tx_large_packets) }, + { "rx_large_packets", IBMVETH_STAT_OFF(rx_large_packets) } }; /* simple methods of getting data from the current rxq entry */ @@ -852,6 +854,10 @@ static int ibmveth_set_features(struct net_device *dev, struct ibmveth_adapter *adapter = netdev_priv(dev); int rx_csum = !!(features & NETIF_F_RXCSUM); int rc; + netdev_features_t changed = features ^ dev->features; + + if (features & NETIF_F_TSO & changed) + netdev_info(dev, "TSO feature requires all partitions to have updated driver"); if (rx_csum == adapter->rx_csum) return 0; @@ -1035,6 +1041,15 @@ retry_bounce: descs[i+1].fields.address = dma_addr; } + if (skb_is_gso(skb) && !skb_is_gso_v6(skb)) { + /* Put -1 in the IP checksum to tell phyp it + * is a largesend packet and put the mss in the TCP checksum. + */ + ip_hdr(skb)->check = 0xffff; + tcp_hdr(skb)->check = cpu_to_be16(skb_shinfo(skb)->gso_size); + adapter->tx_large_packets++; + } + if (ibmveth_send(adapter, descs)) { adapter->tx_send_failed++; netdev->stats.tx_dropped++; @@ -1080,6 +1095,7 @@ static int ibmveth_poll(struct napi_struct *napi, int budget) struct net_device *netdev = adapter->netdev; int frames_processed = 0; unsigned long lpar_rc; + struct iphdr *iph; restart_poll: while (frames_processed < budget) { @@ -1122,10 +1138,23 @@ restart_poll: skb_put(skb, length); skb->protocol = eth_type_trans(skb, netdev); - if (csum_good) + if (csum_good) { skb->ip_summed = CHECKSUM_UNNECESSARY; + if (be16_to_cpu(skb->protocol) == ETH_P_IP) { + iph = (struct iphdr *)skb->data; + + /* If the IP checksum is not offloaded and if the packet + * is large send, the checksum must be rebuilt. + */ + if (iph->check == 0xffff) { + iph->check = 0; + iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); + adapter->rx_large_packets++; + } + } + } - netif_receive_skb(skb); /* send it up */ + napi_gro_receive(napi, skb); /* send it up */ netdev->stats.rx_packets++; netdev->stats.rx_bytes += length; @@ -1422,8 +1451,14 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id) NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; netdev->features |= netdev->hw_features; + /* TSO is disabled by default */ + netdev->hw_features |= NETIF_F_TSO; + memcpy(netdev->dev_addr, mac_addr_p, ETH_ALEN); + if (firmware_has_feature(FW_FEATURE_CMO)) + memcpy(pool_count, pool_count_cmo, sizeof(pool_count)); + for (i = 0; i < IBMVETH_NUM_BUFF_POOLS; i++) { struct kobject *kobj = &adapter->rx_buff_pool[i].kobj; int error; diff --git a/drivers/net/ethernet/ibm/ibmveth.h b/drivers/net/ethernet/ibm/ibmveth.h index 1f37499d4398..41dedb1fb2ae 100644 --- a/drivers/net/ethernet/ibm/ibmveth.h +++ b/drivers/net/ethernet/ibm/ibmveth.h @@ -104,7 +104,8 @@ static inline long h_illan_attributes(unsigned long unit_address, static int pool_size[] = { 512, 1024 * 2, 1024 * 16, 1024 * 32, 1024 * 64 }; static int pool_count[] = { 256, 512, 256, 256, 256 }; -static int pool_active[] = { 1, 1, 0, 0, 0}; +static int pool_count_cmo[] = { 256, 512, 256, 256, 64 }; +static int pool_active[] = { 1, 1, 0, 0, 1}; #define IBM_VETH_INVALID_MAP ((u16)0xffff) @@ -160,6 +161,8 @@ struct ibmveth_adapter { u64 rx_no_buffer; u64 tx_map_failed; u64 tx_send_failed; + u64 tx_large_packets; + u64 rx_large_packets; }; /* diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c index 1a450f4b6b12..d2657a412768 100644 --- a/drivers/net/ethernet/intel/e100.c +++ b/drivers/net/ethernet/intel/e100.c @@ -874,7 +874,7 @@ static int e100_exec_cb(struct nic *nic, struct sk_buff *skb, { struct cb *cb; unsigned long flags; - int err = 0; + int err; spin_lock_irqsave(&nic->cb_lock, flags); @@ -2922,9 +2922,7 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_master(pdev); - init_timer(&nic->watchdog); - nic->watchdog.function = e100_watchdog; - nic->watchdog.data = (unsigned long)nic; + setup_timer(&nic->watchdog, e100_watchdog, (unsigned long)nic); INIT_WORK(&nic->tx_timeout_task, e100_tx_timeout_task); diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 983eb4e6f7aa..74dc15055971 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -2079,11 +2079,6 @@ static void *e1000_alloc_frag(const struct e1000_adapter *a) return data; } -static void e1000_free_frag(const void *data) -{ - put_page(virt_to_head_page(data)); -} - /** * e1000_clean_rx_ring - Free Rx Buffers per Queue * @adapter: board private structure @@ -2107,7 +2102,7 @@ static void e1000_clean_rx_ring(struct e1000_adapter *adapter, adapter->rx_buffer_len, DMA_FROM_DEVICE); if (buffer_info->rxbuf.data) { - e1000_free_frag(buffer_info->rxbuf.data); + skb_free_frag(buffer_info->rxbuf.data); buffer_info->rxbuf.data = NULL; } } else if (adapter->clean_rx == e1000_clean_jumbo_rx_irq) { @@ -4594,28 +4589,28 @@ static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter, data = e1000_alloc_frag(adapter); /* Failed allocation, critical failure */ if (!data) { - e1000_free_frag(olddata); + skb_free_frag(olddata); adapter->alloc_rx_buff_failed++; break; } if (!e1000_check_64k_bound(adapter, data, bufsz)) { /* give up */ - e1000_free_frag(data); - e1000_free_frag(olddata); + skb_free_frag(data); + skb_free_frag(olddata); adapter->alloc_rx_buff_failed++; break; } /* Use new allocation */ - e1000_free_frag(olddata); + skb_free_frag(olddata); } buffer_info->dma = dma_map_single(&pdev->dev, data, adapter->rx_buffer_len, DMA_FROM_DEVICE); if (dma_mapping_error(&pdev->dev, buffer_info->dma)) { - e1000_free_frag(data); + skb_free_frag(data); buffer_info->dma = 0; adapter->alloc_rx_buff_failed++; break; @@ -4637,7 +4632,7 @@ static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter, adapter->rx_buffer_len, DMA_FROM_DEVICE); - e1000_free_frag(data); + skb_free_frag(data); buffer_info->rxbuf.data = NULL; buffer_info->dma = 0; diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c index dc79ed85030b..32e77755a9c6 100644 --- a/drivers/net/ethernet/intel/e1000e/82571.c +++ b/drivers/net/ethernet/intel/e1000e/82571.c @@ -2010,7 +2010,7 @@ const struct e1000_info e1000_82573_info = { .flags2 = FLAG2_DISABLE_ASPM_L1 | FLAG2_DISABLE_ASPM_L0S, .pba = 20, - .max_hw_frame_size = ETH_FRAME_LEN + ETH_FCS_LEN, + .max_hw_frame_size = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN, .get_variants = e1000_get_variants_82571, .mac_ops = &e82571_mac_ops, .phy_ops = &e82_phy_ops_m88, diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index 9d81c0317433..e18443a00bdb 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -1015,7 +1015,7 @@ static s32 e1000_platform_pm_pch_lpt(struct e1000_hw *hw, bool link) u16 max_snoop, max_nosnoop; u16 max_ltr_enc; /* max LTR latency encoded */ s64 lat_ns; /* latency (ns) */ - s64 value; + u64 value; u32 rxa; if (!hw->adapter->max_frame_size) { @@ -1042,12 +1042,13 @@ static s32 e1000_platform_pm_pch_lpt(struct e1000_hw *hw, bool link) */ lat_ns = ((s64)rxa * 1024 - (2 * (s64)hw->adapter->max_frame_size)) * 8 * 1000; - if (lat_ns < 0) - lat_ns = 0; - else - do_div(lat_ns, speed); + if (lat_ns < 0) { + value = 0; + } else { + value = lat_ns; + do_div(value, speed); + } - value = lat_ns; while (value > PCI_LTR_VALUE_MASK) { scale++; value = DIV_ROUND_UP(value, (1 << 5)); @@ -1563,7 +1564,7 @@ static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter) ((adapter->hw.mac.type >= e1000_pch2lan) && (!(er32(CTRL_EXT) & E1000_CTRL_EXT_LSECCK)))) { adapter->flags &= ~FLAG_HAS_JUMBO_FRAMES; - adapter->max_hw_frame_size = ETH_FRAME_LEN + ETH_FCS_LEN; + adapter->max_hw_frame_size = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN; hw->mac.ops.blink_led = NULL; } @@ -5681,7 +5682,7 @@ const struct e1000_info e1000_ich8_info = { | FLAG_HAS_FLASH | FLAG_APME_IN_WUC, .pba = 8, - .max_hw_frame_size = ETH_FRAME_LEN + ETH_FCS_LEN, + .max_hw_frame_size = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN, .get_variants = e1000_get_variants_ich8lan, .mac_ops = &ich8_mac_ops, .phy_ops = &ich8_phy_ops, @@ -5754,7 +5755,7 @@ const struct e1000_info e1000_pch2_info = { .flags2 = FLAG2_HAS_PHY_STATS | FLAG2_HAS_EEE, .pba = 26, - .max_hw_frame_size = 9018, + .max_hw_frame_size = 9022, .get_variants = e1000_get_variants_ich8lan, .mac_ops = &ich8_mac_ops, .phy_ops = &ich8_phy_ops, @@ -5774,7 +5775,7 @@ const struct e1000_info e1000_pch_lpt_info = { .flags2 = FLAG2_HAS_PHY_STATS | FLAG2_HAS_EEE, .pba = 26, - .max_hw_frame_size = 9018, + .max_hw_frame_size = 9022, .get_variants = e1000_get_variants_ich8lan, .mac_ops = &ich8_mac_ops, .phy_ops = &ich8_phy_ops, @@ -5794,7 +5795,7 @@ const struct e1000_info e1000_pch_spt_info = { .flags2 = FLAG2_HAS_PHY_STATS | FLAG2_HAS_EEE, .pba = 26, - .max_hw_frame_size = 9018, + .max_hw_frame_size = 9022, .get_variants = e1000_get_variants_ich8lan, .mac_ops = &ich8_mac_ops, .phy_ops = &ich8_phy_ops, diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index c509a5c900f5..7dd2c11c3f61 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -3807,7 +3807,7 @@ void e1000e_reset(struct e1000_adapter *adapter) /* reset Packet Buffer Allocation to default */ ew32(PBA, pba); - if (adapter->max_frame_size > ETH_FRAME_LEN + ETH_FCS_LEN) { + if (adapter->max_frame_size > (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN)) { /* To maintain wire speed transmits, the Tx FIFO should be * large enough to accommodate two full transmit packets, * rounded up to the next 1KB and expressed in KB. Likewise, @@ -4196,9 +4196,9 @@ static int e1000_sw_init(struct e1000_adapter *adapter) { struct net_device *netdev = adapter->netdev; - adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN; + adapter->rx_buffer_len = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN; adapter->rx_ps_bsize0 = 128; - adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN; + adapter->max_frame_size = netdev->mtu + VLAN_ETH_HLEN + ETH_FCS_LEN; adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN; adapter->tx_ring_count = E1000_DEFAULT_TXD; adapter->rx_ring_count = E1000_DEFAULT_RXD; @@ -5781,17 +5781,17 @@ struct rtnl_link_stats64 *e1000e_get_stats64(struct net_device *netdev, static int e1000_change_mtu(struct net_device *netdev, int new_mtu) { struct e1000_adapter *adapter = netdev_priv(netdev); - int max_frame = new_mtu + VLAN_HLEN + ETH_HLEN + ETH_FCS_LEN; + int max_frame = new_mtu + VLAN_ETH_HLEN + ETH_FCS_LEN; /* Jumbo frame support */ - if ((max_frame > ETH_FRAME_LEN + ETH_FCS_LEN) && + if ((max_frame > (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN)) && !(adapter->flags & FLAG_HAS_JUMBO_FRAMES)) { e_err("Jumbo Frames not supported.\n"); return -EINVAL; } /* Supported frame sizes */ - if ((new_mtu < ETH_ZLEN + ETH_FCS_LEN + VLAN_HLEN) || + if ((new_mtu < (VLAN_ETH_ZLEN + ETH_FCS_LEN)) || (max_frame > adapter->max_hw_frame_size)) { e_err("Unsupported MTU setting\n"); return -EINVAL; @@ -5831,10 +5831,8 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu) adapter->rx_buffer_len = 4096; /* adjust allocation if LPE protects us, and we aren't using SBP */ - if ((max_frame == ETH_FRAME_LEN + ETH_FCS_LEN) || - (max_frame == ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN)) - adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN - + ETH_FCS_LEN; + if (max_frame <= (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN)) + adapter->rx_buffer_len = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN; if (netif_running(netdev)) e1000e_up(adapter); @@ -6678,6 +6676,19 @@ static void e1000_eeprom_checks(struct e1000_adapter *adapter) } } +static netdev_features_t e1000_fix_features(struct net_device *netdev, + netdev_features_t features) +{ + struct e1000_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + /* Jumbo frame workaround on 82579 and newer requires CRC be stripped */ + if ((hw->mac.type >= e1000_pch2lan) && (netdev->mtu > ETH_DATA_LEN)) + features &= ~NETIF_F_RXFCS; + + return features; +} + static int e1000_set_features(struct net_device *netdev, netdev_features_t features) { @@ -6734,6 +6745,7 @@ static const struct net_device_ops e1000e_netdev_ops = { .ndo_poll_controller = e1000_netpoll, #endif .ndo_set_features = e1000_set_features, + .ndo_fix_features = e1000_fix_features, }; /** diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 33c35d3b7420..aca9cef50d81 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -182,6 +182,7 @@ struct i40e_lump_tracking { enum i40e_fd_stat_idx { I40E_FD_STAT_ATR, I40E_FD_STAT_SB, + I40E_FD_STAT_ATR_TUNNEL, I40E_FD_STAT_PF_COUNT }; #define I40E_FD_STAT_PF_IDX(pf_id) ((pf_id) * I40E_FD_STAT_PF_COUNT) @@ -189,6 +190,8 @@ enum i40e_fd_stat_idx { (I40E_FD_STAT_PF_IDX(pf_id) + I40E_FD_STAT_ATR) #define I40E_FD_SB_STAT_IDX(pf_id) \ (I40E_FD_STAT_PF_IDX(pf_id) + I40E_FD_STAT_SB) +#define I40E_FD_ATR_TUNNEL_STAT_IDX(pf_id) \ + (I40E_FD_STAT_PF_IDX(pf_id) + I40E_FD_STAT_ATR_TUNNEL) struct i40e_fdir_filter { struct hlist_node fdir_node; @@ -263,8 +266,6 @@ struct i40e_pf { struct hlist_head fdir_filter_list; u16 fdir_pf_active_filters; - u16 fd_sb_cnt_idx; - u16 fd_atr_cnt_idx; unsigned long fd_flush_timestamp; u32 fd_flush_cnt; u32 fd_add_err; diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 4cbaaeb902c4..9a68c65b17ea 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -147,6 +147,7 @@ static struct i40e_stats i40e_gstrings_stats[] = { I40E_PF_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared), I40E_PF_STAT("fdir_flush_cnt", fd_flush_cnt), I40E_PF_STAT("fdir_atr_match", stats.fd_atr_match), + I40E_PF_STAT("fdir_atr_tunnel_match", stats.fd_atr_tunnel_match), I40E_PF_STAT("fdir_sb_match", stats.fd_sb_match), /* LPI stats */ @@ -1548,6 +1549,17 @@ static int i40e_loopback_test(struct net_device *netdev, u64 *data) return *data; } +static inline bool i40e_active_vfs(struct i40e_pf *pf) +{ + struct i40e_vf *vfs = pf->vf; + int i; + + for (i = 0; i < pf->num_alloc_vfs; i++) + if (vfs[i].vf_states & I40E_VF_STAT_ACTIVE) + return true; + return false; +} + static void i40e_diag_test(struct net_device *netdev, struct ethtool_test *eth_test, u64 *data) { @@ -1560,6 +1572,20 @@ static void i40e_diag_test(struct net_device *netdev, netif_info(pf, drv, netdev, "offline testing starting\n"); set_bit(__I40E_TESTING, &pf->state); + + if (i40e_active_vfs(pf)) { + dev_warn(&pf->pdev->dev, + "Please take active VFS offline and restart the adapter before running NIC diagnostics\n"); + data[I40E_ETH_TEST_REG] = 1; + data[I40E_ETH_TEST_EEPROM] = 1; + data[I40E_ETH_TEST_INTR] = 1; + data[I40E_ETH_TEST_LOOPBACK] = 1; + data[I40E_ETH_TEST_LINK] = 1; + eth_test->flags |= ETH_TEST_FL_FAILED; + clear_bit(__I40E_TESTING, &pf->state); + goto skip_ol_tests; + } + /* If the device is online then take it offline */ if (if_running) /* indicate we're in test mode */ @@ -1605,6 +1631,8 @@ static void i40e_diag_test(struct net_device *netdev, data[I40E_ETH_TEST_LOOPBACK] = 0; } +skip_ol_tests: + netif_info(pf, drv, netdev, "testing finished\n"); } @@ -2265,7 +2293,7 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi, input->pctype = 0; input->dest_vsi = vsi->id; input->fd_status = I40E_FILTER_PROGRAM_DESC_FD_STATUS_FD_ID; - input->cnt_index = pf->fd_sb_cnt_idx; + input->cnt_index = I40E_FD_SB_STAT_IDX(pf->hw.pf_id); input->flow_type = fsp->flow_type; input->ip4_proto = fsp->h_u.usr_ip4_spec.proto; diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c index 1803afeef23e..c8b621e0e7cd 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c +++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c @@ -118,7 +118,7 @@ static inline int i40e_fcoe_fc_eof(struct sk_buff *skb, u8 *eof) * * The FC EOF is converted to the value understood by HW for descriptor * programming. Never call this w/o calling i40e_fcoe_eof_is_supported() - * first. + * first and that already checks for all supported valid eof values. **/ static inline u32 i40e_fcoe_ctxt_eof(u8 eof) { @@ -132,9 +132,12 @@ static inline u32 i40e_fcoe_ctxt_eof(u8 eof) case FC_EOF_A: return I40E_TX_DESC_CMD_L4T_EOFT_EOF_A; default: - /* FIXME: still returns 0 */ - pr_err("Unrecognized EOF %x\n", eof); - return 0; + /* Supported valid eof shall be already checked by + * calling i40e_fcoe_eof_is_supported() first, + * therefore this default case shall never hit. + */ + WARN_ON(1); + return -EINVAL; } } diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index a54c14491e3b..0a3e928a2b00 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -39,7 +39,7 @@ static const char i40e_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 3 -#define DRV_VERSION_BUILD 2 +#define DRV_VERSION_BUILD 4 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN @@ -772,9 +772,8 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf) dcb_cfg = &hw->local_dcbx_config; - /* See if DCB enabled with PFC TC */ - if (!(pf->flags & I40E_FLAG_DCB_ENABLED) || - !(dcb_cfg->pfc.pfcenable)) { + /* Collect Link XOFF stats when PFC is disabled */ + if (!dcb_cfg->pfc.pfcenable) { i40e_update_link_xoff_rx(pf); return; } @@ -1097,12 +1096,18 @@ static void i40e_update_pf_stats(struct i40e_pf *pf) &osd->rx_jabber, &nsd->rx_jabber); /* FDIR stats */ - i40e_stat_update32(hw, I40E_GLQF_PCNT(pf->fd_atr_cnt_idx), + i40e_stat_update32(hw, + I40E_GLQF_PCNT(I40E_FD_ATR_STAT_IDX(pf->hw.pf_id)), pf->stat_offsets_loaded, &osd->fd_atr_match, &nsd->fd_atr_match); - i40e_stat_update32(hw, I40E_GLQF_PCNT(pf->fd_sb_cnt_idx), + i40e_stat_update32(hw, + I40E_GLQF_PCNT(I40E_FD_SB_STAT_IDX(pf->hw.pf_id)), pf->stat_offsets_loaded, &osd->fd_sb_match, &nsd->fd_sb_match); + i40e_stat_update32(hw, + I40E_GLQF_PCNT(I40E_FD_ATR_TUNNEL_STAT_IDX(pf->hw.pf_id)), + pf->stat_offsets_loaded, + &osd->fd_atr_tunnel_match, &nsd->fd_atr_tunnel_match); val = rd32(hw, I40E_PRTPM_EEE_STAT); nsd->tx_lpi_status = @@ -4739,7 +4744,8 @@ static int i40e_up_complete(struct i40e_vsi *vsi) pf->fd_add_err = pf->fd_atr_cnt = 0; if (pf->fd_tcp_rule > 0) { pf->flags &= ~I40E_FLAG_FD_ATR_ENABLED; - dev_info(&pf->pdev->dev, "Forcing ATR off, sideband rules for TCP/IPv4 exist\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "Forcing ATR off, sideband rules for TCP/IPv4 exist\n"); pf->fd_tcp_rule = 0; } i40e_fdir_filter_restore(vsi); @@ -5428,7 +5434,8 @@ void i40e_fdir_check_and_reenable(struct i40e_pf *pf) if ((pf->flags & I40E_FLAG_FD_SB_ENABLED) && (pf->auto_disable_flags & I40E_FLAG_FD_SB_ENABLED)) { pf->auto_disable_flags &= ~I40E_FLAG_FD_SB_ENABLED; - dev_info(&pf->pdev->dev, "FD Sideband/ntuple is being enabled since we have space in the table now\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "FD Sideband/ntuple is being enabled since we have space in the table now\n"); } } /* Wait for some more space to be available to turn on ATR */ @@ -5436,7 +5443,8 @@ void i40e_fdir_check_and_reenable(struct i40e_pf *pf) if ((pf->flags & I40E_FLAG_FD_ATR_ENABLED) && (pf->auto_disable_flags & I40E_FLAG_FD_ATR_ENABLED)) { pf->auto_disable_flags &= ~I40E_FLAG_FD_ATR_ENABLED; - dev_info(&pf->pdev->dev, "ATR is being enabled since we have space in the table now\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "ATR is being enabled since we have space in the table now\n"); } } } @@ -5469,7 +5477,8 @@ static void i40e_fdir_flush_and_replay(struct i40e_pf *pf) if (!(time_after(jiffies, min_flush_time)) && (fd_room < I40E_FDIR_BUFFER_HEAD_ROOM_FOR_ATR)) { - dev_info(&pf->pdev->dev, "ATR disabled, not enough FD filter space.\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "ATR disabled, not enough FD filter space.\n"); disable_atr = true; } @@ -5496,7 +5505,8 @@ static void i40e_fdir_flush_and_replay(struct i40e_pf *pf) if (!disable_atr) pf->flags |= I40E_FLAG_FD_ATR_ENABLED; clear_bit(__I40E_FD_FLUSH_REQUESTED, &pf->state); - dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); } } } @@ -7676,12 +7686,8 @@ static int i40e_sw_init(struct i40e_pf *pf) (pf->hw.func_caps.fd_filters_best_effort > 0)) { pf->flags |= I40E_FLAG_FD_ATR_ENABLED; pf->atr_sample_rate = I40E_DEFAULT_ATR_SAMPLE_RATE; - /* Setup a counter for fd_atr per PF */ - pf->fd_atr_cnt_idx = I40E_FD_ATR_STAT_IDX(pf->hw.pf_id); if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) { pf->flags |= I40E_FLAG_FD_SB_ENABLED; - /* Setup a counter for fd_sb per PF */ - pf->fd_sb_cnt_idx = I40E_FD_SB_STAT_IDX(pf->hw.pf_id); } else { dev_info(&pf->pdev->dev, "Flow Director Sideband mode Disabled in MFP mode\n"); @@ -7771,7 +7777,8 @@ bool i40e_set_ntuple(struct i40e_pf *pf, netdev_features_t features) pf->fd_add_err = pf->fd_atr_cnt = pf->fd_tcp_rule = 0; pf->fdir_pf_active_filters = 0; pf->flags |= I40E_FLAG_FD_ATR_ENABLED; - dev_info(&pf->pdev->dev, "ATR re-enabled.\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "ATR re-enabled.\n"); /* if ATR was auto disabled it can be re-enabled. */ if ((pf->flags & I40E_FLAG_FD_ATR_ENABLED) && (pf->auto_disable_flags & I40E_FLAG_FD_ATR_ENABLED)) diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 4bd3a80aba82..cc82a7ffacb0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -165,9 +165,6 @@ int i40e_program_fdir_filter(struct i40e_fdir_filter *fdir_data, u8 *raw_packet, tx_desc->cmd_type_offset_bsz = build_ctob(td_cmd, 0, I40E_FDIR_MAX_RAW_PACKET_SIZE, 0); - /* set the timestamp */ - tx_buf->time_stamp = jiffies; - /* Force memory writes to complete before letting h/w * know there are new descriptors to fetch. */ @@ -283,7 +280,8 @@ static int i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi, if (add) { pf->fd_tcp_rule++; if (pf->flags & I40E_FLAG_FD_ATR_ENABLED) { - dev_info(&pf->pdev->dev, "Forcing ATR off, sideband rules for TCP/IPv4 flow being applied\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "Forcing ATR off, sideband rules for TCP/IPv4 flow being applied\n"); pf->flags &= ~I40E_FLAG_FD_ATR_ENABLED; } } else { @@ -291,7 +289,8 @@ static int i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi, (pf->fd_tcp_rule - 1) : 0; if (pf->fd_tcp_rule == 0) { pf->flags |= I40E_FLAG_FD_ATR_ENABLED; - dev_info(&pf->pdev->dev, "ATR re-enabled due to no sideband TCP/IPv4 rules\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "ATR re-enabled due to no sideband TCP/IPv4 rules\n"); } } @@ -501,7 +500,8 @@ static void i40e_fd_handle_status(struct i40e_ring *rx_ring, if ((pf->flags & I40E_FLAG_FD_SB_ENABLED) && !(pf->auto_disable_flags & I40E_FLAG_FD_SB_ENABLED)) { - dev_warn(&pdev->dev, "FD filter space full, new ntuple rules will not be added\n"); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_warn(&pdev->dev, "FD filter space full, new ntuple rules will not be added\n"); pf->auto_disable_flags |= I40E_FLAG_FD_SB_ENABLED; } @@ -807,10 +807,6 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget) tx_ring->vsi->seid, tx_ring->queue_index, tx_ring->next_to_use, i); - dev_info(tx_ring->dev, "tx_bi[next_to_clean]\n" - " time_stamp <%lx>\n" - " jiffies <%lx>\n", - tx_ring->tx_bi[i].time_stamp, jiffies); netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); @@ -1653,9 +1649,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) /* ERR_MASK will only have valid bits if EOP set */ if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) { dev_kfree_skb_any(skb); - /* TODO: shouldn't we increment a counter indicating the - * drop? - */ continue; } @@ -1688,7 +1681,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) skb_mark_napi_id(skb, &rx_ring->q_vector->napi); i40e_receive_skb(rx_ring, skb, vlan_tag); - rx_ring->netdev->last_rx = jiffies; rx_desc->wb.qword1.status_error_len = 0; } while (likely(total_rx_packets < budget)); @@ -1821,7 +1813,6 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget) #endif i40e_receive_skb(rx_ring, skb, vlan_tag); - rx_ring->netdev->last_rx = jiffies; rx_desc->wb.qword1.status_error_len = 0; } while (likely(total_rx_packets < budget)); @@ -1925,11 +1916,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) * i40e_atr - Add a Flow Director ATR filter * @tx_ring: ring to add programming descriptor to * @skb: send buffer - * @flags: send flags + * @tx_flags: send tx flags * @protocol: wire protocol **/ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb, - u32 flags, __be16 protocol) + u32 tx_flags, __be16 protocol) { struct i40e_filter_program_desc *fdir_desc; struct i40e_pf *pf = tx_ring->vsi->back; @@ -1954,25 +1945,38 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb, if (!tx_ring->atr_sample_rate) return; - /* snag network header to get L4 type and address */ - hdr.network = skb_network_header(skb); + if (!(tx_flags & (I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6))) + return; - /* Currently only IPv4/IPv6 with TCP is supported */ - if (protocol == htons(ETH_P_IP)) { - if (hdr.ipv4->protocol != IPPROTO_TCP) - return; + if (!(tx_flags & I40E_TX_FLAGS_VXLAN_TUNNEL)) { + /* snag network header to get L4 type and address */ + hdr.network = skb_network_header(skb); - /* access ihl as a u8 to avoid unaligned access on ia64 */ - hlen = (hdr.network[0] & 0x0F) << 2; - } else if (protocol == htons(ETH_P_IPV6)) { - if (hdr.ipv6->nexthdr != IPPROTO_TCP) + /* Currently only IPv4/IPv6 with TCP is supported + * access ihl as u8 to avoid unaligned access on ia64 + */ + if (tx_flags & I40E_TX_FLAGS_IPV4) + hlen = (hdr.network[0] & 0x0F) << 2; + else if (protocol == htons(ETH_P_IPV6)) + hlen = sizeof(struct ipv6hdr); + else return; - - hlen = sizeof(struct ipv6hdr); } else { - return; + hdr.network = skb_inner_network_header(skb); + hlen = skb_inner_network_header_len(skb); } + /* Currently only IPv4/IPv6 with TCP is supported + * Note: tx_flags gets modified to reflect inner protocols in + * tx_enable_csum function if encap is enabled. + */ + if ((tx_flags & I40E_TX_FLAGS_IPV4) && + (hdr.ipv4->protocol != IPPROTO_TCP)) + return; + else if ((tx_flags & I40E_TX_FLAGS_IPV6) && + (hdr.ipv6->nexthdr != IPPROTO_TCP)) + return; + th = (struct tcphdr *)(hdr.network + hlen); /* Due to lack of space, no more new filters can be programmed */ @@ -2022,9 +2026,16 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb, I40E_TXD_FLTR_QW1_FD_STATUS_SHIFT; dtype_cmd |= I40E_TXD_FLTR_QW1_CNT_ENA_MASK; - dtype_cmd |= - ((u32)pf->fd_atr_cnt_idx << I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) & - I40E_TXD_FLTR_QW1_CNTINDEX_MASK; + if (!(tx_flags & I40E_TX_FLAGS_VXLAN_TUNNEL)) + dtype_cmd |= + ((u32)I40E_FD_ATR_STAT_IDX(pf->hw.pf_id) << + I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) & + I40E_TXD_FLTR_QW1_CNTINDEX_MASK; + else + dtype_cmd |= + ((u32)I40E_FD_ATR_TUNNEL_STAT_IDX(pf->hw.pf_id) << + I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) & + I40E_TXD_FLTR_QW1_CNTINDEX_MASK; fdir_desc->qindex_flex_ptype_vsi = cpu_to_le32(flex_ptype); fdir_desc->rsvd = cpu_to_le32(0); @@ -2045,13 +2056,13 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb, * otherwise returns 0 to indicate the flags has been set properly. **/ #ifdef I40E_FCOE -int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, - struct i40e_ring *tx_ring, - u32 *flags) -#else -static int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, +inline int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, struct i40e_ring *tx_ring, u32 *flags) +#else +static inline int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, + struct i40e_ring *tx_ring, + u32 *flags) #endif { __be16 protocol = skb->protocol; @@ -2119,16 +2130,14 @@ out: * i40e_tso - set up the tso context descriptor * @tx_ring: ptr to the ring to send * @skb: ptr to the skb we're sending - * @tx_flags: the collected send information - * @protocol: the send protocol * @hdr_len: ptr to the size of the packet header * @cd_tunneling: ptr to context descriptor bits * * Returns 0 if no TSO can happen, 1 if tso is going, or error **/ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb, - u32 tx_flags, __be16 protocol, u8 *hdr_len, - u64 *cd_type_cmd_tso_mss, u32 *cd_tunneling) + u8 *hdr_len, u64 *cd_type_cmd_tso_mss, + u32 *cd_tunneling) { u32 cd_cmd, cd_tso_len, cd_mss; struct ipv6hdr *ipv6h; @@ -2220,12 +2229,12 @@ static int i40e_tsyn(struct i40e_ring *tx_ring, struct sk_buff *skb, /** * i40e_tx_enable_csum - Enable Tx checksum offloads * @skb: send buffer - * @tx_flags: Tx flags currently set + * @tx_flags: pointer to Tx flags currently set * @td_cmd: Tx descriptor command bits to set * @td_offset: Tx descriptor header offsets to set * @cd_tunneling: ptr to context desc bits **/ -static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, +static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags, u32 *td_cmd, u32 *td_offset, struct i40e_ring *tx_ring, u32 *cd_tunneling) @@ -2241,6 +2250,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, switch (ip_hdr(skb)->protocol) { case IPPROTO_UDP: l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING; + *tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL; break; default: return; @@ -2250,18 +2260,17 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, this_ipv6_hdr = inner_ipv6_hdr(skb); this_tcp_hdrlen = inner_tcp_hdrlen(skb); - if (tx_flags & I40E_TX_FLAGS_IPV4) { - - if (tx_flags & I40E_TX_FLAGS_TSO) { + if (*tx_flags & I40E_TX_FLAGS_IPV4) { + if (*tx_flags & I40E_TX_FLAGS_TSO) { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4; ip_hdr(skb)->check = 0; } else { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM; } - } else if (tx_flags & I40E_TX_FLAGS_IPV6) { + } else if (*tx_flags & I40E_TX_FLAGS_IPV6) { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6; - if (tx_flags & I40E_TX_FLAGS_TSO) + if (*tx_flags & I40E_TX_FLAGS_TSO) ip_hdr(skb)->check = 0; } @@ -2273,8 +2282,8 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, skb_transport_offset(skb)) >> 1) << I40E_TXD_CTX_QW0_NATLEN_SHIFT; if (this_ip_hdr->version == 6) { - tx_flags &= ~I40E_TX_FLAGS_IPV4; - tx_flags |= I40E_TX_FLAGS_IPV6; + *tx_flags &= ~I40E_TX_FLAGS_IPV4; + *tx_flags |= I40E_TX_FLAGS_IPV6; } } else { network_hdr_len = skb_network_header_len(skb); @@ -2284,12 +2293,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, } /* Enable IP checksum offloads */ - if (tx_flags & I40E_TX_FLAGS_IPV4) { + if (*tx_flags & I40E_TX_FLAGS_IPV4) { l4_hdr = this_ip_hdr->protocol; /* the stack computes the IP header already, the only time we * need the hardware to recompute it is in the case of TSO. */ - if (tx_flags & I40E_TX_FLAGS_TSO) { + if (*tx_flags & I40E_TX_FLAGS_TSO) { *td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM; this_ip_hdr->check = 0; } else { @@ -2298,7 +2307,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, /* Now set the td_offset for IP header length */ *td_offset = (network_hdr_len >> 2) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT; - } else if (tx_flags & I40E_TX_FLAGS_IPV6) { + } else if (*tx_flags & I40E_TX_FLAGS_IPV6) { l4_hdr = this_ipv6_hdr->nexthdr; *td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6; /* Now set the td_offset for IP header length */ @@ -2396,9 +2405,9 @@ static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) * Returns 0 if stop is not needed **/ #ifdef I40E_FCOE -int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) #else -static int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) #endif { if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) @@ -2478,13 +2487,13 @@ linearize_chk_done: * @td_offset: offset for checksum or crc **/ #ifdef I40E_FCOE -void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, - struct i40e_tx_buffer *first, u32 tx_flags, - const u8 hdr_len, u32 td_cmd, u32 td_offset) -#else -static void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, +inline void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, struct i40e_tx_buffer *first, u32 tx_flags, const u8 hdr_len, u32 td_cmd, u32 td_offset) +#else +static inline void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, + struct i40e_tx_buffer *first, u32 tx_flags, + const u8 hdr_len, u32 td_cmd, u32 td_offset) #endif { unsigned int data_len = skb->data_len; @@ -2590,9 +2599,6 @@ static void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_ring->queue_index), first->bytecount); - /* set the timestamp */ - first->time_stamp = jiffies; - /* Force memory writes to complete before letting h/w * know there are new descriptors to fetch. (Only * applicable for weak-ordered memory model archs, @@ -2645,11 +2651,11 @@ dma_error: * one descriptor. **/ #ifdef I40E_FCOE -int i40e_xmit_descriptor_count(struct sk_buff *skb, - struct i40e_ring *tx_ring) -#else -static int i40e_xmit_descriptor_count(struct sk_buff *skb, +inline int i40e_xmit_descriptor_count(struct sk_buff *skb, struct i40e_ring *tx_ring) +#else +static inline int i40e_xmit_descriptor_count(struct sk_buff *skb, + struct i40e_ring *tx_ring) #endif { unsigned int f; @@ -2711,7 +2717,7 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, else if (protocol == htons(ETH_P_IPV6)) tx_flags |= I40E_TX_FLAGS_IPV6; - tso = i40e_tso(tx_ring, skb, tx_flags, protocol, &hdr_len, + tso = i40e_tso(tx_ring, skb, &hdr_len, &cd_type_cmd_tso_mss, &cd_tunneling); if (tso < 0) @@ -2737,7 +2743,7 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_PARTIAL) { tx_flags |= I40E_TX_FLAGS_CSUM; - i40e_tx_enable_csum(skb, tx_flags, &td_cmd, &td_offset, + i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset, tx_ring, &cd_tunneling); } diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h index 4b0b8102cdc3..0dc48dc9ca61 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h @@ -139,6 +139,7 @@ enum i40e_dyn_idx_t { #define I40E_TX_FLAGS_FSO (u32)(1 << 7) #define I40E_TX_FLAGS_TSYN (u32)(1 << 8) #define I40E_TX_FLAGS_FD_SB (u32)(1 << 9) +#define I40E_TX_FLAGS_VXLAN_TUNNEL (u32)(1 << 10) #define I40E_TX_FLAGS_VLAN_MASK 0xffff0000 #define I40E_TX_FLAGS_VLAN_PRIO_MASK 0xe0000000 #define I40E_TX_FLAGS_VLAN_PRIO_SHIFT 29 @@ -146,7 +147,6 @@ enum i40e_dyn_idx_t { struct i40e_tx_buffer { struct i40e_tx_desc *next_to_watch; - unsigned long time_stamp; union { struct sk_buff *skb; void *raw_buf; diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h index 568e855da0f3..9a5a75b1e2bc 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_type.h +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h @@ -1133,6 +1133,7 @@ struct i40e_hw_port_stats { /* flow director stats */ u64 fd_atr_match; u64 fd_sb_match; + u64 fd_atr_tunnel_match; /* EEE LPI */ u32 tx_lpi_status; u32 rx_lpi_status; diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 78d1c4ff565e..4653b6e653c9 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -980,6 +980,13 @@ static int i40e_pci_sriov_enable(struct pci_dev *pdev, int num_vfs) int pre_existing_vfs = pci_num_vf(pdev); int err = 0; + if (pf->state & __I40E_TESTING) { + dev_warn(&pdev->dev, + "Cannot enable SR-IOV virtual functions while the device is undergoing diagnostic testing\n"); + err = -EPERM; + goto err_out; + } + dev_info(&pdev->dev, "Allocating %d VFs.\n", num_vfs); if (pre_existing_vfs && pre_existing_vfs != num_vfs) i40e_free_vfs(pf); diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c index b077e02a0cc7..ec7e220757db 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c @@ -322,10 +322,6 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget) tx_ring->vsi->seid, tx_ring->queue_index, tx_ring->next_to_use, i); - dev_info(tx_ring->dev, "tx_bi[next_to_clean]\n" - " time_stamp <%lx>\n" - " jiffies <%lx>\n", - tx_ring->tx_bi[i].time_stamp, jiffies); netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); @@ -1128,9 +1124,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) /* ERR_MASK will only have valid bits if EOP set */ if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) { dev_kfree_skb_any(skb); - /* TODO: shouldn't we increment a counter indicating the - * drop? - */ continue; } @@ -1156,7 +1149,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) skb_mark_napi_id(skb, &rx_ring->q_vector->napi); i40e_receive_skb(rx_ring, skb, vlan_tag); - rx_ring->netdev->last_rx = jiffies; rx_desc->wb.qword1.status_error_len = 0; } while (likely(total_rx_packets < budget)); @@ -1271,7 +1263,6 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget) : 0; i40e_receive_skb(rx_ring, skb, vlan_tag); - rx_ring->netdev->last_rx = jiffies; rx_desc->wb.qword1.status_error_len = 0; } while (likely(total_rx_packets < budget)); @@ -1352,7 +1343,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget) } /** - * i40e_tx_prepare_vlan_flags - prepare generic TX VLAN tagging flags for HW + * i40evf_tx_prepare_vlan_flags - prepare generic TX VLAN tagging flags for HW * @skb: send buffer * @tx_ring: ring to send buffer on * @flags: the tx flags to be set @@ -1363,9 +1354,9 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget) * Returns error code indicate the frame should be dropped upon error and the * otherwise returns 0 to indicate the flags has been set properly. **/ -static int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, - struct i40e_ring *tx_ring, - u32 *flags) +static inline int i40evf_tx_prepare_vlan_flags(struct sk_buff *skb, + struct i40e_ring *tx_ring, + u32 *flags) { __be16 protocol = skb->protocol; u32 tx_flags = 0; @@ -1408,16 +1399,14 @@ out: * i40e_tso - set up the tso context descriptor * @tx_ring: ptr to the ring to send * @skb: ptr to the skb we're sending - * @tx_flags: the collected send information - * @protocol: the send protocol * @hdr_len: ptr to the size of the packet header * @cd_tunneling: ptr to context descriptor bits * * Returns 0 if no TSO can happen, 1 if tso is going, or error **/ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb, - u32 tx_flags, __be16 protocol, u8 *hdr_len, - u64 *cd_type_cmd_tso_mss, u32 *cd_tunneling) + u8 *hdr_len, u64 *cd_type_cmd_tso_mss, + u32 *cd_tunneling) { u32 cd_cmd, cd_tso_len, cd_mss; struct ipv6hdr *ipv6h; @@ -1468,12 +1457,12 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb, /** * i40e_tx_enable_csum - Enable Tx checksum offloads * @skb: send buffer - * @tx_flags: Tx flags currently set + * @tx_flags: pointer to Tx flags currently set * @td_cmd: Tx descriptor command bits to set * @td_offset: Tx descriptor header offsets to set * @cd_tunneling: ptr to context desc bits **/ -static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, +static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags, u32 *td_cmd, u32 *td_offset, struct i40e_ring *tx_ring, u32 *cd_tunneling) @@ -1489,6 +1478,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, switch (ip_hdr(skb)->protocol) { case IPPROTO_UDP: l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING; + *tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL; break; default: return; @@ -1498,18 +1488,17 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, this_ipv6_hdr = inner_ipv6_hdr(skb); this_tcp_hdrlen = inner_tcp_hdrlen(skb); - if (tx_flags & I40E_TX_FLAGS_IPV4) { - - if (tx_flags & I40E_TX_FLAGS_TSO) { + if (*tx_flags & I40E_TX_FLAGS_IPV4) { + if (*tx_flags & I40E_TX_FLAGS_TSO) { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4; ip_hdr(skb)->check = 0; } else { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM; } - } else if (tx_flags & I40E_TX_FLAGS_IPV6) { + } else if (*tx_flags & I40E_TX_FLAGS_IPV6) { *cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6; - if (tx_flags & I40E_TX_FLAGS_TSO) + if (*tx_flags & I40E_TX_FLAGS_TSO) ip_hdr(skb)->check = 0; } @@ -1521,8 +1510,8 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, skb_transport_offset(skb)) >> 1) << I40E_TXD_CTX_QW0_NATLEN_SHIFT; if (this_ip_hdr->version == 6) { - tx_flags &= ~I40E_TX_FLAGS_IPV4; - tx_flags |= I40E_TX_FLAGS_IPV6; + *tx_flags &= ~I40E_TX_FLAGS_IPV4; + *tx_flags |= I40E_TX_FLAGS_IPV6; } @@ -1534,12 +1523,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, } /* Enable IP checksum offloads */ - if (tx_flags & I40E_TX_FLAGS_IPV4) { + if (*tx_flags & I40E_TX_FLAGS_IPV4) { l4_hdr = this_ip_hdr->protocol; /* the stack computes the IP header already, the only time we * need the hardware to recompute it is in the case of TSO. */ - if (tx_flags & I40E_TX_FLAGS_TSO) { + if (*tx_flags & I40E_TX_FLAGS_TSO) { *td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM; this_ip_hdr->check = 0; } else { @@ -1548,7 +1537,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 tx_flags, /* Now set the td_offset for IP header length */ *td_offset = (network_hdr_len >> 2) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT; - } else if (tx_flags & I40E_TX_FLAGS_IPV6) { + } else if (*tx_flags & I40E_TX_FLAGS_IPV6) { l4_hdr = this_ipv6_hdr->nexthdr; *td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6; /* Now set the td_offset for IP header length */ @@ -1677,7 +1666,44 @@ linearize_chk_done: } /** - * i40e_tx_map - Build the Tx descriptor + * __i40evf_maybe_stop_tx - 2nd level check for tx stop conditions + * @tx_ring: the ring to be checked + * @size: the size buffer we want to assure is available + * + * Returns -EBUSY if a stop is needed, else 0 + **/ +static inline int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +{ + netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); + /* Memory barrier before checking head and tail */ + smp_mb(); + + /* Check again in a case another CPU has just made room available. */ + if (likely(I40E_DESC_UNUSED(tx_ring) < size)) + return -EBUSY; + + /* A reprieve! - use start_queue because it doesn't call schedule */ + netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index); + ++tx_ring->tx_stats.restart_queue; + return 0; +} + +/** + * i40evf_maybe_stop_tx - 1st level check for tx stop conditions + * @tx_ring: the ring to be checked + * @size: the size buffer we want to assure is available + * + * Returns 0 if stop is not needed + **/ +static inline int i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +{ + if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) + return 0; + return __i40evf_maybe_stop_tx(tx_ring, size); +} + +/** + * i40evf_tx_map - Build the Tx descriptor * @tx_ring: ring to send buffer on * @skb: send buffer * @first: first buffer info buffer to use @@ -1686,9 +1712,9 @@ linearize_chk_done: * @td_cmd: the command field in the descriptor * @td_offset: offset for checksum or crc **/ -static void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, - struct i40e_tx_buffer *first, u32 tx_flags, - const u8 hdr_len, u32 td_cmd, u32 td_offset) +static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, + struct i40e_tx_buffer *first, u32 tx_flags, + const u8 hdr_len, u32 td_cmd, u32 td_offset) { unsigned int data_len = skb->data_len; unsigned int size = skb_headlen(skb); @@ -1794,9 +1820,6 @@ static void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_ring->queue_index), first->bytecount); - /* set the timestamp */ - first->time_stamp = jiffies; - /* Force memory writes to complete before letting h/w * know there are new descriptors to fetch. (Only * applicable for weak-ordered memory model archs, @@ -1813,8 +1836,12 @@ static void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_ring->next_to_use = i; + i40evf_maybe_stop_tx(tx_ring, DESC_NEEDED); /* notify HW of packet */ - writel(i, tx_ring->tail); + if (!skb->xmit_more || + netif_xmit_stopped(netdev_get_tx_queue(tx_ring->netdev, + tx_ring->queue_index))) + writel(i, tx_ring->tail); return; @@ -1836,44 +1863,7 @@ dma_error: } /** - * __i40e_maybe_stop_tx - 2nd level check for tx stop conditions - * @tx_ring: the ring to be checked - * @size: the size buffer we want to assure is available - * - * Returns -EBUSY if a stop is needed, else 0 - **/ -static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) -{ - netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); - /* Memory barrier before checking head and tail */ - smp_mb(); - - /* Check again in a case another CPU has just made room available. */ - if (likely(I40E_DESC_UNUSED(tx_ring) < size)) - return -EBUSY; - - /* A reprieve! - use start_queue because it doesn't call schedule */ - netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index); - ++tx_ring->tx_stats.restart_queue; - return 0; -} - -/** - * i40e_maybe_stop_tx - 1st level check for tx stop conditions - * @tx_ring: the ring to be checked - * @size: the size buffer we want to assure is available - * - * Returns 0 if stop is not needed - **/ -static int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) -{ - if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) - return 0; - return __i40e_maybe_stop_tx(tx_ring, size); -} - -/** - * i40e_xmit_descriptor_count - calculate number of tx descriptors needed + * i40evf_xmit_descriptor_count - calculate number of tx descriptors needed * @skb: send buffer * @tx_ring: ring to send buffer on * @@ -1881,8 +1871,8 @@ static int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) * there is not enough descriptors available in this ring since we need at least * one descriptor. **/ -static int i40e_xmit_descriptor_count(struct sk_buff *skb, - struct i40e_ring *tx_ring) +static inline int i40evf_xmit_descriptor_count(struct sk_buff *skb, + struct i40e_ring *tx_ring) { unsigned int f; int count = 0; @@ -1897,7 +1887,7 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb, count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); count += TXD_USE_COUNT(skb_headlen(skb)); - if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) { + if (i40evf_maybe_stop_tx(tx_ring, count + 4 + 1)) { tx_ring->tx_stats.tx_busy++; return 0; } @@ -1923,11 +1913,11 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, u32 td_cmd = 0; u8 hdr_len = 0; int tso; - if (0 == i40e_xmit_descriptor_count(skb, tx_ring)) + if (0 == i40evf_xmit_descriptor_count(skb, tx_ring)) return NETDEV_TX_BUSY; /* prepare the xmit flags */ - if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) + if (i40evf_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) goto out_drop; /* obtain protocol of skb */ @@ -1942,7 +1932,7 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, else if (protocol == htons(ETH_P_IPV6)) tx_flags |= I40E_TX_FLAGS_IPV6; - tso = i40e_tso(tx_ring, skb, tx_flags, protocol, &hdr_len, + tso = i40e_tso(tx_ring, skb, &hdr_len, &cd_type_cmd_tso_mss, &cd_tunneling); if (tso < 0) @@ -1963,17 +1953,15 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_PARTIAL) { tx_flags |= I40E_TX_FLAGS_CSUM; - i40e_tx_enable_csum(skb, tx_flags, &td_cmd, &td_offset, + i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset, tx_ring, &cd_tunneling); } i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss, cd_tunneling, cd_l2tag2); - i40e_tx_map(tx_ring, skb, first, tx_flags, hdr_len, - td_cmd, td_offset); - - i40e_maybe_stop_tx(tx_ring, DESC_NEEDED); + i40evf_tx_map(tx_ring, skb, first, tx_flags, hdr_len, + td_cmd, td_offset); return NETDEV_TX_OK; diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h index 1e49bb1fbac1..e7a34f899f2c 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h @@ -138,6 +138,7 @@ enum i40e_dyn_idx_t { #define I40E_TX_FLAGS_FCCRC (u32)(1 << 6) #define I40E_TX_FLAGS_FSO (u32)(1 << 7) #define I40E_TX_FLAGS_FD_SB (u32)(1 << 9) +#define I40E_TX_FLAGS_VXLAN_TUNNEL (u32)(1 << 10) #define I40E_TX_FLAGS_VLAN_MASK 0xffff0000 #define I40E_TX_FLAGS_VLAN_PRIO_MASK 0xe0000000 #define I40E_TX_FLAGS_VLAN_PRIO_SHIFT 29 @@ -145,7 +146,6 @@ enum i40e_dyn_idx_t { struct i40e_tx_buffer { struct i40e_tx_desc *next_to_watch; - unsigned long time_stamp; union { struct sk_buff *skb; void *raw_buf; diff --git a/drivers/net/ethernet/intel/i40evf/i40e_type.h b/drivers/net/ethernet/intel/i40evf/i40e_type.h index ec9d83a93379..c463ec41579c 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_type.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_type.h @@ -1108,6 +1108,7 @@ struct i40e_hw_port_stats { /* flow director stats */ u64 fd_atr_match; u64 fd_sb_match; + u64 fd_atr_tunnel_match; /* EEE LPI */ u32 tx_lpi_status; u32 rx_lpi_status; diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index a0a9b1fcb5e8..f287186192bb 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -1836,31 +1836,19 @@ void igb_reinit_locked(struct igb_adapter *adapter) * * @adapter: adapter struct **/ -static s32 igb_enable_mas(struct igb_adapter *adapter) +static void igb_enable_mas(struct igb_adapter *adapter) { struct e1000_hw *hw = &adapter->hw; - u32 connsw; - s32 ret_val = 0; - - connsw = rd32(E1000_CONNSW); - if (!(hw->phy.media_type == e1000_media_type_copper)) - return ret_val; + u32 connsw = rd32(E1000_CONNSW); /* configure for SerDes media detect */ - if (!(connsw & E1000_CONNSW_SERDESD)) { + if ((hw->phy.media_type == e1000_media_type_copper) && + (!(connsw & E1000_CONNSW_SERDESD))) { connsw |= E1000_CONNSW_ENRGSRC; connsw |= E1000_CONNSW_AUTOSENSE_EN; wr32(E1000_CONNSW, connsw); wrfl(); - } else if (connsw & E1000_CONNSW_SERDESD) { - /* already SerDes, no need to enable anything */ - return ret_val; - } else { - netdev_info(adapter->netdev, - "MAS: Unable to configure feature, disabling..\n"); - adapter->flags &= ~IGB_FLAG_MAS_ENABLE; } - return ret_val; } void igb_reset(struct igb_adapter *adapter) @@ -1980,10 +1968,9 @@ void igb_reset(struct igb_adapter *adapter) adapter->ei.get_invariants(hw); adapter->flags &= ~IGB_FLAG_MEDIA_RESET; } - if (adapter->flags & IGB_FLAG_MAS_ENABLE) { - if (igb_enable_mas(adapter)) - dev_err(&pdev->dev, - "Error enabling Media Auto Sense\n"); + if ((mac->type == e1000_82575) && + (adapter->flags & IGB_FLAG_MAS_ENABLE)) { + igb_enable_mas(adapter); } if (hw->mac.ops.init_hw(hw)) dev_err(&pdev->dev, "Hardware Error\n"); @@ -4989,6 +4976,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb, struct igb_tx_buffer *first; int tso; u32 tx_flags = 0; + unsigned short f; u16 count = TXD_USE_COUNT(skb_headlen(skb)); __be16 protocol = vlan_get_protocol(skb); u8 hdr_len = 0; @@ -4999,14 +4987,8 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb, * + 1 desc for context descriptor, * otherwise try next time */ - if (NETDEV_FRAG_PAGE_MAX_SIZE > IGB_MAX_DATA_PER_TXD) { - unsigned short f; - - for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) - count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); - } else { - count += skb_shinfo(skb)->nr_frags; - } + for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) + count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); if (igb_maybe_stop_tx(tx_ring, count + 3)) { /* this is a hard error */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index eafa9ec802ba..9a1d0f142b09 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -2594,18 +2594,35 @@ static int ixgbe_add_ethtool_fdir_entry(struct ixgbe_adapter *adapter, struct ixgbe_hw *hw = &adapter->hw; struct ixgbe_fdir_filter *input; union ixgbe_atr_input mask; + u8 queue; int err; if (!(adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)) return -EOPNOTSUPP; - /* - * Don't allow programming if the action is a queue greater than - * the number of online Rx queues. + /* ring_cookie is a masked into a set of queues and ixgbe pools or + * we use the drop index. */ - if ((fsp->ring_cookie != RX_CLS_FLOW_DISC) && - (fsp->ring_cookie >= adapter->num_rx_queues)) - return -EINVAL; + if (fsp->ring_cookie == RX_CLS_FLOW_DISC) { + queue = IXGBE_FDIR_DROP_QUEUE; + } else { + u32 ring = ethtool_get_flow_spec_ring(fsp->ring_cookie); + u8 vf = ethtool_get_flow_spec_ring_vf(fsp->ring_cookie); + + if (!vf && (ring >= adapter->num_rx_queues)) + return -EINVAL; + else if (vf && + ((vf > adapter->num_vfs) || + ring >= adapter->num_rx_queues_per_pool)) + return -EINVAL; + + /* Map the ring onto the absolute queue index */ + if (!vf) + queue = adapter->rx_ring[ring]->reg_idx; + else + queue = ((vf - 1) * + adapter->num_rx_queues_per_pool) + ring; + } /* Don't allow indexes to exist outside of available space */ if (fsp->location >= ((1024 << adapter->fdir_pballoc) - 2)) { @@ -2683,10 +2700,7 @@ static int ixgbe_add_ethtool_fdir_entry(struct ixgbe_adapter *adapter, /* program filters to filter memory */ err = ixgbe_fdir_write_perfect_filter_82599(hw, - &input->filter, input->sw_idx, - (input->action == IXGBE_FDIR_DROP_QUEUE) ? - IXGBE_FDIR_DROP_QUEUE : - adapter->rx_ring[input->action]->reg_idx); + &input->filter, input->sw_idx, queue); if (err) goto err_out_w_lock; @@ -3053,7 +3067,7 @@ static int ixgbe_get_module_info(struct net_device *dev, { struct ixgbe_adapter *adapter = netdev_priv(dev); struct ixgbe_hw *hw = &adapter->hw; - u32 status; + s32 status; u8 sff8472_rev, addr_mode; bool page_swap = false; @@ -3061,14 +3075,14 @@ static int ixgbe_get_module_info(struct net_device *dev, status = hw->phy.ops.read_i2c_eeprom(hw, IXGBE_SFF_SFF_8472_COMP, &sff8472_rev); - if (status != 0) + if (status) return -EIO; /* addressing mode is not supported */ status = hw->phy.ops.read_i2c_eeprom(hw, IXGBE_SFF_SFF_8472_SWAP, &addr_mode); - if (status != 0) + if (status) return -EIO; if (addr_mode & IXGBE_SFF_ADDRESSING_MODE) { @@ -3095,7 +3109,7 @@ static int ixgbe_get_module_eeprom(struct net_device *dev, { struct ixgbe_adapter *adapter = netdev_priv(dev); struct ixgbe_hw *hw = &adapter->hw; - u32 status = IXGBE_ERR_PHY_ADDR_INVALID; + s32 status = IXGBE_ERR_PHY_ADDR_INVALID; u8 databyte = 0xFF; int i = 0; @@ -3112,7 +3126,7 @@ static int ixgbe_get_module_eeprom(struct net_device *dev, else status = hw->phy.ops.read_i2c_sff8472(hw, i, &databyte); - if (status != 0) + if (status) return -EIO; data[i - ee->offset] = databyte; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 5be12a00e1f4..23d82b34314e 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -4757,7 +4757,7 @@ static int ixgbe_non_sfp_link_config(struct ixgbe_hw *hw) { u32 speed; bool autoneg, link_up = false; - u32 ret = IXGBE_ERR_LINK_SETUP; + int ret = IXGBE_ERR_LINK_SETUP; if (hw->mac.ops.check_link) ret = hw->mac.ops.check_link(hw, &speed, &link_up, false); @@ -8022,7 +8022,7 @@ static int ixgbe_ndo_bridge_setlink(struct net_device *dev, return -EINVAL; nla_for_each_nested(attr, br_spec, rem) { - u32 status; + int status; __u16 mode; if (nla_type(attr) != IFLA_BRIDGE_MODE) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c index 8a2be444113d..af828f89419f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c @@ -317,14 +317,14 @@ bool ixgbe_check_reset_blocked(struct ixgbe_hw *hw) **/ static s32 ixgbe_get_phy_id(struct ixgbe_hw *hw) { - u32 status; + s32 status; u16 phy_id_high = 0; u16 phy_id_low = 0; status = hw->phy.ops.read_reg(hw, MDIO_DEVID1, MDIO_MMD_PMAPMD, &phy_id_high); - if (status == 0) { + if (!status) { hw->phy.id = (u32)(phy_id_high << 16); status = hw->phy.ops.read_reg(hw, MDIO_DEVID2, MDIO_MMD_PMAPMD, &phy_id_low); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c index f5f948d08b43..0a8b5e42e1a9 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c @@ -696,14 +696,14 @@ static void ixgbe_release_swfw_sync_semaphore(struct ixgbe_hw *hw) /* Release both semaphores by writing 0 to the bits REGSMP and SMBI */ - swsm = IXGBE_READ_REG(hw, IXGBE_SWSM); - swsm &= ~IXGBE_SWSM_SMBI; - IXGBE_WRITE_REG(hw, IXGBE_SWSM, swsm); - swsm = IXGBE_READ_REG(hw, IXGBE_SWFW_SYNC); swsm &= ~IXGBE_SWFW_REGSMP; IXGBE_WRITE_REG(hw, IXGBE_SWFW_SYNC, swsm); + swsm = IXGBE_READ_REG(hw, IXGBE_SWSM); + swsm &= ~IXGBE_SWSM_SMBI; + IXGBE_WRITE_REG(hw, IXGBE_SWSM, swsm); + IXGBE_WRITE_FLUSH(hw); } diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index cf5cf819a6b8..b0236985e915 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -103,6 +103,39 @@ static s32 ixgbe_init_eeprom_params_X550(struct ixgbe_hw *hw) return 0; } +/** + * ixgbe_iosf_wait - Wait for IOSF command completion + * @hw: pointer to hardware structure + * @ctrl: pointer to location to receive final IOSF control value + * + * Return: failing status on timeout + * + * Note: ctrl can be NULL if the IOSF control register value is not needed + */ +static s32 ixgbe_iosf_wait(struct ixgbe_hw *hw, u32 *ctrl) +{ + u32 i, command; + + /* Check every 10 usec to see if the address cycle completed. + * The SB IOSF BUSY bit will clear when the operation is + * complete. + */ + for (i = 0; i < IXGBE_MDIO_COMMAND_TIMEOUT; i++) { + command = IXGBE_READ_REG(hw, IXGBE_SB_IOSF_INDIRECT_CTRL); + if (!(command & IXGBE_SB_IOSF_CTRL_BUSY)) + break; + usleep_range(10, 20); + } + if (ctrl) + *ctrl = command; + if (i == IXGBE_MDIO_COMMAND_TIMEOUT) { + hw_dbg(hw, "IOSF wait timed out\n"); + return IXGBE_ERR_PHY; + } + + return 0; +} + /** ixgbe_read_iosf_sb_reg_x550 - Writes a value to specified register of the * IOSF device * @hw: pointer to hardware structure @@ -113,7 +146,17 @@ static s32 ixgbe_init_eeprom_params_X550(struct ixgbe_hw *hw) static s32 ixgbe_read_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, u32 device_type, u32 *data) { - u32 i, command, error; + u32 gssr = IXGBE_GSSR_PHY1_SM | IXGBE_GSSR_PHY0_SM; + u32 command, error; + s32 ret; + + ret = hw->mac.ops.acquire_swfw_sync(hw, gssr); + if (ret) + return ret; + + ret = ixgbe_iosf_wait(hw, NULL); + if (ret) + goto out; command = ((reg_addr << IXGBE_SB_IOSF_CTRL_ADDR_SHIFT) | (device_type << IXGBE_SB_IOSF_CTRL_TARGET_SELECT_SHIFT)); @@ -121,17 +164,7 @@ static s32 ixgbe_read_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, /* Write IOSF control register */ IXGBE_WRITE_REG(hw, IXGBE_SB_IOSF_INDIRECT_CTRL, command); - /* Check every 10 usec to see if the address cycle completed. - * The SB IOSF BUSY bit will clear when the operation is - * complete - */ - for (i = 0; i < IXGBE_MDIO_COMMAND_TIMEOUT; i++) { - usleep_range(10, 20); - - command = IXGBE_READ_REG(hw, IXGBE_SB_IOSF_INDIRECT_CTRL); - if ((command & IXGBE_SB_IOSF_CTRL_BUSY) == 0) - break; - } + ret = ixgbe_iosf_wait(hw, &command); if ((command & IXGBE_SB_IOSF_CTRL_RESP_STAT_MASK) != 0) { error = (command & IXGBE_SB_IOSF_CTRL_CMPL_ERR_MASK) >> @@ -140,14 +173,12 @@ static s32 ixgbe_read_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, return IXGBE_ERR_PHY; } - if (i == IXGBE_MDIO_COMMAND_TIMEOUT) { - hw_dbg(hw, "Read timed out\n"); - return IXGBE_ERR_PHY; - } - - *data = IXGBE_READ_REG(hw, IXGBE_SB_IOSF_INDIRECT_DATA); + if (!ret) + *data = IXGBE_READ_REG(hw, IXGBE_SB_IOSF_INDIRECT_DATA); - return 0; +out: + hw->mac.ops.release_swfw_sync(hw, gssr); + return ret; } /** ixgbe_read_ee_hostif_data_X550 - Read EEPROM word using a host interface @@ -789,7 +820,17 @@ static s32 ixgbe_get_link_capabilities_X550em(struct ixgbe_hw *hw, static s32 ixgbe_write_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, u32 device_type, u32 data) { - u32 i, command, error; + u32 gssr = IXGBE_GSSR_PHY1_SM | IXGBE_GSSR_PHY0_SM; + u32 command, error; + s32 ret; + + ret = hw->mac.ops.acquire_swfw_sync(hw, gssr); + if (ret) + return ret; + + ret = ixgbe_iosf_wait(hw, NULL); + if (ret) + goto out; command = ((reg_addr << IXGBE_SB_IOSF_CTRL_ADDR_SHIFT) | (device_type << IXGBE_SB_IOSF_CTRL_TARGET_SELECT_SHIFT)); @@ -800,17 +841,7 @@ static s32 ixgbe_write_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, /* Write IOSF data register */ IXGBE_WRITE_REG(hw, IXGBE_SB_IOSF_INDIRECT_DATA, data); - /* Check every 10 usec to see if the address cycle completed. - * The SB IOSF BUSY bit will clear when the operation is - * complete - */ - for (i = 0; i < IXGBE_MDIO_COMMAND_TIMEOUT; i++) { - usleep_range(10, 20); - - command = IXGBE_READ_REG(hw, IXGBE_SB_IOSF_INDIRECT_CTRL); - if ((command & IXGBE_SB_IOSF_CTRL_BUSY) == 0) - break; - } + ret = ixgbe_iosf_wait(hw, &command); if ((command & IXGBE_SB_IOSF_CTRL_RESP_STAT_MASK) != 0) { error = (command & IXGBE_SB_IOSF_CTRL_CMPL_ERR_MASK) >> @@ -819,12 +850,9 @@ static s32 ixgbe_write_iosf_sb_reg_x550(struct ixgbe_hw *hw, u32 reg_addr, return IXGBE_ERR_PHY; } - if (i == IXGBE_MDIO_COMMAND_TIMEOUT) { - hw_dbg(hw, "Write timed out\n"); - return IXGBE_ERR_PHY; - } - - return 0; +out: + hw->mac.ops.release_swfw_sync(hw, gssr); + return ret; } /** ixgbe_setup_ixfi_x550em - Configure the KR PHY for iXFI mode. @@ -1035,7 +1063,7 @@ static s32 ixgbe_setup_kr_x550em(struct ixgbe_hw *hw) **/ static s32 ixgbe_setup_internal_phy_x550em(struct ixgbe_hw *hw) { - u32 status; + s32 status; u16 lasi, autoneg_status, speed; ixgbe_link_speed force_speed; @@ -1177,7 +1205,7 @@ static enum ixgbe_media_type ixgbe_get_media_type_X550em(struct ixgbe_hw *hw) **/ static s32 ixgbe_init_ext_t_x550em(struct ixgbe_hw *hw) { - u32 status; + s32 status; u16 reg; u32 retries = 2; diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 1c75829eb166..d52639bc491f 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -3125,9 +3125,8 @@ static int mv643xx_eth_probe(struct platform_device *pdev) mib_counters_clear(mp); - init_timer(&mp->mib_counters_timer); - mp->mib_counters_timer.data = (unsigned long)mp; - mp->mib_counters_timer.function = mib_counters_timer_wrapper; + setup_timer(&mp->mib_counters_timer, mib_counters_timer_wrapper, + (unsigned long)mp); mp->mib_counters_timer.expires = jiffies + 30 * HZ; spin_lock_init(&mp->mib_counters_lock); @@ -3136,9 +3135,7 @@ static int mv643xx_eth_probe(struct platform_device *pdev) netif_napi_add(dev, &mp->napi, mv643xx_eth_poll, NAPI_POLL_WEIGHT); - init_timer(&mp->rx_oom); - mp->rx_oom.data = (unsigned long)mp; - mp->rx_oom.function = oom_timer_wrapper; + setup_timer(&mp->rx_oom, oom_timer_wrapper, (unsigned long)mp); res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index ce5f7f9cff06..ecce8261ce3b 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -1359,7 +1359,7 @@ static void *mvneta_frag_alloc(const struct mvneta_port *pp) static void mvneta_frag_free(const struct mvneta_port *pp, void *data) { if (likely(pp->frag_size <= PAGE_SIZE)) - put_page(virt_to_head_page(data)); + skb_free_frag(data); else kfree(data); } diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 529ef0594b90..91d834495fb7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -882,7 +882,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave, { struct ib_smp *smp = inbox->buf; u32 index; - u8 port; + u8 port, slave_port; u8 opcode_modifier; u16 *table; int err; @@ -894,7 +894,8 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave, __be32 slave_cap_mask; __be64 slave_node_guid; - port = vhcr->in_modifier; + slave_port = vhcr->in_modifier; + port = mlx4_slave_convert_port(dev, slave, slave_port); /* network-view bit is for driver use only, and should not be passed to FW */ opcode_modifier = vhcr->op_modifier & ~0x8; /* clear netw view bit */ @@ -930,8 +931,9 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave, if (smp->attr_id == IB_SMP_ATTR_PORT_INFO) { /*get the slave specific caps:*/ /*do the command */ + smp->attr_mod = cpu_to_be32(port); err = mlx4_cmd_box(dev, inbox->dma, outbox->dma, - vhcr->in_modifier, opcode_modifier, + port, opcode_modifier, vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE); /* modify the response for slaves */ if (!err && slave != mlx4_master_func_num(dev)) { @@ -975,7 +977,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave, } if (smp->attr_id == IB_SMP_ATTR_NODE_INFO) { err = mlx4_cmd_box(dev, inbox->dma, outbox->dma, - vhcr->in_modifier, opcode_modifier, + port, opcode_modifier, vhcr->op, MLX4_CMD_TIME_CLASS_C, MLX4_CMD_NATIVE); if (!err) { slave_node_guid = mlx4_get_slave_node_guid(dev, slave); @@ -3197,6 +3199,12 @@ int mlx4_vf_set_enable_smi_admin(struct mlx4_dev *dev, int slave, int port, int enabled) { struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_active_ports actv_ports = mlx4_get_active_ports( + &priv->dev, slave); + int min_port = find_first_bit(actv_ports.ports, + priv->dev.caps.num_ports) + 1; + int max_port = min_port - 1 + + bitmap_weight(actv_ports.ports, priv->dev.caps.num_ports); if (slave == mlx4_master_func_num(dev)) return 0; @@ -3206,6 +3214,11 @@ int mlx4_vf_set_enable_smi_admin(struct mlx4_dev *dev, int slave, int port, enabled < 0 || enabled > 1) return -EINVAL; + if (min_port == max_port && dev->caps.num_ports > 1) { + mlx4_info(dev, "SMI access disallowed for single ported VFs\n"); + return -EPROTONOSUPPORT; + } + priv->mfunc.master.vf_admin[slave].enable_smi[port] = enabled; return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c index e71f31387ac6..3348e646db70 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cq.c +++ b/drivers/net/ethernet/mellanox/mlx4/cq.c @@ -292,7 +292,7 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, u64 mtt_addr; int err; - if (vector > dev->caps.num_comp_vectors + dev->caps.comp_pool) + if (vector >= dev->caps.num_comp_vectors) return -EINVAL; cq->vector = vector; @@ -319,7 +319,7 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, cq_context->flags |= cpu_to_be32(1 << 19); cq_context->logsize_usrpage = cpu_to_be32((ilog2(nent) << 24) | uar->index); - cq_context->comp_eqn = priv->eq_table.eq[vector].eqn; + cq_context->comp_eqn = priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(vector)].eqn; cq_context->log_page_size = mtt->page_shift - MLX4_ICM_PAGE_SHIFT; mtt_addr = mlx4_mtt_addr(dev, mtt); @@ -339,11 +339,11 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, init_completion(&cq->free); cq->comp = mlx4_add_cq_to_tasklet; cq->tasklet_ctx.priv = - &priv->eq_table.eq[cq->vector].tasklet_ctx; + &priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(vector)].tasklet_ctx; INIT_LIST_HEAD(&cq->tasklet_ctx.list); - cq->irq = priv->eq_table.eq[cq->vector].irq; + cq->irq = priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(vector)].irq; return 0; err_radix: @@ -368,7 +368,10 @@ void mlx4_cq_free(struct mlx4_dev *dev, struct mlx4_cq *cq) if (err) mlx4_warn(dev, "HW2SW_CQ failed (%d) for CQN %06x\n", err, cq->cqn); - synchronize_irq(priv->eq_table.eq[cq->vector].irq); + synchronize_irq(priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(cq->vector)].irq); + if (priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(cq->vector)].irq != + priv->eq_table.eq[MLX4_EQ_ASYNC].irq) + synchronize_irq(priv->eq_table.eq[MLX4_EQ_ASYNC].irq); spin_lock_irq(&cq_table->lock); radix_tree_delete(&cq_table->tree, cq->cqn); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c index 22da4d0d0f05..63769df872a4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c @@ -66,6 +66,7 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv, cq->ring = ring; cq->is_tx = mode; + cq->vector = mdev->dev->caps.num_comp_vectors; /* Allocate HW buffers on provided NUMA node. * dev->numa_node is used in mtt range allocation flow. @@ -101,12 +102,7 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, int err = 0; char name[25]; int timestamp_en = 0; - struct cpu_rmap *rmap = -#ifdef CONFIG_RFS_ACCEL - priv->dev->rx_cpu_rmap; -#else - NULL; -#endif + bool assigned_eq = false; cq->dev = mdev->pndev[priv->port]; cq->mcq.set_ci_db = cq->wqres.db.db; @@ -116,23 +112,19 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, memset(cq->buf, 0, cq->buf_size); if (cq->is_tx == RX) { - if (mdev->dev->caps.comp_pool) { - if (!cq->vector) { - sprintf(name, "%s-%d", priv->dev->name, - cq->ring); - /* Set IRQ for specific name (per ring) */ - if (mlx4_assign_eq(mdev->dev, name, rmap, - &cq->vector)) { - cq->vector = (cq->ring + 1 + priv->port) - % mdev->dev->caps.num_comp_vectors; - mlx4_warn(mdev, "Failed assigning an EQ to %s, falling back to legacy EQ's\n", - name); - } - + if (!mlx4_is_eq_vector_valid(mdev->dev, priv->port, + cq->vector)) { + cq->vector = cpumask_first(priv->rx_ring[cq->ring]->affinity_mask); + + err = mlx4_assign_eq(mdev->dev, priv->port, + &cq->vector); + if (err) { + mlx4_err(mdev, "Failed assigning an EQ to %s\n", + name); + goto free_eq; } - } else { - cq->vector = (cq->ring + 1 + priv->port) % - mdev->dev->caps.num_comp_vectors; + + assigned_eq = true; } cq->irq_desc = @@ -159,7 +151,7 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, &mdev->priv_uar, cq->wqres.db.dma, &cq->mcq, cq->vector, 0, timestamp_en); if (err) - return err; + goto free_eq; cq->mcq.comp = cq->is_tx ? mlx4_en_tx_irq : mlx4_en_rx_irq; cq->mcq.event = mlx4_en_cq_event; @@ -168,13 +160,6 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_tx_cq, NAPI_POLL_WEIGHT); } else { - struct mlx4_en_rx_ring *ring = priv->rx_ring[cq->ring]; - - err = irq_set_affinity_hint(cq->mcq.irq, - ring->affinity_mask); - if (err) - mlx4_warn(mdev, "Failed setting affinity hint\n"); - netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq, 64); napi_hash_add(&cq->napi); } @@ -182,6 +167,12 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, napi_enable(&cq->napi); return 0; + +free_eq: + if (assigned_eq) + mlx4_release_eq(mdev->dev, cq->vector); + cq->vector = mdev->dev->caps.num_comp_vectors; + return err; } void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq **pcq) @@ -191,9 +182,9 @@ void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq **pcq) mlx4_en_unmap_buffer(&cq->wqres.buf); mlx4_free_hwq_res(mdev->dev, &cq->wqres, cq->buf_size); - if (priv->mdev->dev->caps.comp_pool && cq->vector) { + if (mlx4_is_eq_vector_valid(mdev->dev, priv->port, cq->vector) && + cq->is_tx == RX) mlx4_release_eq(priv->mdev->dev, cq->vector); - } cq->vector = 0; cq->buf_size = 0; cq->buf = NULL; @@ -207,7 +198,6 @@ void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) if (!cq->is_tx) { napi_hash_del(&cq->napi); synchronize_rcu(); - irq_set_affinity_hint(cq->mcq.irq, NULL); } netif_napi_del(&cq->napi); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index cf467a9f6cc7..98efb5842fca 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1954,7 +1954,6 @@ void mlx4_en_free_resources(struct mlx4_en_priv *priv) int i; #ifdef CONFIG_RFS_ACCEL - free_irq_cpu_rmap(priv->dev->rx_cpu_rmap); priv->dev->rx_cpu_rmap = NULL; #endif @@ -2012,11 +2011,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv) } #ifdef CONFIG_RFS_ACCEL - if (priv->mdev->dev->caps.comp_pool) { - priv->dev->rx_cpu_rmap = alloc_irq_cpu_rmap(priv->mdev->dev->caps.comp_pool); - if (!priv->dev->rx_cpu_rmap) - goto err; - } + priv->dev->rx_cpu_rmap = mlx4_get_cpu_rmap(priv->mdev->dev, priv->port); #endif return 0; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 2a77a6b19121..35f726c17e48 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -337,15 +337,10 @@ void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev) struct mlx4_dev *dev = mdev->dev; mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) { - if (!dev->caps.comp_pool) - num_of_eqs = max_t(int, MIN_RX_RINGS, - min_t(int, - dev->caps.num_comp_vectors, - DEF_RX_RINGS)); - else - num_of_eqs = min_t(int, MAX_MSIX_P_PORT, - dev->caps.comp_pool/ - dev->caps.num_ports) - 1; + num_of_eqs = max_t(int, MIN_RX_RINGS, + min_t(int, + mlx4_get_eqs_per_port(mdev->dev, i), + DEF_RX_RINGS)); num_rx_rings = mlx4_low_memory_profile() ? MIN_RX_RINGS : min_t(int, num_of_eqs, diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c index 2619c9fbf42d..11168825a9fa 100644 --- a/drivers/net/ethernet/mellanox/mlx4/eq.c +++ b/drivers/net/ethernet/mellanox/mlx4/eq.c @@ -145,7 +145,7 @@ void mlx4_gen_slave_eqe(struct work_struct *work) struct mlx4_slave_event_eq *slave_eq = &mfunc->master.slave_eq; struct mlx4_eqe *eqe; u8 slave; - int i; + int i, phys_port, slave_port; for (eqe = next_slave_event_eqe(slave_eq); eqe; eqe = next_slave_event_eqe(slave_eq)) { @@ -154,9 +154,20 @@ void mlx4_gen_slave_eqe(struct work_struct *work) /* All active slaves need to receive the event */ if (slave == ALL_SLAVES) { for (i = 0; i <= dev->persist->num_vfs; i++) { + phys_port = 0; + if (eqe->type == MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT && + eqe->subtype == MLX4_DEV_PMC_SUBTYPE_PORT_INFO) { + phys_port = eqe->event.port_mgmt_change.port; + slave_port = mlx4_phys_to_slave_port(dev, i, phys_port); + if (slave_port < 0) /* VF doesn't have this port */ + continue; + eqe->event.port_mgmt_change.port = slave_port; + } if (mlx4_GEN_EQE(dev, i, eqe)) mlx4_warn(dev, "Failed to generate event for slave %d\n", i); + if (phys_port) + eqe->event.port_mgmt_change.port = phys_port; } } else { if (mlx4_GEN_EQE(dev, slave, eqe)) @@ -210,6 +221,20 @@ static void mlx4_slave_event(struct mlx4_dev *dev, int slave, slave_event(dev, slave, eqe); } +static void mlx4_set_eq_affinity_hint(struct mlx4_priv *priv, int vec) +{ + int hint_err; + struct mlx4_dev *dev = &priv->dev; + struct mlx4_eq *eq = &priv->eq_table.eq[vec]; + + if (!eq->affinity_mask || cpumask_empty(eq->affinity_mask)) + return; + + hint_err = irq_set_affinity_hint(eq->irq, eq->affinity_mask); + if (hint_err) + mlx4_warn(dev, "irq_set_affinity_hint failed, err %d\n", hint_err); +} + int mlx4_gen_pkey_eqe(struct mlx4_dev *dev, int slave, u8 port) { struct mlx4_eqe eqe; @@ -224,7 +249,7 @@ int mlx4_gen_pkey_eqe(struct mlx4_dev *dev, int slave, u8 port) eqe.type = MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT; eqe.subtype = MLX4_DEV_PMC_SUBTYPE_PKEY_TABLE; - eqe.event.port_mgmt_change.port = port; + eqe.event.port_mgmt_change.port = mlx4_phys_to_slave_port(dev, slave, port); return mlx4_GEN_EQE(dev, slave, &eqe); } @@ -241,7 +266,7 @@ int mlx4_gen_guid_change_eqe(struct mlx4_dev *dev, int slave, u8 port) eqe.type = MLX4_EVENT_TYPE_PORT_MNG_CHG_EVENT; eqe.subtype = MLX4_DEV_PMC_SUBTYPE_GUID_INFO; - eqe.event.port_mgmt_change.port = port; + eqe.event.port_mgmt_change.port = mlx4_phys_to_slave_port(dev, slave, port); return mlx4_GEN_EQE(dev, slave, &eqe); } @@ -251,6 +276,7 @@ int mlx4_gen_port_state_change_eqe(struct mlx4_dev *dev, int slave, u8 port, u8 port_subtype_change) { struct mlx4_eqe eqe; + u8 slave_port = mlx4_phys_to_slave_port(dev, slave, port); /*don't send if we don't have the that slave */ if (dev->persist->num_vfs < slave) @@ -259,7 +285,7 @@ int mlx4_gen_port_state_change_eqe(struct mlx4_dev *dev, int slave, u8 port, eqe.type = MLX4_EVENT_TYPE_PORT_CHANGE; eqe.subtype = port_subtype_change; - eqe.event.port_change.port = cpu_to_be32(port << 28); + eqe.event.port_change.port = cpu_to_be32(slave_port << 28); mlx4_dbg(dev, "%s: sending: %d to slave: %d on port: %d\n", __func__, port_subtype_change, slave, port); @@ -589,6 +615,10 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq) if (SLAVE_PORT_GEN_EVENT_DOWN == gen_event) { if (i == mlx4_master_func_num(dev)) continue; + eqe->event.port_change.port = + cpu_to_be32( + (be32_to_cpu(eqe->event.port_change.port) & 0xFFFFFFF) + | (mlx4_phys_to_slave_port(dev, i, port) << 28)); mlx4_slave_event(dev, i, eqe); } } @@ -879,8 +909,8 @@ static int mlx4_num_eq_uar(struct mlx4_dev *dev) * we need to map, take the difference of highest index and * the lowest index we'll use and add 1. */ - return (dev->caps.num_comp_vectors + 1 + dev->caps.reserved_eqs + - dev->caps.comp_pool)/4 - dev->caps.reserved_eqs/4 + 1; + return (dev->caps.num_comp_vectors + 1 + dev->caps.reserved_eqs) / 4 - + dev->caps.reserved_eqs / 4 + 1; } static void __iomem *mlx4_get_eq_uar(struct mlx4_dev *dev, struct mlx4_eq *eq) @@ -1069,32 +1099,21 @@ static void mlx4_free_eq(struct mlx4_dev *dev, static void mlx4_free_irqs(struct mlx4_dev *dev) { struct mlx4_eq_table *eq_table = &mlx4_priv(dev)->eq_table; - struct mlx4_priv *priv = mlx4_priv(dev); - int i, vec; + int i; if (eq_table->have_irq) free_irq(dev->persist->pdev->irq, dev); for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) if (eq_table->eq[i].have_irq) { + free_cpumask_var(eq_table->eq[i].affinity_mask); +#if defined(CONFIG_SMP) + irq_set_affinity_hint(eq_table->eq[i].irq, NULL); +#endif free_irq(eq_table->eq[i].irq, eq_table->eq + i); eq_table->eq[i].have_irq = 0; } - for (i = 0; i < dev->caps.comp_pool; i++) { - /* - * Freeing the assigned irq's - * all bits should be 0, but we need to validate - */ - if (priv->msix_ctl.pool_bm & 1ULL << i) { - /* NO need protecting*/ - vec = dev->caps.num_comp_vectors + 1 + i; - free_irq(priv->eq_table.eq[vec].irq, - &priv->eq_table.eq[vec]); - } - } - - kfree(eq_table->irq_names); } @@ -1175,76 +1194,73 @@ int mlx4_init_eq_table(struct mlx4_dev *dev) } priv->eq_table.irq_names = - kmalloc(MLX4_IRQNAME_SIZE * (dev->caps.num_comp_vectors + 1 + - dev->caps.comp_pool), + kmalloc(MLX4_IRQNAME_SIZE * (dev->caps.num_comp_vectors + 1), GFP_KERNEL); if (!priv->eq_table.irq_names) { err = -ENOMEM; - goto err_out_bitmap; + goto err_out_clr_int; } - for (i = 0; i < dev->caps.num_comp_vectors; ++i) { - err = mlx4_create_eq(dev, dev->caps.num_cqs - - dev->caps.reserved_cqs + - MLX4_NUM_SPARE_EQE, - (dev->flags & MLX4_FLAG_MSI_X) ? i : 0, - &priv->eq_table.eq[i]); - if (err) { - --i; - goto err_out_unmap; - } - } - - err = mlx4_create_eq(dev, MLX4_NUM_ASYNC_EQE + MLX4_NUM_SPARE_EQE, - (dev->flags & MLX4_FLAG_MSI_X) ? dev->caps.num_comp_vectors : 0, - &priv->eq_table.eq[dev->caps.num_comp_vectors]); - if (err) - goto err_out_comp; - - /*if additional completion vectors poolsize is 0 this loop will not run*/ - for (i = dev->caps.num_comp_vectors + 1; - i < dev->caps.num_comp_vectors + dev->caps.comp_pool + 1; ++i) { + for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) { + if (i == MLX4_EQ_ASYNC) { + err = mlx4_create_eq(dev, + MLX4_NUM_ASYNC_EQE + MLX4_NUM_SPARE_EQE, + 0, &priv->eq_table.eq[MLX4_EQ_ASYNC]); + } else { +#ifdef CONFIG_RFS_ACCEL + struct mlx4_eq *eq = &priv->eq_table.eq[i]; + int port = find_first_bit(eq->actv_ports.ports, + dev->caps.num_ports) + 1; + + if (port <= dev->caps.num_ports) { + struct mlx4_port_info *info = + &mlx4_priv(dev)->port[port]; + + if (!info->rmap) { + info->rmap = alloc_irq_cpu_rmap( + mlx4_get_eqs_per_port(dev, port)); + if (!info->rmap) { + mlx4_warn(dev, "Failed to allocate cpu rmap\n"); + err = -ENOMEM; + goto err_out_unmap; + } + } - err = mlx4_create_eq(dev, dev->caps.num_cqs - - dev->caps.reserved_cqs + - MLX4_NUM_SPARE_EQE, - (dev->flags & MLX4_FLAG_MSI_X) ? i : 0, - &priv->eq_table.eq[i]); - if (err) { - --i; - goto err_out_unmap; + err = irq_cpu_rmap_add( + info->rmap, eq->irq); + if (err) + mlx4_warn(dev, "Failed adding irq rmap\n"); + } +#endif + err = mlx4_create_eq(dev, dev->caps.num_cqs - + dev->caps.reserved_cqs + + MLX4_NUM_SPARE_EQE, + (dev->flags & MLX4_FLAG_MSI_X) ? + i + 1 - !!(i > MLX4_EQ_ASYNC) : 0, + eq); } + if (err) + goto err_out_unmap; } - if (dev->flags & MLX4_FLAG_MSI_X) { const char *eq_name; - for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) { - if (i < dev->caps.num_comp_vectors) { - snprintf(priv->eq_table.irq_names + - i * MLX4_IRQNAME_SIZE, - MLX4_IRQNAME_SIZE, - "mlx4-comp-%d@pci:%s", i, - pci_name(dev->persist->pdev)); - } else { - snprintf(priv->eq_table.irq_names + - i * MLX4_IRQNAME_SIZE, - MLX4_IRQNAME_SIZE, - "mlx4-async@pci:%s", - pci_name(dev->persist->pdev)); - } + snprintf(priv->eq_table.irq_names + + MLX4_EQ_ASYNC * MLX4_IRQNAME_SIZE, + MLX4_IRQNAME_SIZE, + "mlx4-async@pci:%s", + pci_name(dev->persist->pdev)); + eq_name = priv->eq_table.irq_names + + MLX4_EQ_ASYNC * MLX4_IRQNAME_SIZE; - eq_name = priv->eq_table.irq_names + - i * MLX4_IRQNAME_SIZE; - err = request_irq(priv->eq_table.eq[i].irq, - mlx4_msi_x_interrupt, 0, eq_name, - priv->eq_table.eq + i); - if (err) - goto err_out_async; + err = request_irq(priv->eq_table.eq[MLX4_EQ_ASYNC].irq, + mlx4_msi_x_interrupt, 0, eq_name, + priv->eq_table.eq + MLX4_EQ_ASYNC); + if (err) + goto err_out_unmap; - priv->eq_table.eq[i].have_irq = 1; - } + priv->eq_table.eq[MLX4_EQ_ASYNC].have_irq = 1; } else { snprintf(priv->eq_table.irq_names, MLX4_IRQNAME_SIZE, @@ -1253,36 +1269,38 @@ int mlx4_init_eq_table(struct mlx4_dev *dev) err = request_irq(dev->persist->pdev->irq, mlx4_interrupt, IRQF_SHARED, priv->eq_table.irq_names, dev); if (err) - goto err_out_async; + goto err_out_unmap; priv->eq_table.have_irq = 1; } err = mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0, - priv->eq_table.eq[dev->caps.num_comp_vectors].eqn); + priv->eq_table.eq[MLX4_EQ_ASYNC].eqn); if (err) mlx4_warn(dev, "MAP_EQ for async EQ %d failed (%d)\n", - priv->eq_table.eq[dev->caps.num_comp_vectors].eqn, err); + priv->eq_table.eq[MLX4_EQ_ASYNC].eqn, err); - for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) - eq_set_ci(&priv->eq_table.eq[i], 1); + /* arm ASYNC eq */ + eq_set_ci(&priv->eq_table.eq[MLX4_EQ_ASYNC], 1); return 0; -err_out_async: - mlx4_free_eq(dev, &priv->eq_table.eq[dev->caps.num_comp_vectors]); - -err_out_comp: - i = dev->caps.num_comp_vectors - 1; - err_out_unmap: - while (i >= 0) { - mlx4_free_eq(dev, &priv->eq_table.eq[i]); - --i; + while (i >= 0) + mlx4_free_eq(dev, &priv->eq_table.eq[i--]); +#ifdef CONFIG_RFS_ACCEL + for (i = 1; i <= dev->caps.num_ports; i++) { + if (mlx4_priv(dev)->port[i].rmap) { + free_irq_cpu_rmap(mlx4_priv(dev)->port[i].rmap); + mlx4_priv(dev)->port[i].rmap = NULL; + } } +#endif + mlx4_free_irqs(dev); + +err_out_clr_int: if (!mlx4_is_slave(dev)) mlx4_unmap_clr_int(dev); - mlx4_free_irqs(dev); err_out_bitmap: mlx4_unmap_uar(dev); @@ -1300,11 +1318,19 @@ void mlx4_cleanup_eq_table(struct mlx4_dev *dev) int i; mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 1, - priv->eq_table.eq[dev->caps.num_comp_vectors].eqn); + priv->eq_table.eq[MLX4_EQ_ASYNC].eqn); +#ifdef CONFIG_RFS_ACCEL + for (i = 1; i <= dev->caps.num_ports; i++) { + if (mlx4_priv(dev)->port[i].rmap) { + free_irq_cpu_rmap(mlx4_priv(dev)->port[i].rmap); + mlx4_priv(dev)->port[i].rmap = NULL; + } + } +#endif mlx4_free_irqs(dev); - for (i = 0; i < dev->caps.num_comp_vectors + dev->caps.comp_pool + 1; ++i) + for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) mlx4_free_eq(dev, &priv->eq_table.eq[i]); if (!mlx4_is_slave(dev)) @@ -1355,87 +1381,169 @@ int mlx4_test_interrupts(struct mlx4_dev *dev) /* Return to default */ mlx4_MAP_EQ(dev, get_async_ev_mask(dev), 0, - priv->eq_table.eq[dev->caps.num_comp_vectors].eqn); + priv->eq_table.eq[MLX4_EQ_ASYNC].eqn); return err; } EXPORT_SYMBOL(mlx4_test_interrupts); -int mlx4_assign_eq(struct mlx4_dev *dev, char *name, struct cpu_rmap *rmap, - int *vector) +bool mlx4_is_eq_vector_valid(struct mlx4_dev *dev, u8 port, int vector) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + + vector = MLX4_CQ_TO_EQ_VECTOR(vector); + if (vector < 0 || (vector >= dev->caps.num_comp_vectors + 1) || + (vector == MLX4_EQ_ASYNC)) + return false; + + return test_bit(port - 1, priv->eq_table.eq[vector].actv_ports.ports); +} +EXPORT_SYMBOL(mlx4_is_eq_vector_valid); + +u32 mlx4_get_eqs_per_port(struct mlx4_dev *dev, u8 port) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + unsigned int i; + unsigned int sum = 0; + + for (i = 0; i < dev->caps.num_comp_vectors + 1; i++) + sum += !!test_bit(port - 1, + priv->eq_table.eq[i].actv_ports.ports); + + return sum; +} +EXPORT_SYMBOL(mlx4_get_eqs_per_port); + +int mlx4_is_eq_shared(struct mlx4_dev *dev, int vector) { + struct mlx4_priv *priv = mlx4_priv(dev); + + vector = MLX4_CQ_TO_EQ_VECTOR(vector); + if (vector <= 0 || (vector >= dev->caps.num_comp_vectors + 1)) + return -EINVAL; + + return !!(bitmap_weight(priv->eq_table.eq[vector].actv_ports.ports, + dev->caps.num_ports) > 1); +} +EXPORT_SYMBOL(mlx4_is_eq_shared); +struct cpu_rmap *mlx4_get_cpu_rmap(struct mlx4_dev *dev, int port) +{ + return mlx4_priv(dev)->port[port].rmap; +} +EXPORT_SYMBOL(mlx4_get_cpu_rmap); + +int mlx4_assign_eq(struct mlx4_dev *dev, u8 port, int *vector) +{ struct mlx4_priv *priv = mlx4_priv(dev); - int vec = 0, err = 0, i; + int err = 0, i = 0; + u32 min_ref_count_val = (u32)-1; + int requested_vector = MLX4_CQ_TO_EQ_VECTOR(*vector); + int *prequested_vector = NULL; + mutex_lock(&priv->msix_ctl.pool_lock); - for (i = 0; !vec && i < dev->caps.comp_pool; i++) { - if (~priv->msix_ctl.pool_bm & 1ULL << i) { - priv->msix_ctl.pool_bm |= 1ULL << i; - vec = dev->caps.num_comp_vectors + 1 + i; - snprintf(priv->eq_table.irq_names + - vec * MLX4_IRQNAME_SIZE, - MLX4_IRQNAME_SIZE, "%s", name); -#ifdef CONFIG_RFS_ACCEL - if (rmap) { - err = irq_cpu_rmap_add(rmap, - priv->eq_table.eq[vec].irq); - if (err) - mlx4_warn(dev, "Failed adding irq rmap\n"); + if (requested_vector < (dev->caps.num_comp_vectors + 1) && + (requested_vector >= 0) && + (requested_vector != MLX4_EQ_ASYNC)) { + if (test_bit(port - 1, + priv->eq_table.eq[requested_vector].actv_ports.ports)) { + prequested_vector = &requested_vector; + } else { + struct mlx4_eq *eq; + + for (i = 1; i < port; + requested_vector += mlx4_get_eqs_per_port(dev, i++)) + ; + + eq = &priv->eq_table.eq[requested_vector]; + if (requested_vector < dev->caps.num_comp_vectors + 1 && + test_bit(port - 1, eq->actv_ports.ports)) { + prequested_vector = &requested_vector; } -#endif - err = request_irq(priv->eq_table.eq[vec].irq, - mlx4_msi_x_interrupt, 0, - &priv->eq_table.irq_names[vec<<5], - priv->eq_table.eq + vec); - if (err) { - /*zero out bit by fliping it*/ - priv->msix_ctl.pool_bm ^= 1 << i; - vec = 0; - continue; - /*we dont want to break here*/ + } + } + + if (!prequested_vector) { + requested_vector = -1; + for (i = 0; min_ref_count_val && i < dev->caps.num_comp_vectors + 1; + i++) { + struct mlx4_eq *eq = &priv->eq_table.eq[i]; + + if (min_ref_count_val > eq->ref_count && + test_bit(port - 1, eq->actv_ports.ports)) { + min_ref_count_val = eq->ref_count; + requested_vector = i; } + } + + if (requested_vector < 0) { + err = -ENOSPC; + goto err_unlock; + } + + prequested_vector = &requested_vector; + } + + if (!test_bit(*prequested_vector, priv->msix_ctl.pool_bm) && + dev->flags & MLX4_FLAG_MSI_X) { + set_bit(*prequested_vector, priv->msix_ctl.pool_bm); + snprintf(priv->eq_table.irq_names + + *prequested_vector * MLX4_IRQNAME_SIZE, + MLX4_IRQNAME_SIZE, "mlx4-%d@%s", + *prequested_vector, dev_name(&dev->persist->pdev->dev)); - eq_set_ci(&priv->eq_table.eq[vec], 1); + err = request_irq(priv->eq_table.eq[*prequested_vector].irq, + mlx4_msi_x_interrupt, 0, + &priv->eq_table.irq_names[*prequested_vector << 5], + priv->eq_table.eq + *prequested_vector); + + if (err) { + clear_bit(*prequested_vector, priv->msix_ctl.pool_bm); + *prequested_vector = -1; + } else { +#if defined(CONFIG_SMP) + mlx4_set_eq_affinity_hint(priv, *prequested_vector); +#endif + eq_set_ci(&priv->eq_table.eq[*prequested_vector], 1); + priv->eq_table.eq[*prequested_vector].have_irq = 1; } } + + if (!err && *prequested_vector >= 0) + priv->eq_table.eq[*prequested_vector].ref_count++; + +err_unlock: mutex_unlock(&priv->msix_ctl.pool_lock); - if (vec) { - *vector = vec; - } else { + if (!err && *prequested_vector >= 0) + *vector = MLX4_EQ_TO_CQ_VECTOR(*prequested_vector); + else *vector = 0; - err = (i == dev->caps.comp_pool) ? -ENOSPC : err; - } + return err; } EXPORT_SYMBOL(mlx4_assign_eq); -int mlx4_eq_get_irq(struct mlx4_dev *dev, int vec) +int mlx4_eq_get_irq(struct mlx4_dev *dev, int cq_vec) { struct mlx4_priv *priv = mlx4_priv(dev); - return priv->eq_table.eq[vec].irq; + return priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(cq_vec)].irq; } EXPORT_SYMBOL(mlx4_eq_get_irq); void mlx4_release_eq(struct mlx4_dev *dev, int vec) { struct mlx4_priv *priv = mlx4_priv(dev); - /*bm index*/ - int i = vec - dev->caps.num_comp_vectors - 1; - - if (likely(i >= 0)) { - /*sanity check , making sure were not trying to free irq's - Belonging to a legacy EQ*/ - mutex_lock(&priv->msix_ctl.pool_lock); - if (priv->msix_ctl.pool_bm & 1ULL << i) { - free_irq(priv->eq_table.eq[vec].irq, - &priv->eq_table.eq[vec]); - priv->msix_ctl.pool_bm &= ~(1ULL << i); - } - mutex_unlock(&priv->msix_ctl.pool_lock); - } + int eq_vec = MLX4_CQ_TO_EQ_VECTOR(vec); + mutex_lock(&priv->msix_ctl.pool_lock); + priv->eq_table.eq[eq_vec].ref_count--; + + /* once we allocated EQ, we don't release it because it might be binded + * to cpu_rmap. + */ + mutex_unlock(&priv->msix_ctl.pool_lock); } EXPORT_SYMBOL(mlx4_release_eq); diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index ced5ecab5aa7..0dbd70427221 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -2364,11 +2364,11 @@ static int mlx4_setup_hca(struct mlx4_dev *dev) if (err) { if (dev->flags & MLX4_FLAG_MSI_X) { mlx4_warn(dev, "NOP command failed to generate MSI-X interrupt IRQ %d)\n", - priv->eq_table.eq[dev->caps.num_comp_vectors].irq); + priv->eq_table.eq[MLX4_EQ_ASYNC].irq); mlx4_warn(dev, "Trying again without MSI-X\n"); } else { mlx4_err(dev, "NOP command failed to generate interrupt (IRQ %d), aborting\n", - priv->eq_table.eq[dev->caps.num_comp_vectors].irq); + priv->eq_table.eq[MLX4_EQ_ASYNC].irq); mlx4_err(dev, "BIOS or ACPI interrupt routing problem?\n"); } @@ -2481,14 +2481,45 @@ err_uar_table_free: return err; } +static int mlx4_init_affinity_hint(struct mlx4_dev *dev, int port, int eqn) +{ + int requested_cpu = 0; + struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_eq *eq; + int off = 0; + int i; + + if (eqn > dev->caps.num_comp_vectors) + return -EINVAL; + + for (i = 1; i < port; i++) + off += mlx4_get_eqs_per_port(dev, i); + + requested_cpu = eqn - off - !!(eqn > MLX4_EQ_ASYNC); + + /* Meaning EQs are shared, and this call comes from the second port */ + if (requested_cpu < 0) + return 0; + + eq = &priv->eq_table.eq[eqn]; + + if (!zalloc_cpumask_var(&eq->affinity_mask, GFP_KERNEL)) + return -ENOMEM; + + cpumask_set_cpu(requested_cpu, eq->affinity_mask); + + return 0; +} + static void mlx4_enable_msi_x(struct mlx4_dev *dev) { struct mlx4_priv *priv = mlx4_priv(dev); struct msix_entry *entries; int i; + int port = 0; if (msi_x) { - int nreq = dev->caps.num_ports * num_online_cpus() + MSIX_LEGACY_SZ; + int nreq = dev->caps.num_ports * num_online_cpus() + 1; nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs, nreq); @@ -2503,20 +2534,55 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) nreq = pci_enable_msix_range(dev->persist->pdev, entries, 2, nreq); - if (nreq < 0) { + if (nreq < 0 || nreq < MLX4_EQ_ASYNC) { kfree(entries); goto no_msi; - } else if (nreq < MSIX_LEGACY_SZ + - dev->caps.num_ports * MIN_MSIX_P_PORT) { - /*Working in legacy mode , all EQ's shared*/ - dev->caps.comp_pool = 0; - dev->caps.num_comp_vectors = nreq - 1; - } else { - dev->caps.comp_pool = nreq - MSIX_LEGACY_SZ; - dev->caps.num_comp_vectors = MSIX_LEGACY_SZ - 1; } - for (i = 0; i < nreq; ++i) - priv->eq_table.eq[i].irq = entries[i].vector; + /* 1 is reserved for events (asyncrounous EQ) */ + dev->caps.num_comp_vectors = nreq - 1; + + priv->eq_table.eq[MLX4_EQ_ASYNC].irq = entries[0].vector; + bitmap_zero(priv->eq_table.eq[MLX4_EQ_ASYNC].actv_ports.ports, + dev->caps.num_ports); + + for (i = 0; i < dev->caps.num_comp_vectors + 1; i++) { + if (i == MLX4_EQ_ASYNC) + continue; + + priv->eq_table.eq[i].irq = + entries[i + 1 - !!(i > MLX4_EQ_ASYNC)].vector; + + if (MLX4_IS_LEGACY_EQ_MODE(dev->caps)) { + bitmap_fill(priv->eq_table.eq[i].actv_ports.ports, + dev->caps.num_ports); + /* We don't set affinity hint when there + * aren't enough EQs + */ + } else { + set_bit(port, + priv->eq_table.eq[i].actv_ports.ports); + if (mlx4_init_affinity_hint(dev, port + 1, i)) + mlx4_warn(dev, "Couldn't init hint cpumask for EQ %d\n", + i); + } + /* We divide the Eqs evenly between the two ports. + * (dev->caps.num_comp_vectors / dev->caps.num_ports) + * refers to the number of Eqs per port + * (i.e eqs_per_port). Theoretically, we would like to + * write something like (i + 1) % eqs_per_port == 0. + * However, since there's an asynchronous Eq, we have + * to skip over it by comparing this condition to + * !!((i + 1) > MLX4_EQ_ASYNC). + */ + if ((dev->caps.num_comp_vectors > dev->caps.num_ports) && + ((i + 1) % + (dev->caps.num_comp_vectors / dev->caps.num_ports)) == + !!((i + 1) > MLX4_EQ_ASYNC)) + /* If dev->caps.num_comp_vectors < dev->caps.num_ports, + * everything is shared anyway. + */ + port++; + } dev->flags |= MLX4_FLAG_MSI_X; @@ -2526,10 +2592,15 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) no_msi: dev->caps.num_comp_vectors = 1; - dev->caps.comp_pool = 0; - for (i = 0; i < 2; ++i) + BUG_ON(MLX4_EQ_ASYNC >= 2); + for (i = 0; i < 2; ++i) { priv->eq_table.eq[i].irq = dev->persist->pdev->irq; + if (i != MLX4_EQ_ASYNC) { + bitmap_fill(priv->eq_table.eq[i].actv_ports.ports, + dev->caps.num_ports); + } + } } static int mlx4_init_port_info(struct mlx4_dev *dev, int port) @@ -2594,6 +2665,10 @@ static void mlx4_cleanup_port_info(struct mlx4_port_info *info) device_remove_file(&info->dev->persist->pdev->dev, &info->port_attr); device_remove_file(&info->dev->persist->pdev->dev, &info->port_mtu_attr); +#ifdef CONFIG_RFS_ACCEL + free_irq_cpu_rmap(info->rmap); + info->rmap = NULL; +#endif } static int mlx4_init_steering(struct mlx4_dev *dev) @@ -2988,18 +3063,6 @@ slave_start: /* In master functions, the communication channel must be initialized * after obtaining its address from fw */ if (mlx4_is_master(dev)) { - int ib_ports = 0; - - mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB) - ib_ports++; - - if (ib_ports && - (num_vfs_argc > 1 || probe_vfs_argc > 1)) { - mlx4_err(dev, - "Invalid syntax of num_vfs/probe_vfs with IB port - single port VFs syntax is only supported when all ports are configured as ethernet\n"); - err = -EINVAL; - goto err_close; - } if (dev->caps.num_ports < 2 && num_vfs_argc > 1) { err = -EINVAL; @@ -3036,7 +3099,7 @@ slave_start: if (err) goto err_master_mfunc; - priv->msix_ctl.pool_bm = 0; + bitmap_zero(priv->msix_ctl.pool_bm, MAX_MSIX); mutex_init(&priv->msix_ctl.pool_lock); mlx4_enable_msi_x(dev); @@ -3058,7 +3121,6 @@ slave_start: !mlx4_is_mfunc(dev)) { dev->flags &= ~MLX4_FLAG_MSI_X; dev->caps.num_comp_vectors = 1; - dev->caps.comp_pool = 0; pci_disable_msix(pdev); err = mlx4_setup_hca(dev); } diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 502d3dd2c888..f424900d23a6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -287,6 +287,12 @@ struct mlx4_icm_table { #define MLX4_CQE_SIZE_MASK_STRIDE 0x3 #define MLX4_EQE_SIZE_MASK_STRIDE 0x30 +#define MLX4_EQ_ASYNC 0 +#define MLX4_EQ_TO_CQ_VECTOR(vector) ((vector) - \ + !!((int)(vector) >= MLX4_EQ_ASYNC)) +#define MLX4_CQ_TO_EQ_VECTOR(vector) ((vector) + \ + !!((int)(vector) >= MLX4_EQ_ASYNC)) + /* * Must be packed because mtt_seg is 64 bits but only aligned to 32 bits. */ @@ -391,6 +397,9 @@ struct mlx4_eq { struct mlx4_buf_list *page_list; struct mlx4_mtt mtt; struct mlx4_eq_tasklet tasklet_ctx; + struct mlx4_active_ports actv_ports; + u32 ref_count; + cpumask_var_t affinity_mask; }; struct mlx4_slave_eqe { @@ -808,6 +817,7 @@ struct mlx4_port_info { struct mlx4_vlan_table vlan_table; struct mlx4_roce_gid_table gid_table; int base_qpn; + struct cpu_rmap *rmap; }; struct mlx4_sense { @@ -818,7 +828,7 @@ struct mlx4_sense { }; struct mlx4_msix_ctl { - u64 pool_bm; + DECLARE_BITMAP(pool_bm, MAX_MSIX); struct mutex pool_lock; }; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index d021f079f181..edd8fd69ec9a 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -338,7 +338,7 @@ struct mlx4_en_cq { struct napi_struct napi; int size; int buf_size; - unsigned vector; + int vector; enum cq_type is_tx; u16 moder_time; u16 moder_cnt; diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index bafe2180cf0c..ab48386bfefc 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -2703,6 +2703,10 @@ static void adjust_proxy_tun_qkey(struct mlx4_dev *dev, struct mlx4_vhcr *vhcr, context->qkey = cpu_to_be32(qkey); } +static int adjust_qp_sched_queue(struct mlx4_dev *dev, int slave, + struct mlx4_qp_context *qpc, + struct mlx4_cmd_mailbox *inbox); + int mlx4_RST2INIT_QP_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr, struct mlx4_cmd_mailbox *inbox, @@ -2725,6 +2729,10 @@ int mlx4_RST2INIT_QP_wrapper(struct mlx4_dev *dev, int slave, struct res_srq *srq; int local_qpn = be32_to_cpu(qpc->local_qpn) & 0xffffff; + err = adjust_qp_sched_queue(dev, slave, qpc, inbox); + if (err) + return err; + err = qp_res_start_move_to(dev, slave, qpn, RES_QP_HW, &qp, 0); if (err) return err; @@ -3526,8 +3534,8 @@ static int adjust_qp_sched_queue(struct mlx4_dev *dev, int slave, pri_sched_queue = (qpc->pri_path.sched_queue & ~(1 << 6)) | ((port & 1) << 6); - if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH || - mlx4_is_eth(dev, port + 1)) { + if (optpar & (MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH | MLX4_QP_OPTPAR_SCHED_QUEUE) || + qpc->pri_path.sched_queue || mlx4_is_eth(dev, port + 1)) { qpc->pri_path.sched_queue = pri_sched_queue; } @@ -3965,6 +3973,22 @@ static int validate_eth_header_mac(int slave, struct _rule_hw *eth_header, return 0; } +static void handle_eth_header_mcast_prio(struct mlx4_net_trans_rule_hw_ctrl *ctrl, + struct _rule_hw *eth_header) +{ + if (is_multicast_ether_addr(eth_header->eth.dst_mac) || + is_broadcast_ether_addr(eth_header->eth.dst_mac)) { + struct mlx4_net_trans_rule_hw_eth *eth = + (struct mlx4_net_trans_rule_hw_eth *)eth_header; + struct _rule_hw *next_rule = (struct _rule_hw *)(eth + 1); + bool last_rule = next_rule->size == 0 && next_rule->id == 0 && + next_rule->rsvd == 0; + + if (last_rule) + ctrl->prio = cpu_to_be16(MLX4_DOMAIN_NIC); + } +} + /* * In case of missing eth header, append eth header with a MAC address * assigned to the VF. @@ -4117,6 +4141,12 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave, rule_header = (struct _rule_hw *)(ctrl + 1); header_id = map_hw_to_sw_id(be16_to_cpu(rule_header->id)); + if (header_id == MLX4_NET_TRANS_RULE_ID_ETH) + handle_eth_header_mcast_prio(ctrl, rule_header); + + if (slave == dev->caps.function) + goto execute; + switch (header_id) { case MLX4_NET_TRANS_RULE_ID_ETH: if (validate_eth_header_mac(slave, rule_header, rlist)) { @@ -4143,6 +4173,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave, goto err_put; } +execute: err = mlx4_cmd_imm(dev, inbox->dma, &vhcr->out_param, vhcr->in_modifier, 0, MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index 8ff57e8e3e91..0d7aef040fb0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -3,6 +3,18 @@ # config MLX5_CORE - tristate + tristate "Mellanox Technologies ConnectX-4 and Connect-IB core driver" depends on PCI default n + ---help--- + Core driver for low level functionality of the ConnectX-4 and + Connect-IB cards by Mellanox Technologies. + +config MLX5_CORE_EN + bool "Mellanox Technologies ConnectX-4 Ethernet support" + depends on MLX5_INFINIBAND=n && NETDEVICES && ETHERNET && PCI && MLX5_CORE + default n + ---help--- + Ethernet support in Mellanox Technologies ConnectX-4 NIC. + Ethernet and Infiniband support in ConnectX-4 are currently mutually + exclusive. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 105780bb980b..87e9e606596a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -3,3 +3,6 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ mad.o +mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o flow_table.o vport.o transobj.o \ + en_main.o en_flow_table.o en_ethtool.o en_tx.o en_rx.o \ + en_txrx.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c index ac0f7bf4be95..0715b497511f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c @@ -42,95 +42,36 @@ #include "mlx5_core.h" /* Handling for queue buffers -- we allocate a bunch of memory and - * register it in a memory region at HCA virtual address 0. If the - * requested size is > max_direct, we split the allocation into - * multiple pages, so we don't require too much contiguous memory. + * register it in a memory region at HCA virtual address 0. */ -int mlx5_buf_alloc(struct mlx5_core_dev *dev, int size, int max_direct, - struct mlx5_buf *buf) +int mlx5_buf_alloc(struct mlx5_core_dev *dev, int size, struct mlx5_buf *buf) { dma_addr_t t; buf->size = size; - if (size <= max_direct) { - buf->nbufs = 1; - buf->npages = 1; - buf->page_shift = (u8)get_order(size) + PAGE_SHIFT; - buf->direct.buf = dma_zalloc_coherent(&dev->pdev->dev, - size, &t, GFP_KERNEL); - if (!buf->direct.buf) - return -ENOMEM; - - buf->direct.map = t; - - while (t & ((1 << buf->page_shift) - 1)) { - --buf->page_shift; - buf->npages *= 2; - } - } else { - int i; - - buf->direct.buf = NULL; - buf->nbufs = (size + PAGE_SIZE - 1) / PAGE_SIZE; - buf->npages = buf->nbufs; - buf->page_shift = PAGE_SHIFT; - buf->page_list = kcalloc(buf->nbufs, sizeof(*buf->page_list), - GFP_KERNEL); - if (!buf->page_list) - return -ENOMEM; - - for (i = 0; i < buf->nbufs; i++) { - buf->page_list[i].buf = - dma_zalloc_coherent(&dev->pdev->dev, PAGE_SIZE, - &t, GFP_KERNEL); - if (!buf->page_list[i].buf) - goto err_free; - - buf->page_list[i].map = t; - } - - if (BITS_PER_LONG == 64) { - struct page **pages; - pages = kmalloc(sizeof(*pages) * buf->nbufs, GFP_KERNEL); - if (!pages) - goto err_free; - for (i = 0; i < buf->nbufs; i++) - pages[i] = virt_to_page(buf->page_list[i].buf); - buf->direct.buf = vmap(pages, buf->nbufs, VM_MAP, PAGE_KERNEL); - kfree(pages); - if (!buf->direct.buf) - goto err_free; - } - } + buf->npages = 1; + buf->page_shift = (u8)get_order(size) + PAGE_SHIFT; + buf->direct.buf = dma_zalloc_coherent(&dev->pdev->dev, + size, &t, GFP_KERNEL); + if (!buf->direct.buf) + return -ENOMEM; - return 0; + buf->direct.map = t; -err_free: - mlx5_buf_free(dev, buf); + while (t & ((1 << buf->page_shift) - 1)) { + --buf->page_shift; + buf->npages *= 2; + } - return -ENOMEM; + return 0; } EXPORT_SYMBOL_GPL(mlx5_buf_alloc); void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_buf *buf) { - int i; - - if (buf->nbufs == 1) - dma_free_coherent(&dev->pdev->dev, buf->size, buf->direct.buf, - buf->direct.map); - else { - if (BITS_PER_LONG == 64) - vunmap(buf->direct.buf); - - for (i = 0; i < buf->nbufs; i++) - if (buf->page_list[i].buf) - dma_free_coherent(&dev->pdev->dev, PAGE_SIZE, - buf->page_list[i].buf, - buf->page_list[i].map); - kfree(buf->page_list); - } + dma_free_coherent(&dev->pdev->dev, buf->size, buf->direct.buf, + buf->direct.map); } EXPORT_SYMBOL_GPL(mlx5_buf_free); @@ -230,10 +171,7 @@ void mlx5_fill_page_array(struct mlx5_buf *buf, __be64 *pas) int i; for (i = 0; i < buf->npages; i++) { - if (buf->nbufs == 1) - addr = buf->direct.map + (i << buf->page_shift); - else - addr = buf->page_list[i].map; + addr = buf->direct.map + (i << buf->page_shift); pas[i] = cpu_to_be64(addr); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index e3273faf4568..75ff58dc1ff5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -75,25 +75,6 @@ enum { MLX5_CMD_DELIVERY_STAT_CMD_DESCR_ERR = 0x10, }; -enum { - MLX5_CMD_STAT_OK = 0x0, - MLX5_CMD_STAT_INT_ERR = 0x1, - MLX5_CMD_STAT_BAD_OP_ERR = 0x2, - MLX5_CMD_STAT_BAD_PARAM_ERR = 0x3, - MLX5_CMD_STAT_BAD_SYS_STATE_ERR = 0x4, - MLX5_CMD_STAT_BAD_RES_ERR = 0x5, - MLX5_CMD_STAT_RES_BUSY = 0x6, - MLX5_CMD_STAT_LIM_ERR = 0x8, - MLX5_CMD_STAT_BAD_RES_STATE_ERR = 0x9, - MLX5_CMD_STAT_IX_ERR = 0xa, - MLX5_CMD_STAT_NO_RES_ERR = 0xf, - MLX5_CMD_STAT_BAD_INP_LEN_ERR = 0x50, - MLX5_CMD_STAT_BAD_OUTP_LEN_ERR = 0x51, - MLX5_CMD_STAT_BAD_QP_STATE_ERR = 0x10, - MLX5_CMD_STAT_BAD_PKT_ERR = 0x30, - MLX5_CMD_STAT_BAD_SIZE_OUTS_CQES_ERR = 0x40, -}; - static struct mlx5_cmd_work_ent *alloc_cmd(struct mlx5_cmd *cmd, struct mlx5_cmd_msg *in, struct mlx5_cmd_msg *out, @@ -390,8 +371,17 @@ const char *mlx5_command_str(int command) case MLX5_CMD_OP_ARM_RQ: return "ARM_RQ"; - case MLX5_CMD_OP_RESIZE_SRQ: - return "RESIZE_SRQ"; + case MLX5_CMD_OP_CREATE_XRC_SRQ: + return "CREATE_XRC_SRQ"; + + case MLX5_CMD_OP_DESTROY_XRC_SRQ: + return "DESTROY_XRC_SRQ"; + + case MLX5_CMD_OP_QUERY_XRC_SRQ: + return "QUERY_XRC_SRQ"; + + case MLX5_CMD_OP_ARM_XRC_SRQ: + return "ARM_XRC_SRQ"; case MLX5_CMD_OP_ALLOC_PD: return "ALLOC_PD"; @@ -408,8 +398,8 @@ const char *mlx5_command_str(int command) case MLX5_CMD_OP_ATTACH_TO_MCG: return "ATTACH_TO_MCG"; - case MLX5_CMD_OP_DETACH_FROM_MCG: - return "DETACH_FROM_MCG"; + case MLX5_CMD_OP_DETTACH_FROM_MCG: + return "DETTACH_FROM_MCG"; case MLX5_CMD_OP_ALLOC_XRCD: return "ALLOC_XRCD"; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c index eb0cf81f5f45..04ab7e445eae 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c @@ -219,6 +219,24 @@ int mlx5_core_modify_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq, } EXPORT_SYMBOL(mlx5_core_modify_cq); +int mlx5_core_modify_cq_moderation(struct mlx5_core_dev *dev, + struct mlx5_core_cq *cq, + u16 cq_period, + u16 cq_max_count) +{ + struct mlx5_modify_cq_mbox_in in; + + memset(&in, 0, sizeof(in)); + + in.cqn = cpu_to_be32(cq->cqn); + in.ctx.cq_period = cpu_to_be16(cq_period); + in.ctx.cq_max_count = cpu_to_be16(cq_max_count); + in.field_select = cpu_to_be32(MLX5_CQ_MODIFY_PERIOD | + MLX5_CQ_MODIFY_COUNT); + + return mlx5_core_modify_cq(dev, cq, &in, sizeof(in)); +} + int mlx5_init_cq_table(struct mlx5_core_dev *dev) { struct mlx5_cq_table *table = &dev->priv.cq_table; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h new file mode 100644 index 000000000000..cbb3c7cb53f7 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -0,0 +1,520 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/if_vlan.h> +#include <linux/etherdevice.h> +#include <linux/mlx5/driver.h> +#include <linux/mlx5/qp.h> +#include <linux/mlx5/cq.h> +#include "vport.h" +#include "wq.h" +#include "transobj.h" +#include "mlx5_core.h" + +#define MLX5E_MAX_NUM_TC 8 + +#define MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE 0x7 +#define MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE 0xa +#define MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE 0xd + +#define MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE 0x7 +#define MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE 0xa +#define MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE 0xd + +#define MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ (16 * 1024) +#define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC 0x10 +#define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS 0x20 +#define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC 0x10 +#define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS 0x20 +#define MLX5E_PARAMS_DEFAULT_MIN_RX_WQES 0x80 +#define MLX5E_PARAMS_DEFAULT_RX_HASH_LOG_TBL_SZ 0x7 +#define MLX5E_PARAMS_MIN_MTU 46 + +#define MLX5E_TX_CQ_POLL_BUDGET 128 +#define MLX5E_UPDATE_STATS_INTERVAL 200 /* msecs */ + +static const char vport_strings[][ETH_GSTRING_LEN] = { + /* vport statistics */ + "rx_packets", + "rx_bytes", + "tx_packets", + "tx_bytes", + "rx_error_packets", + "rx_error_bytes", + "tx_error_packets", + "tx_error_bytes", + "rx_unicast_packets", + "rx_unicast_bytes", + "tx_unicast_packets", + "tx_unicast_bytes", + "rx_multicast_packets", + "rx_multicast_bytes", + "tx_multicast_packets", + "tx_multicast_bytes", + "rx_broadcast_packets", + "rx_broadcast_bytes", + "tx_broadcast_packets", + "tx_broadcast_bytes", + + /* SW counters */ + "tso_packets", + "tso_bytes", + "lro_packets", + "lro_bytes", + "rx_csum_good", + "rx_csum_none", + "tx_csum_offload", + "tx_queue_stopped", + "tx_queue_wake", + "tx_queue_dropped", + "rx_wqe_err", +}; + +struct mlx5e_vport_stats { + /* HW counters */ + u64 rx_packets; + u64 rx_bytes; + u64 tx_packets; + u64 tx_bytes; + u64 rx_error_packets; + u64 rx_error_bytes; + u64 tx_error_packets; + u64 tx_error_bytes; + u64 rx_unicast_packets; + u64 rx_unicast_bytes; + u64 tx_unicast_packets; + u64 tx_unicast_bytes; + u64 rx_multicast_packets; + u64 rx_multicast_bytes; + u64 tx_multicast_packets; + u64 tx_multicast_bytes; + u64 rx_broadcast_packets; + u64 rx_broadcast_bytes; + u64 tx_broadcast_packets; + u64 tx_broadcast_bytes; + + /* SW counters */ + u64 tso_packets; + u64 tso_bytes; + u64 lro_packets; + u64 lro_bytes; + u64 rx_csum_good; + u64 rx_csum_none; + u64 tx_csum_offload; + u64 tx_queue_stopped; + u64 tx_queue_wake; + u64 tx_queue_dropped; + u64 rx_wqe_err; + +#define NUM_VPORT_COUNTERS 31 +}; + +static const char rq_stats_strings[][ETH_GSTRING_LEN] = { + "packets", + "csum_none", + "lro_packets", + "lro_bytes", + "wqe_err" +}; + +struct mlx5e_rq_stats { + u64 packets; + u64 csum_none; + u64 lro_packets; + u64 lro_bytes; + u64 wqe_err; +#define NUM_RQ_STATS 5 +}; + +static const char sq_stats_strings[][ETH_GSTRING_LEN] = { + "packets", + "tso_packets", + "tso_bytes", + "csum_offload_none", + "stopped", + "wake", + "dropped", + "nop" +}; + +struct mlx5e_sq_stats { + u64 packets; + u64 tso_packets; + u64 tso_bytes; + u64 csum_offload_none; + u64 stopped; + u64 wake; + u64 dropped; + u64 nop; +#define NUM_SQ_STATS 8 +}; + +struct mlx5e_stats { + struct mlx5e_vport_stats vport; +}; + +struct mlx5e_params { + u8 log_sq_size; + u8 log_rq_size; + u16 num_channels; + u8 default_vlan_prio; + u8 num_tc; + u16 rx_cq_moderation_usec; + u16 rx_cq_moderation_pkts; + u16 tx_cq_moderation_usec; + u16 tx_cq_moderation_pkts; + u16 min_rx_wqes; + u16 rx_hash_log_tbl_sz; + bool lro_en; + u32 lro_wqe_sz; +}; + +enum { + MLX5E_RQ_STATE_POST_WQES_ENABLE, +}; + +enum cq_flags { + MLX5E_CQ_HAS_CQES = 1, +}; + +struct mlx5e_cq { + /* data path - accessed per cqe */ + struct mlx5_cqwq wq; + void *sqrq; + unsigned long flags; + + /* data path - accessed per napi poll */ + struct napi_struct *napi; + struct mlx5_core_cq mcq; + struct mlx5e_channel *channel; + + /* control */ + struct mlx5_wq_ctrl wq_ctrl; +} ____cacheline_aligned_in_smp; + +struct mlx5e_rq { + /* data path */ + struct mlx5_wq_ll wq; + u32 wqe_sz; + struct sk_buff **skb; + + struct device *pdev; + struct net_device *netdev; + struct mlx5e_rq_stats stats; + struct mlx5e_cq cq; + + unsigned long state; + int ix; + + /* control */ + struct mlx5_wq_ctrl wq_ctrl; + u32 rqn; + struct mlx5e_channel *channel; +} ____cacheline_aligned_in_smp; + +struct mlx5e_tx_skb_cb { + u32 num_bytes; + u8 num_wqebbs; + u8 num_dma; +}; + +#define MLX5E_TX_SKB_CB(__skb) ((struct mlx5e_tx_skb_cb *)__skb->cb) + +struct mlx5e_sq_dma { + dma_addr_t addr; + u32 size; +}; + +enum { + MLX5E_SQ_STATE_WAKE_TXQ_ENABLE, +}; + +struct mlx5e_sq { + /* data path */ + + /* dirtied @completion */ + u16 cc; + u32 dma_fifo_cc; + + /* dirtied @xmit */ + u16 pc ____cacheline_aligned_in_smp; + u32 dma_fifo_pc; + u32 bf_offset; + struct mlx5e_sq_stats stats; + + struct mlx5e_cq cq; + + /* pointers to per packet info: write@xmit, read@completion */ + struct sk_buff **skb; + struct mlx5e_sq_dma *dma_fifo; + + /* read only */ + struct mlx5_wq_cyc wq; + u32 dma_fifo_mask; + void __iomem *uar_map; + struct netdev_queue *txq; + u32 sqn; + u32 bf_buf_size; + struct device *pdev; + __be32 mkey_be; + unsigned long state; + + /* control path */ + struct mlx5_wq_ctrl wq_ctrl; + struct mlx5_uar uar; + struct mlx5e_channel *channel; + int tc; +} ____cacheline_aligned_in_smp; + +static inline bool mlx5e_sq_has_room_for(struct mlx5e_sq *sq, u16 n) +{ + return (((sq->wq.sz_m1 & (sq->cc - sq->pc)) >= n) || + (sq->cc == sq->pc)); +} + +enum channel_flags { + MLX5E_CHANNEL_NAPI_SCHED = 1, +}; + +struct mlx5e_channel { + /* data path */ + struct mlx5e_rq rq; + struct mlx5e_sq sq[MLX5E_MAX_NUM_TC]; + struct napi_struct napi; + struct device *pdev; + struct net_device *netdev; + __be32 mkey_be; + u8 num_tc; + unsigned long flags; + + /* control */ + struct mlx5e_priv *priv; + int ix; + int cpu; +}; + +enum mlx5e_traffic_types { + MLX5E_TT_IPV4_TCP = 0, + MLX5E_TT_IPV6_TCP = 1, + MLX5E_TT_IPV4_UDP = 2, + MLX5E_TT_IPV6_UDP = 3, + MLX5E_TT_IPV4 = 4, + MLX5E_TT_IPV6 = 5, + MLX5E_TT_ANY = 6, + MLX5E_NUM_TT = 7, +}; + +enum { + MLX5E_RQT_SPREADING = 0, + MLX5E_RQT_DEFAULT_RQ = 1, + MLX5E_NUM_RQT = 2, +}; + +struct mlx5e_eth_addr_info { + u8 addr[ETH_ALEN + 2]; + u32 tt_vec; + u32 ft_ix[MLX5E_NUM_TT]; /* flow table index per traffic type */ +}; + +#define MLX5E_ETH_ADDR_HASH_SIZE (1 << BITS_PER_BYTE) + +struct mlx5e_eth_addr_db { + struct hlist_head netdev_uc[MLX5E_ETH_ADDR_HASH_SIZE]; + struct hlist_head netdev_mc[MLX5E_ETH_ADDR_HASH_SIZE]; + struct mlx5e_eth_addr_info broadcast; + struct mlx5e_eth_addr_info allmulti; + struct mlx5e_eth_addr_info promisc; + bool broadcast_enabled; + bool allmulti_enabled; + bool promisc_enabled; +}; + +enum { + MLX5E_STATE_ASYNC_EVENTS_ENABLE, + MLX5E_STATE_OPENED, +}; + +struct mlx5e_vlan_db { + unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; + u32 active_vlans_ft_ix[VLAN_N_VID]; + u32 untagged_rule_ft_ix; + u32 any_vlan_rule_ft_ix; + bool filter_disabled; +}; + +struct mlx5e_flow_table { + void *vlan; + void *main; +}; + +struct mlx5e_priv { + /* priv data path fields - start */ + int order_base_2_num_channels; + int queue_mapping_channel_mask; + int num_tc; + int default_vlan_prio; + /* priv data path fields - end */ + + unsigned long state; + struct mutex state_lock; /* Protects Interface state */ + struct mlx5_uar cq_uar; + u32 pdn; + struct mlx5_core_mr mr; + + struct mlx5e_channel **channel; + u32 tisn[MLX5E_MAX_NUM_TC]; + u32 rqtn; + u32 tirn[MLX5E_NUM_TT]; + + struct mlx5e_flow_table ft; + struct mlx5e_eth_addr_db eth_addr; + struct mlx5e_vlan_db vlan; + + struct mlx5e_params params; + spinlock_t async_events_spinlock; /* sync hw events */ + struct work_struct update_carrier_work; + struct work_struct set_rx_mode_work; + struct delayed_work update_stats_work; + + struct mlx5_core_dev *mdev; + struct net_device *netdev; + struct mlx5e_stats stats; +}; + +#define MLX5E_NET_IP_ALIGN 2 + +struct mlx5e_tx_wqe { + struct mlx5_wqe_ctrl_seg ctrl; + struct mlx5_wqe_eth_seg eth; +}; + +struct mlx5e_rx_wqe { + struct mlx5_wqe_srq_next_seg next; + struct mlx5_wqe_data_seg data; +}; + +enum mlx5e_link_mode { + MLX5E_1000BASE_CX_SGMII = 0, + MLX5E_1000BASE_KX = 1, + MLX5E_10GBASE_CX4 = 2, + MLX5E_10GBASE_KX4 = 3, + MLX5E_10GBASE_KR = 4, + MLX5E_20GBASE_KR2 = 5, + MLX5E_40GBASE_CR4 = 6, + MLX5E_40GBASE_KR4 = 7, + MLX5E_56GBASE_R4 = 8, + MLX5E_10GBASE_CR = 12, + MLX5E_10GBASE_SR = 13, + MLX5E_10GBASE_ER = 14, + MLX5E_40GBASE_SR4 = 15, + MLX5E_40GBASE_LR4 = 16, + MLX5E_100GBASE_CR4 = 20, + MLX5E_100GBASE_SR4 = 21, + MLX5E_100GBASE_KR4 = 22, + MLX5E_100GBASE_LR4 = 23, + MLX5E_100BASE_TX = 24, + MLX5E_100BASE_T = 25, + MLX5E_10GBASE_T = 26, + MLX5E_25GBASE_CR = 27, + MLX5E_25GBASE_KR = 28, + MLX5E_25GBASE_SR = 29, + MLX5E_50GBASE_CR2 = 30, + MLX5E_50GBASE_KR2 = 31, + MLX5E_LINK_MODES_NUMBER, +}; + +#define MLX5E_PROT_MASK(link_mode) (1 << link_mode) + +u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, + void *accel_priv, select_queue_fallback_t fallback); +netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev); +netdev_tx_t mlx5e_xmit_multi_tc(struct sk_buff *skb, struct net_device *dev); + +void mlx5e_completion_event(struct mlx5_core_cq *mcq); +void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event); +int mlx5e_napi_poll(struct napi_struct *napi, int budget); +bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq); +bool mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget); +bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq); +struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq); + +void mlx5e_update_stats(struct mlx5e_priv *priv); + +int mlx5e_open_flow_table(struct mlx5e_priv *priv); +void mlx5e_close_flow_table(struct mlx5e_priv *priv); +void mlx5e_init_eth_addr(struct mlx5e_priv *priv); +void mlx5e_set_rx_mode_core(struct mlx5e_priv *priv); +void mlx5e_set_rx_mode_work(struct work_struct *work); + +int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto, + u16 vid); +int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __always_unused __be16 proto, + u16 vid); +void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv); +void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv); +int mlx5e_add_all_vlan_rules(struct mlx5e_priv *priv); +void mlx5e_del_all_vlan_rules(struct mlx5e_priv *priv); + +int mlx5e_open_locked(struct net_device *netdev); +int mlx5e_close_locked(struct net_device *netdev); +int mlx5e_update_priv_params(struct mlx5e_priv *priv, + struct mlx5e_params *new_params); + +static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq, + struct mlx5e_tx_wqe *wqe) +{ + /* ensure wqe is visible to device before updating doorbell record */ + dma_wmb(); + + *sq->wq.db = cpu_to_be32(sq->pc); + + /* ensure doorbell record is visible to device before ringing the + * doorbell + */ + wmb(); + + mlx5_write64((__be32 *)&wqe->ctrl, + sq->uar_map + MLX5_BF_OFFSET + sq->bf_offset, + NULL); + + sq->bf_offset ^= sq->bf_buf_size; +} + +static inline void mlx5e_cq_arm(struct mlx5e_cq *cq) +{ + struct mlx5_core_cq *mcq; + + mcq = &cq->mcq; + mlx5_cq_arm(mcq, MLX5_CQ_DB_REQ_NOT, mcq->uar->map, NULL, cq->wq.cc); +} + +extern const struct ethtool_ops mlx5e_ethtool_ops; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c new file mode 100644 index 000000000000..de7aec8abca1 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -0,0 +1,679 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +static void mlx5e_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *drvinfo) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_core_dev *mdev = priv->mdev; + + strlcpy(drvinfo->driver, DRIVER_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRIVER_VERSION " (" DRIVER_RELDATE ")", + sizeof(drvinfo->version)); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", + fw_rev_maj(mdev), fw_rev_min(mdev), fw_rev_sub(mdev)); + strlcpy(drvinfo->bus_info, pci_name(mdev->pdev), + sizeof(drvinfo->bus_info)); +} + +static const struct { + u32 supported; + u32 advertised; + u32 speed; +} ptys2ethtool_table[MLX5E_LINK_MODES_NUMBER] = { + [MLX5E_1000BASE_CX_SGMII] = { + .supported = SUPPORTED_1000baseKX_Full, + .advertised = ADVERTISED_1000baseKX_Full, + .speed = 1000, + }, + [MLX5E_1000BASE_KX] = { + .supported = SUPPORTED_1000baseKX_Full, + .advertised = ADVERTISED_1000baseKX_Full, + .speed = 1000, + }, + [MLX5E_10GBASE_CX4] = { + .supported = SUPPORTED_10000baseKX4_Full, + .advertised = ADVERTISED_10000baseKX4_Full, + .speed = 10000, + }, + [MLX5E_10GBASE_KX4] = { + .supported = SUPPORTED_10000baseKX4_Full, + .advertised = ADVERTISED_10000baseKX4_Full, + .speed = 10000, + }, + [MLX5E_10GBASE_KR] = { + .supported = SUPPORTED_10000baseKR_Full, + .advertised = ADVERTISED_10000baseKR_Full, + .speed = 10000, + }, + [MLX5E_20GBASE_KR2] = { + .supported = SUPPORTED_20000baseKR2_Full, + .advertised = ADVERTISED_20000baseKR2_Full, + .speed = 20000, + }, + [MLX5E_40GBASE_CR4] = { + .supported = SUPPORTED_40000baseCR4_Full, + .advertised = ADVERTISED_40000baseCR4_Full, + .speed = 40000, + }, + [MLX5E_40GBASE_KR4] = { + .supported = SUPPORTED_40000baseKR4_Full, + .advertised = ADVERTISED_40000baseKR4_Full, + .speed = 40000, + }, + [MLX5E_56GBASE_R4] = { + .supported = SUPPORTED_56000baseKR4_Full, + .advertised = ADVERTISED_56000baseKR4_Full, + .speed = 56000, + }, + [MLX5E_10GBASE_CR] = { + .supported = SUPPORTED_10000baseKR_Full, + .advertised = ADVERTISED_10000baseKR_Full, + .speed = 10000, + }, + [MLX5E_10GBASE_SR] = { + .supported = SUPPORTED_10000baseKR_Full, + .advertised = ADVERTISED_10000baseKR_Full, + .speed = 10000, + }, + [MLX5E_10GBASE_ER] = { + .supported = SUPPORTED_10000baseKR_Full, + .advertised = ADVERTISED_10000baseKR_Full, + .speed = 10000, + }, + [MLX5E_40GBASE_SR4] = { + .supported = SUPPORTED_40000baseSR4_Full, + .advertised = ADVERTISED_40000baseSR4_Full, + .speed = 40000, + }, + [MLX5E_40GBASE_LR4] = { + .supported = SUPPORTED_40000baseLR4_Full, + .advertised = ADVERTISED_40000baseLR4_Full, + .speed = 40000, + }, + [MLX5E_100GBASE_CR4] = { + .speed = 100000, + }, + [MLX5E_100GBASE_SR4] = { + .speed = 100000, + }, + [MLX5E_100GBASE_KR4] = { + .speed = 100000, + }, + [MLX5E_100GBASE_LR4] = { + .speed = 100000, + }, + [MLX5E_100BASE_TX] = { + .speed = 100, + }, + [MLX5E_100BASE_T] = { + .supported = SUPPORTED_100baseT_Full, + .advertised = ADVERTISED_100baseT_Full, + .speed = 100, + }, + [MLX5E_10GBASE_T] = { + .supported = SUPPORTED_10000baseT_Full, + .advertised = ADVERTISED_10000baseT_Full, + .speed = 1000, + }, + [MLX5E_25GBASE_CR] = { + .speed = 25000, + }, + [MLX5E_25GBASE_KR] = { + .speed = 25000, + }, + [MLX5E_25GBASE_SR] = { + .speed = 25000, + }, + [MLX5E_50GBASE_CR2] = { + .speed = 50000, + }, + [MLX5E_50GBASE_KR2] = { + .speed = 50000, + }, +}; + +static int mlx5e_get_sset_count(struct net_device *dev, int sset) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + switch (sset) { + case ETH_SS_STATS: + return NUM_VPORT_COUNTERS + + priv->params.num_channels * NUM_RQ_STATS + + priv->params.num_channels * priv->num_tc * + NUM_SQ_STATS; + /* fallthrough */ + default: + return -EOPNOTSUPP; + } +} + +static void mlx5e_get_strings(struct net_device *dev, + uint32_t stringset, uint8_t *data) +{ + int i, j, tc, idx = 0; + struct mlx5e_priv *priv = netdev_priv(dev); + + switch (stringset) { + case ETH_SS_PRIV_FLAGS: + break; + + case ETH_SS_TEST: + break; + + case ETH_SS_STATS: + /* VPORT counters */ + for (i = 0; i < NUM_VPORT_COUNTERS; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + vport_strings[i]); + + /* per channel counters */ + for (i = 0; i < priv->params.num_channels; i++) + for (j = 0; j < NUM_RQ_STATS; j++) + sprintf(data + (idx++) * ETH_GSTRING_LEN, + "rx%d_%s", i, rq_stats_strings[j]); + + for (i = 0; i < priv->params.num_channels; i++) + for (tc = 0; tc < priv->num_tc; tc++) + for (j = 0; j < NUM_SQ_STATS; j++) + sprintf(data + + (idx++) * ETH_GSTRING_LEN, + "tx%d_%d_%s", i, tc, + sq_stats_strings[j]); + break; + } +} + +static void mlx5e_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int i, j, tc, idx = 0; + + if (!data) + return; + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_update_stats(priv); + mutex_unlock(&priv->state_lock); + + for (i = 0; i < NUM_VPORT_COUNTERS; i++) + data[idx++] = ((u64 *)&priv->stats.vport)[i]; + + /* per channel counters */ + for (i = 0; i < priv->params.num_channels; i++) + for (j = 0; j < NUM_RQ_STATS; j++) + data[idx++] = !test_bit(MLX5E_STATE_OPENED, + &priv->state) ? 0 : + ((u64 *)&priv->channel[i]->rq.stats)[j]; + + for (i = 0; i < priv->params.num_channels; i++) + for (tc = 0; tc < priv->num_tc; tc++) + for (j = 0; j < NUM_SQ_STATS; j++) + data[idx++] = !test_bit(MLX5E_STATE_OPENED, + &priv->state) ? 0 : + ((u64 *)&priv->channel[i]->sq[tc].stats)[j]; +} + +static void mlx5e_get_ringparam(struct net_device *dev, + struct ethtool_ringparam *param) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + param->rx_max_pending = 1 << MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE; + param->tx_max_pending = 1 << MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE; + param->rx_pending = 1 << priv->params.log_rq_size; + param->tx_pending = 1 << priv->params.log_sq_size; +} + +static int mlx5e_set_ringparam(struct net_device *dev, + struct ethtool_ringparam *param) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_params new_params; + u16 min_rx_wqes; + u8 log_rq_size; + u8 log_sq_size; + int err = 0; + + if (param->rx_jumbo_pending) { + netdev_info(dev, "%s: rx_jumbo_pending not supported\n", + __func__); + return -EINVAL; + } + if (param->rx_mini_pending) { + netdev_info(dev, "%s: rx_mini_pending not supported\n", + __func__); + return -EINVAL; + } + if (param->rx_pending < (1 << MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE)) { + netdev_info(dev, "%s: rx_pending (%d) < min (%d)\n", + __func__, param->rx_pending, + 1 << MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE); + return -EINVAL; + } + if (param->rx_pending > (1 << MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE)) { + netdev_info(dev, "%s: rx_pending (%d) > max (%d)\n", + __func__, param->rx_pending, + 1 << MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE); + return -EINVAL; + } + if (param->tx_pending < (1 << MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE)) { + netdev_info(dev, "%s: tx_pending (%d) < min (%d)\n", + __func__, param->tx_pending, + 1 << MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE); + return -EINVAL; + } + if (param->tx_pending > (1 << MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE)) { + netdev_info(dev, "%s: tx_pending (%d) > max (%d)\n", + __func__, param->tx_pending, + 1 << MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE); + return -EINVAL; + } + + log_rq_size = order_base_2(param->rx_pending); + log_sq_size = order_base_2(param->tx_pending); + min_rx_wqes = min_t(u16, param->rx_pending - 1, + MLX5E_PARAMS_DEFAULT_MIN_RX_WQES); + + if (log_rq_size == priv->params.log_rq_size && + log_sq_size == priv->params.log_sq_size && + min_rx_wqes == priv->params.min_rx_wqes) + return 0; + + mutex_lock(&priv->state_lock); + new_params = priv->params; + new_params.log_rq_size = log_rq_size; + new_params.log_sq_size = log_sq_size; + new_params.min_rx_wqes = min_rx_wqes; + err = mlx5e_update_priv_params(priv, &new_params); + mutex_unlock(&priv->state_lock); + + return err; +} + +static void mlx5e_get_channels(struct net_device *dev, + struct ethtool_channels *ch) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int ncv = priv->mdev->priv.eq_table.num_comp_vectors; + + ch->max_combined = ncv; + ch->combined_count = priv->params.num_channels; +} + +static int mlx5e_set_channels(struct net_device *dev, + struct ethtool_channels *ch) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int ncv = priv->mdev->priv.eq_table.num_comp_vectors; + unsigned int count = ch->combined_count; + struct mlx5e_params new_params; + int err = 0; + + if (!count) { + netdev_info(dev, "%s: combined_count=0 not supported\n", + __func__); + return -EINVAL; + } + if (ch->rx_count || ch->tx_count) { + netdev_info(dev, "%s: separate rx/tx count not supported\n", + __func__); + return -EINVAL; + } + if (count > ncv) { + netdev_info(dev, "%s: count (%d) > max (%d)\n", + __func__, count, ncv); + return -EINVAL; + } + + if (priv->params.num_channels == count) + return 0; + + mutex_lock(&priv->state_lock); + new_params = priv->params; + new_params.num_channels = count; + err = mlx5e_update_priv_params(priv, &new_params); + mutex_unlock(&priv->state_lock); + + return err; +} + +static int mlx5e_get_coalesce(struct net_device *netdev, + struct ethtool_coalesce *coal) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + coal->rx_coalesce_usecs = priv->params.rx_cq_moderation_usec; + coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation_pkts; + coal->tx_coalesce_usecs = priv->params.tx_cq_moderation_usec; + coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation_pkts; + + return 0; +} + +static int mlx5e_set_coalesce(struct net_device *netdev, + struct ethtool_coalesce *coal) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_channel *c; + int tc; + int i; + + priv->params.tx_cq_moderation_usec = coal->tx_coalesce_usecs; + priv->params.tx_cq_moderation_pkts = coal->tx_max_coalesced_frames; + priv->params.rx_cq_moderation_usec = coal->rx_coalesce_usecs; + priv->params.rx_cq_moderation_pkts = coal->rx_max_coalesced_frames; + + for (i = 0; i < priv->params.num_channels; ++i) { + c = priv->channel[i]; + + for (tc = 0; tc < c->num_tc; tc++) { + mlx5_core_modify_cq_moderation(mdev, + &c->sq[tc].cq.mcq, + coal->tx_coalesce_usecs, + coal->tx_max_coalesced_frames); + } + + mlx5_core_modify_cq_moderation(mdev, &c->rq.cq.mcq, + coal->rx_coalesce_usecs, + coal->rx_max_coalesced_frames); + } + + return 0; +} + +static u32 ptys2ethtool_supported_link(u32 eth_proto_cap) +{ + int i; + u32 supported_modes = 0; + + for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { + if (eth_proto_cap & MLX5E_PROT_MASK(i)) + supported_modes |= ptys2ethtool_table[i].supported; + } + return supported_modes; +} + +static u32 ptys2ethtool_adver_link(u32 eth_proto_cap) +{ + int i; + u32 advertising_modes = 0; + + for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { + if (eth_proto_cap & MLX5E_PROT_MASK(i)) + advertising_modes |= ptys2ethtool_table[i].advertised; + } + return advertising_modes; +} + +static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) +{ + if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_10GBASE_CR) + | MLX5E_PROT_MASK(MLX5E_10GBASE_SR) + | MLX5E_PROT_MASK(MLX5E_40GBASE_CR4) + | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4) + | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4) + | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) { + return SUPPORTED_FIBRE; + } + + if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_100GBASE_KR4) + | MLX5E_PROT_MASK(MLX5E_40GBASE_KR4) + | MLX5E_PROT_MASK(MLX5E_10GBASE_KR) + | MLX5E_PROT_MASK(MLX5E_10GBASE_KX4) + | MLX5E_PROT_MASK(MLX5E_1000BASE_KX))) { + return SUPPORTED_Backplane; + } + return 0; +} + +static void get_speed_duplex(struct net_device *netdev, + u32 eth_proto_oper, + struct ethtool_cmd *cmd) +{ + int i; + u32 speed = SPEED_UNKNOWN; + u8 duplex = DUPLEX_UNKNOWN; + + if (!netif_carrier_ok(netdev)) + goto out; + + for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { + if (eth_proto_oper & MLX5E_PROT_MASK(i)) { + speed = ptys2ethtool_table[i].speed; + duplex = DUPLEX_FULL; + break; + } + } +out: + ethtool_cmd_speed_set(cmd, speed); + cmd->duplex = duplex; +} + +static void get_supported(u32 eth_proto_cap, u32 *supported) +{ + *supported |= ptys2ethtool_supported_port(eth_proto_cap); + *supported |= ptys2ethtool_supported_link(eth_proto_cap); + *supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; +} + +static void get_advertising(u32 eth_proto_cap, u8 tx_pause, + u8 rx_pause, u32 *advertising) +{ + *advertising |= ptys2ethtool_adver_link(eth_proto_cap); + *advertising |= tx_pause ? ADVERTISED_Pause : 0; + *advertising |= (tx_pause ^ rx_pause) ? ADVERTISED_Asym_Pause : 0; +} + +static u8 get_connector_port(u32 eth_proto) +{ + if (eth_proto & (MLX5E_PROT_MASK(MLX5E_10GBASE_SR) + | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4) + | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4) + | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) { + return PORT_FIBRE; + } + + if (eth_proto & (MLX5E_PROT_MASK(MLX5E_40GBASE_CR4) + | MLX5E_PROT_MASK(MLX5E_10GBASE_CR) + | MLX5E_PROT_MASK(MLX5E_100GBASE_CR4))) { + return PORT_DA; + } + + if (eth_proto & (MLX5E_PROT_MASK(MLX5E_10GBASE_KX4) + | MLX5E_PROT_MASK(MLX5E_10GBASE_KR) + | MLX5E_PROT_MASK(MLX5E_40GBASE_KR4) + | MLX5E_PROT_MASK(MLX5E_100GBASE_KR4))) { + return PORT_NONE; + } + + return PORT_OTHER; +} + +static void get_lp_advertising(u32 eth_proto_lp, u32 *lp_advertising) +{ + *lp_advertising = ptys2ethtool_adver_link(eth_proto_lp); +} + +static int mlx5e_get_settings(struct net_device *netdev, + struct ethtool_cmd *cmd) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + u32 eth_proto_cap; + u32 eth_proto_admin; + u32 eth_proto_lp; + u32 eth_proto_oper; + int err; + + err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN); + + if (err) { + netdev_err(netdev, "%s: query port ptys failed: %d\n", + __func__, err); + goto err_query_ptys; + } + + eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); + eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); + eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); + eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + + cmd->supported = 0; + cmd->advertising = 0; + + get_supported(eth_proto_cap, &cmd->supported); + get_advertising(eth_proto_admin, 0, 0, &cmd->advertising); + get_speed_duplex(netdev, eth_proto_oper, cmd); + + eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; + + cmd->port = get_connector_port(eth_proto_oper); + get_lp_advertising(eth_proto_lp, &cmd->lp_advertising); + + cmd->transceiver = XCVR_INTERNAL; + +err_query_ptys: + return err; +} + +static u32 mlx5e_ethtool2ptys_adver_link(u32 link_modes) +{ + u32 i, ptys_modes = 0; + + for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { + if (ptys2ethtool_table[i].advertised & link_modes) + ptys_modes |= MLX5E_PROT_MASK(i); + } + + return ptys_modes; +} + +static u32 mlx5e_ethtool2ptys_speed_link(u32 speed) +{ + u32 i, speed_links = 0; + + for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { + if (ptys2ethtool_table[i].speed == speed) + speed_links |= MLX5E_PROT_MASK(i); + } + + return speed_links; +} + +static int mlx5e_set_settings(struct net_device *netdev, + struct ethtool_cmd *cmd) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + u32 link_modes; + u32 speed; + u32 eth_proto_cap, eth_proto_admin; + u8 port_status; + int err; + + speed = ethtool_cmd_speed(cmd); + + link_modes = cmd->autoneg == AUTONEG_ENABLE ? + mlx5e_ethtool2ptys_adver_link(cmd->advertising) : + mlx5e_ethtool2ptys_speed_link(speed); + + err = mlx5_query_port_proto_cap(mdev, ð_proto_cap, MLX5_PTYS_EN); + if (err) { + netdev_err(netdev, "%s: query port eth proto cap failed: %d\n", + __func__, err); + goto out; + } + + link_modes = link_modes & eth_proto_cap; + if (!link_modes) { + netdev_err(netdev, "%s: Not supported link mode(s) requested", + __func__); + err = -EINVAL; + goto out; + } + + err = mlx5_query_port_proto_admin(mdev, ð_proto_admin, MLX5_PTYS_EN); + if (err) { + netdev_err(netdev, "%s: query port eth proto admin failed: %d\n", + __func__, err); + goto out; + } + + if (link_modes == eth_proto_admin) + goto out; + + err = mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); + if (err) { + netdev_err(netdev, "%s: set port eth proto admin failed: %d\n", + __func__, err); + goto out; + } + + err = mlx5_query_port_status(mdev, &port_status); + if (err) + goto out; + + if (port_status == MLX5_PORT_DOWN) + return 0; + + err = mlx5_set_port_status(mdev, MLX5_PORT_DOWN); + if (err) + goto out; + err = mlx5_set_port_status(mdev, MLX5_PORT_UP); +out: + return err; +} + +const struct ethtool_ops mlx5e_ethtool_ops = { + .get_drvinfo = mlx5e_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = mlx5e_get_strings, + .get_sset_count = mlx5e_get_sset_count, + .get_ethtool_stats = mlx5e_get_ethtool_stats, + .get_ringparam = mlx5e_get_ringparam, + .set_ringparam = mlx5e_set_ringparam, + .get_channels = mlx5e_get_channels, + .set_channels = mlx5e_set_channels, + .get_coalesce = mlx5e_get_coalesce, + .set_coalesce = mlx5e_set_coalesce, + .get_settings = mlx5e_get_settings, + .set_settings = mlx5e_set_settings, +}; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c new file mode 100644 index 000000000000..6feebda4b3e4 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_flow_table.c @@ -0,0 +1,858 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/list.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/tcp.h> +#include <linux/mlx5/flow_table.h> +#include "en.h" + +enum { + MLX5E_FULLMATCH = 0, + MLX5E_ALLMULTI = 1, + MLX5E_PROMISC = 2, +}; + +enum { + MLX5E_UC = 0, + MLX5E_MC_IPV4 = 1, + MLX5E_MC_IPV6 = 2, + MLX5E_MC_OTHER = 3, +}; + +enum { + MLX5E_ACTION_NONE = 0, + MLX5E_ACTION_ADD = 1, + MLX5E_ACTION_DEL = 2, +}; + +struct mlx5e_eth_addr_hash_node { + struct hlist_node hlist; + u8 action; + struct mlx5e_eth_addr_info ai; +}; + +static inline int mlx5e_hash_eth_addr(u8 *addr) +{ + return addr[5]; +} + +static void mlx5e_add_eth_addr_to_hash(struct hlist_head *hash, u8 *addr) +{ + struct mlx5e_eth_addr_hash_node *hn; + int ix = mlx5e_hash_eth_addr(addr); + int found = 0; + + hlist_for_each_entry(hn, &hash[ix], hlist) + if (ether_addr_equal_64bits(hn->ai.addr, addr)) { + found = 1; + break; + } + + if (found) { + hn->action = MLX5E_ACTION_NONE; + return; + } + + hn = kzalloc(sizeof(*hn), GFP_ATOMIC); + if (!hn) + return; + + ether_addr_copy(hn->ai.addr, addr); + hn->action = MLX5E_ACTION_ADD; + + hlist_add_head(&hn->hlist, &hash[ix]); +} + +static void mlx5e_del_eth_addr_from_hash(struct mlx5e_eth_addr_hash_node *hn) +{ + hlist_del(&hn->hlist); + kfree(hn); +} + +static void mlx5e_del_eth_addr_from_flow_table(struct mlx5e_priv *priv, + struct mlx5e_eth_addr_info *ai) +{ + void *ft = priv->ft.main; + + if (ai->tt_vec & (1 << MLX5E_TT_IPV6_TCP)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_TCP]); + + if (ai->tt_vec & (1 << MLX5E_TT_IPV4_TCP)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_TCP]); + + if (ai->tt_vec & (1 << MLX5E_TT_IPV6_UDP)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6_UDP]); + + if (ai->tt_vec & (1 << MLX5E_TT_IPV4_UDP)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4_UDP]); + + if (ai->tt_vec & (1 << MLX5E_TT_IPV6)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV6]); + + if (ai->tt_vec & (1 << MLX5E_TT_IPV4)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_IPV4]); + + if (ai->tt_vec & (1 << MLX5E_TT_ANY)) + mlx5_del_flow_table_entry(ft, ai->ft_ix[MLX5E_TT_ANY]); +} + +static int mlx5e_get_eth_addr_type(u8 *addr) +{ + if (is_unicast_ether_addr(addr)) + return MLX5E_UC; + + if ((addr[0] == 0x01) && + (addr[1] == 0x00) && + (addr[2] == 0x5e) && + !(addr[3] & 0x80)) + return MLX5E_MC_IPV4; + + if ((addr[0] == 0x33) && + (addr[1] == 0x33)) + return MLX5E_MC_IPV6; + + return MLX5E_MC_OTHER; +} + +static u32 mlx5e_get_tt_vec(struct mlx5e_eth_addr_info *ai, int type) +{ + int eth_addr_type; + u32 ret; + + switch (type) { + case MLX5E_FULLMATCH: + eth_addr_type = mlx5e_get_eth_addr_type(ai->addr); + switch (eth_addr_type) { + case MLX5E_UC: + ret = + (1 << MLX5E_TT_IPV4_TCP) | + (1 << MLX5E_TT_IPV6_TCP) | + (1 << MLX5E_TT_IPV4_UDP) | + (1 << MLX5E_TT_IPV6_UDP) | + (1 << MLX5E_TT_IPV4) | + (1 << MLX5E_TT_IPV6) | + (1 << MLX5E_TT_ANY) | + 0; + break; + + case MLX5E_MC_IPV4: + ret = + (1 << MLX5E_TT_IPV4_UDP) | + (1 << MLX5E_TT_IPV4) | + 0; + break; + + case MLX5E_MC_IPV6: + ret = + (1 << MLX5E_TT_IPV6_UDP) | + (1 << MLX5E_TT_IPV6) | + 0; + break; + + case MLX5E_MC_OTHER: + ret = + (1 << MLX5E_TT_ANY) | + 0; + break; + } + + break; + + case MLX5E_ALLMULTI: + ret = + (1 << MLX5E_TT_IPV4_UDP) | + (1 << MLX5E_TT_IPV6_UDP) | + (1 << MLX5E_TT_IPV4) | + (1 << MLX5E_TT_IPV6) | + (1 << MLX5E_TT_ANY) | + 0; + break; + + default: /* MLX5E_PROMISC */ + ret = + (1 << MLX5E_TT_IPV4_TCP) | + (1 << MLX5E_TT_IPV6_TCP) | + (1 << MLX5E_TT_IPV4_UDP) | + (1 << MLX5E_TT_IPV6_UDP) | + (1 << MLX5E_TT_IPV4) | + (1 << MLX5E_TT_IPV6) | + (1 << MLX5E_TT_ANY) | + 0; + break; + } + + return ret; +} + +static int __mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv, + struct mlx5e_eth_addr_info *ai, int type, + void *flow_context, void *match_criteria) +{ + u8 match_criteria_enable = 0; + void *match_value; + void *dest; + u8 *dmac; + u8 *match_criteria_dmac; + void *ft = priv->ft.main; + u32 *tirn = priv->tirn; + u32 tt_vec; + int err; + + match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); + dmac = MLX5_ADDR_OF(fte_match_param, match_value, + outer_headers.dmac_47_16); + match_criteria_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers.dmac_47_16); + dest = MLX5_ADDR_OF(flow_context, flow_context, destination); + + MLX5_SET(flow_context, flow_context, action, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST); + MLX5_SET(flow_context, flow_context, destination_list_size, 1); + MLX5_SET(dest_format_struct, dest, destination_type, + MLX5_FLOW_CONTEXT_DEST_TYPE_TIR); + + switch (type) { + case MLX5E_FULLMATCH: + match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + memset(match_criteria_dmac, 0xff, ETH_ALEN); + ether_addr_copy(dmac, ai->addr); + break; + + case MLX5E_ALLMULTI: + match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + match_criteria_dmac[0] = 0x01; + dmac[0] = 0x01; + break; + + case MLX5E_PROMISC: + break; + } + + tt_vec = mlx5e_get_tt_vec(ai, type); + + if (tt_vec & (1 << MLX5E_TT_ANY)) { + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_ANY]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_ANY]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_ANY); + } + + match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, match_criteria, + outer_headers.ethertype); + + if (tt_vec & (1 << MLX5E_TT_IPV4)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IP); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV4]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV4]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV4); + } + + if (tt_vec & (1 << MLX5E_TT_IPV6)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IPV6); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV6]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV6]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV6); + } + + MLX5_SET_TO_ONES(fte_match_param, match_criteria, + outer_headers.ip_protocol); + MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, + IPPROTO_UDP); + + if (tt_vec & (1 << MLX5E_TT_IPV4_UDP)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IP); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV4_UDP]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV4_UDP]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV4_UDP); + } + + if (tt_vec & (1 << MLX5E_TT_IPV6_UDP)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IPV6); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV6_UDP]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV6_UDP]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV6_UDP); + } + + MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, + IPPROTO_TCP); + + if (tt_vec & (1 << MLX5E_TT_IPV4_TCP)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IP); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV4_TCP]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV4_TCP]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV4_TCP); + } + + if (tt_vec & (1 << MLX5E_TT_IPV6_TCP)) { + MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + ETH_P_IPV6); + MLX5_SET(dest_format_struct, dest, destination_id, + tirn[MLX5E_TT_IPV6_TCP]); + err = mlx5_add_flow_table_entry(ft, match_criteria_enable, + match_criteria, flow_context, + &ai->ft_ix[MLX5E_TT_IPV6_TCP]); + if (err) { + mlx5e_del_eth_addr_from_flow_table(priv, ai); + return err; + } + ai->tt_vec |= (1 << MLX5E_TT_IPV6_TCP); + } + + return 0; +} + +static int mlx5e_add_eth_addr_rule(struct mlx5e_priv *priv, + struct mlx5e_eth_addr_info *ai, int type) +{ + u32 *flow_context; + u32 *match_criteria; + int err; + + flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + + MLX5_ST_SZ_BYTES(dest_format_struct)); + match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + if (!flow_context || !match_criteria) { + netdev_err(priv->netdev, "%s: alloc failed\n", __func__); + err = -ENOMEM; + goto add_eth_addr_rule_out; + } + + err = __mlx5e_add_eth_addr_rule(priv, ai, type, flow_context, + match_criteria); + if (err) + netdev_err(priv->netdev, "%s: failed\n", __func__); + +add_eth_addr_rule_out: + kvfree(match_criteria); + kvfree(flow_context); + return err; +} + +enum mlx5e_vlan_rule_type { + MLX5E_VLAN_RULE_TYPE_UNTAGGED, + MLX5E_VLAN_RULE_TYPE_ANY_VID, + MLX5E_VLAN_RULE_TYPE_MATCH_VID, +}; + +static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, + enum mlx5e_vlan_rule_type rule_type, u16 vid) +{ + u8 match_criteria_enable = 0; + u32 *flow_context; + void *match_value; + void *dest; + u32 *match_criteria; + u32 *ft_ix; + int err; + + flow_context = mlx5_vzalloc(MLX5_ST_SZ_BYTES(flow_context) + + MLX5_ST_SZ_BYTES(dest_format_struct)); + match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); + if (!flow_context || !match_criteria) { + netdev_err(priv->netdev, "%s: alloc failed\n", __func__); + err = -ENOMEM; + goto add_vlan_rule_out; + } + match_value = MLX5_ADDR_OF(flow_context, flow_context, match_value); + dest = MLX5_ADDR_OF(flow_context, flow_context, destination); + + MLX5_SET(flow_context, flow_context, action, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST); + MLX5_SET(flow_context, flow_context, destination_list_size, 1); + MLX5_SET(dest_format_struct, dest, destination_type, + MLX5_FLOW_CONTEXT_DEST_TYPE_FLOW_TABLE); + MLX5_SET(dest_format_struct, dest, destination_id, + mlx5_get_flow_table_id(priv->ft.main)); + + match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, match_criteria, + outer_headers.vlan_tag); + + switch (rule_type) { + case MLX5E_VLAN_RULE_TYPE_UNTAGGED: + ft_ix = &priv->vlan.untagged_rule_ft_ix; + break; + case MLX5E_VLAN_RULE_TYPE_ANY_VID: + ft_ix = &priv->vlan.any_vlan_rule_ft_ix; + MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag, + 1); + break; + default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */ + ft_ix = &priv->vlan.active_vlans_ft_ix[vid]; + MLX5_SET(fte_match_param, match_value, outer_headers.vlan_tag, + 1); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, + outer_headers.first_vid); + MLX5_SET(fte_match_param, match_value, outer_headers.first_vid, + vid); + break; + } + + err = mlx5_add_flow_table_entry(priv->ft.vlan, match_criteria_enable, + match_criteria, flow_context, ft_ix); + if (err) + netdev_err(priv->netdev, "%s: failed\n", __func__); + +add_vlan_rule_out: + kvfree(match_criteria); + kvfree(flow_context); + return err; +} + +static void mlx5e_del_vlan_rule(struct mlx5e_priv *priv, + enum mlx5e_vlan_rule_type rule_type, u16 vid) +{ + switch (rule_type) { + case MLX5E_VLAN_RULE_TYPE_UNTAGGED: + mlx5_del_flow_table_entry(priv->ft.vlan, + priv->vlan.untagged_rule_ft_ix); + break; + case MLX5E_VLAN_RULE_TYPE_ANY_VID: + mlx5_del_flow_table_entry(priv->ft.vlan, + priv->vlan.any_vlan_rule_ft_ix); + break; + case MLX5E_VLAN_RULE_TYPE_MATCH_VID: + mlx5_del_flow_table_entry(priv->ft.vlan, + priv->vlan.active_vlans_ft_ix[vid]); + break; + } +} + +void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv) +{ + WARN_ON(!mutex_is_locked(&priv->state_lock)); + + if (priv->vlan.filter_disabled) { + priv->vlan.filter_disabled = false; + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, + 0); + } +} + +void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv) +{ + WARN_ON(!mutex_is_locked(&priv->state_lock)); + + if (!priv->vlan.filter_disabled) { + priv->vlan.filter_disabled = true; + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, + 0); + } +} + +int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto, + u16 vid) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int err = 0; + + mutex_lock(&priv->state_lock); + + set_bit(vid, priv->vlan.active_vlans); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, + vid); + + mutex_unlock(&priv->state_lock); + + return err; +} + +int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __always_unused __be16 proto, + u16 vid) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + mutex_lock(&priv->state_lock); + + clear_bit(vid, priv->vlan.active_vlans); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid); + + mutex_unlock(&priv->state_lock); + + return 0; +} + +int mlx5e_add_all_vlan_rules(struct mlx5e_priv *priv) +{ + u16 vid; + int err; + + for_each_set_bit(vid, priv->vlan.active_vlans, VLAN_N_VID) { + err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, + vid); + if (err) + return err; + } + + err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0); + if (err) + return err; + + if (priv->vlan.filter_disabled) { + err = mlx5e_add_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, + 0); + if (err) + return err; + } + + return 0; +} + +void mlx5e_del_all_vlan_rules(struct mlx5e_priv *priv) +{ + u16 vid; + + if (priv->vlan.filter_disabled) + mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_ANY_VID, 0); + + mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_UNTAGGED, 0); + + for_each_set_bit(vid, priv->vlan.active_vlans, VLAN_N_VID) + mlx5e_del_vlan_rule(priv, MLX5E_VLAN_RULE_TYPE_MATCH_VID, vid); +} + +#define mlx5e_for_each_hash_node(hn, tmp, hash, i) \ + for (i = 0; i < MLX5E_ETH_ADDR_HASH_SIZE; i++) \ + hlist_for_each_entry_safe(hn, tmp, &hash[i], hlist) + +static void mlx5e_execute_action(struct mlx5e_priv *priv, + struct mlx5e_eth_addr_hash_node *hn) +{ + switch (hn->action) { + case MLX5E_ACTION_ADD: + mlx5e_add_eth_addr_rule(priv, &hn->ai, MLX5E_FULLMATCH); + hn->action = MLX5E_ACTION_NONE; + break; + + case MLX5E_ACTION_DEL: + mlx5e_del_eth_addr_from_flow_table(priv, &hn->ai); + mlx5e_del_eth_addr_from_hash(hn); + break; + } +} + +static void mlx5e_sync_netdev_addr(struct mlx5e_priv *priv) +{ + struct net_device *netdev = priv->netdev; + struct netdev_hw_addr *ha; + + netif_addr_lock_bh(netdev); + + mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc, + priv->netdev->dev_addr); + + netdev_for_each_uc_addr(ha, netdev) + mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_uc, ha->addr); + + netdev_for_each_mc_addr(ha, netdev) + mlx5e_add_eth_addr_to_hash(priv->eth_addr.netdev_mc, ha->addr); + + netif_addr_unlock_bh(netdev); +} + +static void mlx5e_apply_netdev_addr(struct mlx5e_priv *priv) +{ + struct mlx5e_eth_addr_hash_node *hn; + struct hlist_node *tmp; + int i; + + mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i) + mlx5e_execute_action(priv, hn); + + mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i) + mlx5e_execute_action(priv, hn); +} + +static void mlx5e_handle_netdev_addr(struct mlx5e_priv *priv) +{ + struct mlx5e_eth_addr_hash_node *hn; + struct hlist_node *tmp; + int i; + + mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_uc, i) + hn->action = MLX5E_ACTION_DEL; + mlx5e_for_each_hash_node(hn, tmp, priv->eth_addr.netdev_mc, i) + hn->action = MLX5E_ACTION_DEL; + + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_sync_netdev_addr(priv); + + mlx5e_apply_netdev_addr(priv); +} + +void mlx5e_set_rx_mode_core(struct mlx5e_priv *priv) +{ + struct mlx5e_eth_addr_db *ea = &priv->eth_addr; + struct net_device *ndev = priv->netdev; + + bool rx_mode_enable = test_bit(MLX5E_STATE_OPENED, &priv->state); + bool promisc_enabled = rx_mode_enable && (ndev->flags & IFF_PROMISC); + bool allmulti_enabled = rx_mode_enable && (ndev->flags & IFF_ALLMULTI); + bool broadcast_enabled = rx_mode_enable; + + bool enable_promisc = !ea->promisc_enabled && promisc_enabled; + bool disable_promisc = ea->promisc_enabled && !promisc_enabled; + bool enable_allmulti = !ea->allmulti_enabled && allmulti_enabled; + bool disable_allmulti = ea->allmulti_enabled && !allmulti_enabled; + bool enable_broadcast = !ea->broadcast_enabled && broadcast_enabled; + bool disable_broadcast = ea->broadcast_enabled && !broadcast_enabled; + + if (enable_promisc) + mlx5e_add_eth_addr_rule(priv, &ea->promisc, MLX5E_PROMISC); + if (enable_allmulti) + mlx5e_add_eth_addr_rule(priv, &ea->allmulti, MLX5E_ALLMULTI); + if (enable_broadcast) + mlx5e_add_eth_addr_rule(priv, &ea->broadcast, MLX5E_FULLMATCH); + + mlx5e_handle_netdev_addr(priv); + + if (disable_broadcast) + mlx5e_del_eth_addr_from_flow_table(priv, &ea->broadcast); + if (disable_allmulti) + mlx5e_del_eth_addr_from_flow_table(priv, &ea->allmulti); + if (disable_promisc) + mlx5e_del_eth_addr_from_flow_table(priv, &ea->promisc); + + ea->promisc_enabled = promisc_enabled; + ea->allmulti_enabled = allmulti_enabled; + ea->broadcast_enabled = broadcast_enabled; +} + +void mlx5e_set_rx_mode_work(struct work_struct *work) +{ + struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, + set_rx_mode_work); + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_set_rx_mode_core(priv); + mutex_unlock(&priv->state_lock); +} + +void mlx5e_init_eth_addr(struct mlx5e_priv *priv) +{ + ether_addr_copy(priv->eth_addr.broadcast.addr, priv->netdev->broadcast); +} + +static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv) +{ + struct mlx5_flow_table_group *g; + u8 *dmac; + + g = kcalloc(9, sizeof(*g), GFP_KERNEL); + + g[0].log_sz = 2; + g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, + outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, + outer_headers.ip_protocol); + + g[1].log_sz = 1; + g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria, + outer_headers.ethertype); + + g[2].log_sz = 0; + + g[3].log_sz = 14; + g[3].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[3].match_criteria, + outer_headers.dmac_47_16); + memset(dmac, 0xff, ETH_ALEN); + MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria, + outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, g[3].match_criteria, + outer_headers.ip_protocol); + + g[4].log_sz = 13; + g[4].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[4].match_criteria, + outer_headers.dmac_47_16); + memset(dmac, 0xff, ETH_ALEN); + MLX5_SET_TO_ONES(fte_match_param, g[4].match_criteria, + outer_headers.ethertype); + + g[5].log_sz = 11; + g[5].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[5].match_criteria, + outer_headers.dmac_47_16); + memset(dmac, 0xff, ETH_ALEN); + + g[6].log_sz = 2; + g[6].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[6].match_criteria, + outer_headers.dmac_47_16); + dmac[0] = 0x01; + MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria, + outer_headers.ethertype); + MLX5_SET_TO_ONES(fte_match_param, g[6].match_criteria, + outer_headers.ip_protocol); + + g[7].log_sz = 1; + g[7].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[7].match_criteria, + outer_headers.dmac_47_16); + dmac[0] = 0x01; + MLX5_SET_TO_ONES(fte_match_param, g[7].match_criteria, + outer_headers.ethertype); + + g[8].log_sz = 0; + g[8].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + dmac = MLX5_ADDR_OF(fte_match_param, g[8].match_criteria, + outer_headers.dmac_47_16); + dmac[0] = 0x01; + priv->ft.main = mlx5_create_flow_table(priv->mdev, 1, + MLX5_FLOW_TABLE_TYPE_NIC_RCV, + 9, g); + kfree(g); + + return priv->ft.main ? 0 : -ENOMEM; +} + +static void mlx5e_destroy_main_flow_table(struct mlx5e_priv *priv) +{ + mlx5_destroy_flow_table(priv->ft.main); +} + +static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv) +{ + struct mlx5_flow_table_group *g; + + g = kcalloc(2, sizeof(*g), GFP_KERNEL); + if (!g) + return -ENOMEM; + + g[0].log_sz = 12; + g[0].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, + outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, g[0].match_criteria, + outer_headers.first_vid); + + /* untagged + any vlan id */ + g[1].log_sz = 1; + g[1].match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, g[1].match_criteria, + outer_headers.vlan_tag); + + priv->ft.vlan = mlx5_create_flow_table(priv->mdev, 0, + MLX5_FLOW_TABLE_TYPE_NIC_RCV, + 2, g); + + kfree(g); + return priv->ft.vlan ? 0 : -ENOMEM; +} + +static void mlx5e_destroy_vlan_flow_table(struct mlx5e_priv *priv) +{ + mlx5_destroy_flow_table(priv->ft.vlan); +} + +int mlx5e_open_flow_table(struct mlx5e_priv *priv) +{ + int err; + + err = mlx5e_create_main_flow_table(priv); + if (err) + return err; + + err = mlx5e_create_vlan_flow_table(priv); + if (err) + goto err_destroy_main_flow_table; + + return 0; + +err_destroy_main_flow_table: + mlx5e_destroy_main_flow_table(priv); + + return err; +} + +void mlx5e_close_flow_table(struct mlx5e_priv *priv) +{ + mlx5e_destroy_vlan_flow_table(priv); + mlx5e_destroy_main_flow_table(priv); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c new file mode 100644 index 000000000000..eee829d119f9 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -0,0 +1,1899 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/mlx5/flow_table.h> +#include "en.h" + +struct mlx5e_rq_param { + u32 rqc[MLX5_ST_SZ_DW(rqc)]; + struct mlx5_wq_param wq; +}; + +struct mlx5e_sq_param { + u32 sqc[MLX5_ST_SZ_DW(sqc)]; + struct mlx5_wq_param wq; +}; + +struct mlx5e_cq_param { + u32 cqc[MLX5_ST_SZ_DW(cqc)]; + struct mlx5_wq_param wq; + u16 eq_ix; +}; + +struct mlx5e_channel_param { + struct mlx5e_rq_param rq; + struct mlx5e_sq_param sq; + struct mlx5e_cq_param rx_cq; + struct mlx5e_cq_param tx_cq; +}; + +static void mlx5e_update_carrier(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u8 port_state; + + port_state = mlx5_query_vport_state(mdev, + MLX5_QUERY_VPORT_STATE_IN_OP_MOD_VNIC_VPORT); + + if (port_state == VPORT_STATE_UP) + netif_carrier_on(priv->netdev); + else + netif_carrier_off(priv->netdev); +} + +static void mlx5e_update_carrier_work(struct work_struct *work) +{ + struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, + update_carrier_work); + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_update_carrier(priv); + mutex_unlock(&priv->state_lock); +} + +void mlx5e_update_stats(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_vport_stats *s = &priv->stats.vport; + struct mlx5e_rq_stats *rq_stats; + struct mlx5e_sq_stats *sq_stats; + u32 in[MLX5_ST_SZ_DW(query_vport_counter_in)]; + u32 *out; + int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out); + u64 tx_offload_none; + int i, j; + + out = mlx5_vzalloc(outlen); + if (!out) + return; + + /* Collect firts the SW counters and then HW for consistency */ + s->tso_packets = 0; + s->tso_bytes = 0; + s->tx_queue_stopped = 0; + s->tx_queue_wake = 0; + s->tx_queue_dropped = 0; + tx_offload_none = 0; + s->lro_packets = 0; + s->lro_bytes = 0; + s->rx_csum_none = 0; + s->rx_wqe_err = 0; + for (i = 0; i < priv->params.num_channels; i++) { + rq_stats = &priv->channel[i]->rq.stats; + + s->lro_packets += rq_stats->lro_packets; + s->lro_bytes += rq_stats->lro_bytes; + s->rx_csum_none += rq_stats->csum_none; + s->rx_wqe_err += rq_stats->wqe_err; + + for (j = 0; j < priv->num_tc; j++) { + sq_stats = &priv->channel[i]->sq[j].stats; + + s->tso_packets += sq_stats->tso_packets; + s->tso_bytes += sq_stats->tso_bytes; + s->tx_queue_stopped += sq_stats->stopped; + s->tx_queue_wake += sq_stats->wake; + s->tx_queue_dropped += sq_stats->dropped; + tx_offload_none += sq_stats->csum_offload_none; + } + } + + /* HW counters */ + memset(in, 0, sizeof(in)); + + MLX5_SET(query_vport_counter_in, in, opcode, + MLX5_CMD_OP_QUERY_VPORT_COUNTER); + MLX5_SET(query_vport_counter_in, in, op_mod, 0); + MLX5_SET(query_vport_counter_in, in, other_vport, 0); + + memset(out, 0, outlen); + + if (mlx5_cmd_exec(mdev, in, sizeof(in), out, outlen)) + goto free_out; + +#define MLX5_GET_CTR(p, x) \ + MLX5_GET64(query_vport_counter_out, p, x) + + s->rx_error_packets = + MLX5_GET_CTR(out, received_errors.packets); + s->rx_error_bytes = + MLX5_GET_CTR(out, received_errors.octets); + s->tx_error_packets = + MLX5_GET_CTR(out, transmit_errors.packets); + s->tx_error_bytes = + MLX5_GET_CTR(out, transmit_errors.octets); + + s->rx_unicast_packets = + MLX5_GET_CTR(out, received_eth_unicast.packets); + s->rx_unicast_bytes = + MLX5_GET_CTR(out, received_eth_unicast.octets); + s->tx_unicast_packets = + MLX5_GET_CTR(out, transmitted_eth_unicast.packets); + s->tx_unicast_bytes = + MLX5_GET_CTR(out, transmitted_eth_unicast.octets); + + s->rx_multicast_packets = + MLX5_GET_CTR(out, received_eth_multicast.packets); + s->rx_multicast_bytes = + MLX5_GET_CTR(out, received_eth_multicast.octets); + s->tx_multicast_packets = + MLX5_GET_CTR(out, transmitted_eth_multicast.packets); + s->tx_multicast_bytes = + MLX5_GET_CTR(out, transmitted_eth_multicast.octets); + + s->rx_broadcast_packets = + MLX5_GET_CTR(out, received_eth_broadcast.packets); + s->rx_broadcast_bytes = + MLX5_GET_CTR(out, received_eth_broadcast.octets); + s->tx_broadcast_packets = + MLX5_GET_CTR(out, transmitted_eth_broadcast.packets); + s->tx_broadcast_bytes = + MLX5_GET_CTR(out, transmitted_eth_broadcast.octets); + + s->rx_packets = + s->rx_unicast_packets + + s->rx_multicast_packets + + s->rx_broadcast_packets; + s->rx_bytes = + s->rx_unicast_bytes + + s->rx_multicast_bytes + + s->rx_broadcast_bytes; + s->tx_packets = + s->tx_unicast_packets + + s->tx_multicast_packets + + s->tx_broadcast_packets; + s->tx_bytes = + s->tx_unicast_bytes + + s->tx_multicast_bytes + + s->tx_broadcast_bytes; + + /* Update calculated offload counters */ + s->tx_csum_offload = s->tx_packets - tx_offload_none; + s->rx_csum_good = s->rx_packets - s->rx_csum_none; + +free_out: + kvfree(out); +} + +static void mlx5e_update_stats_work(struct work_struct *work) +{ + struct delayed_work *dwork = to_delayed_work(work); + struct mlx5e_priv *priv = container_of(dwork, struct mlx5e_priv, + update_stats_work); + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) { + mlx5e_update_stats(priv); + schedule_delayed_work(dwork, + msecs_to_jiffies( + MLX5E_UPDATE_STATS_INTERVAL)); + } + mutex_unlock(&priv->state_lock); +} + +static void __mlx5e_async_event(struct mlx5e_priv *priv, + enum mlx5_dev_event event) +{ + switch (event) { + case MLX5_DEV_EVENT_PORT_UP: + case MLX5_DEV_EVENT_PORT_DOWN: + schedule_work(&priv->update_carrier_work); + break; + + default: + break; + } +} + +static void mlx5e_async_event(struct mlx5_core_dev *mdev, void *vpriv, + enum mlx5_dev_event event, unsigned long param) +{ + struct mlx5e_priv *priv = vpriv; + + spin_lock(&priv->async_events_spinlock); + if (test_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state)) + __mlx5e_async_event(priv, event); + spin_unlock(&priv->async_events_spinlock); +} + +static void mlx5e_enable_async_events(struct mlx5e_priv *priv) +{ + set_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state); +} + +static void mlx5e_disable_async_events(struct mlx5e_priv *priv) +{ + spin_lock_irq(&priv->async_events_spinlock); + clear_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state); + spin_unlock_irq(&priv->async_events_spinlock); +} + +static void mlx5e_send_nop(struct mlx5e_sq *sq) +{ + struct mlx5_wq_cyc *wq = &sq->wq; + + u16 pi = sq->pc & wq->sz_m1; + struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(wq, pi); + + struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; + + memset(cseg, 0, sizeof(*cseg)); + + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_NOP); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | 0x01); + cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; + + sq->skb[pi] = NULL; + sq->pc++; + mlx5e_tx_notify_hw(sq, wqe); +} + +static int mlx5e_create_rq(struct mlx5e_channel *c, + struct mlx5e_rq_param *param, + struct mlx5e_rq *rq) +{ + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + void *rqc = param->rqc; + void *rqc_wq = MLX5_ADDR_OF(rqc, rqc, wq); + int wq_sz; + int err; + int i; + + err = mlx5_wq_ll_create(mdev, ¶m->wq, rqc_wq, &rq->wq, + &rq->wq_ctrl); + if (err) + return err; + + rq->wq.db = &rq->wq.db[MLX5_RCV_DBR]; + + wq_sz = mlx5_wq_ll_get_size(&rq->wq); + rq->skb = kzalloc_node(wq_sz * sizeof(*rq->skb), GFP_KERNEL, + cpu_to_node(c->cpu)); + if (!rq->skb) { + err = -ENOMEM; + goto err_rq_wq_destroy; + } + + rq->wqe_sz = (priv->params.lro_en) ? priv->params.lro_wqe_sz : + priv->netdev->mtu + ETH_HLEN + VLAN_HLEN; + + for (i = 0; i < wq_sz; i++) { + struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(&rq->wq, i); + + wqe->data.lkey = c->mkey_be; + wqe->data.byte_count = cpu_to_be32(rq->wqe_sz); + } + + rq->pdev = c->pdev; + rq->netdev = c->netdev; + rq->channel = c; + rq->ix = c->ix; + + return 0; + +err_rq_wq_destroy: + mlx5_wq_destroy(&rq->wq_ctrl); + + return err; +} + +static void mlx5e_destroy_rq(struct mlx5e_rq *rq) +{ + kfree(rq->skb); + mlx5_wq_destroy(&rq->wq_ctrl); +} + +static int mlx5e_enable_rq(struct mlx5e_rq *rq, struct mlx5e_rq_param *param) +{ + struct mlx5e_channel *c = rq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + void *in; + void *rqc; + void *wq; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(create_rq_in) + + sizeof(u64) * rq->wq_ctrl.buf.npages; + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + rqc = MLX5_ADDR_OF(create_rq_in, in, ctx); + wq = MLX5_ADDR_OF(rqc, rqc, wq); + + memcpy(rqc, param->rqc, sizeof(param->rqc)); + + MLX5_SET(rqc, rqc, cqn, c->rq.cq.mcq.cqn); + MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RST); + MLX5_SET(rqc, rqc, flush_in_error_en, 1); + MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST); + MLX5_SET(wq, wq, log_wq_pg_sz, rq->wq_ctrl.buf.page_shift - + PAGE_SHIFT); + MLX5_SET64(wq, wq, dbr_addr, rq->wq_ctrl.db.dma); + + mlx5_fill_page_array(&rq->wq_ctrl.buf, + (__be64 *)MLX5_ADDR_OF(wq, wq, pas)); + + err = mlx5_create_rq(mdev, in, inlen, &rq->rqn); + + kvfree(in); + + return err; +} + +static int mlx5e_modify_rq(struct mlx5e_rq *rq, int curr_state, int next_state) +{ + struct mlx5e_channel *c = rq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + void *in; + void *rqc; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(modify_rq_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + rqc = MLX5_ADDR_OF(modify_rq_in, in, ctx); + + MLX5_SET(modify_rq_in, in, rq_state, curr_state); + MLX5_SET(rqc, rqc, state, next_state); + + err = mlx5_modify_rq(mdev, rq->rqn, in, inlen); + + kvfree(in); + + return err; +} + +static void mlx5e_disable_rq(struct mlx5e_rq *rq) +{ + struct mlx5e_channel *c = rq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + mlx5_destroy_rq(mdev, rq->rqn); +} + +static int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq) +{ + struct mlx5e_channel *c = rq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_wq_ll *wq = &rq->wq; + int i; + + for (i = 0; i < 1000; i++) { + if (wq->cur_sz >= priv->params.min_rx_wqes) + return 0; + + msleep(20); + } + + return -ETIMEDOUT; +} + +static int mlx5e_open_rq(struct mlx5e_channel *c, + struct mlx5e_rq_param *param, + struct mlx5e_rq *rq) +{ + int err; + + err = mlx5e_create_rq(c, param, rq); + if (err) + return err; + + err = mlx5e_enable_rq(rq, param); + if (err) + goto err_destroy_rq; + + err = mlx5e_modify_rq(rq, MLX5_RQC_STATE_RST, MLX5_RQC_STATE_RDY); + if (err) + goto err_disable_rq; + + set_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state); + mlx5e_send_nop(&c->sq[0]); /* trigger mlx5e_post_rx_wqes() */ + + return 0; + +err_disable_rq: + mlx5e_disable_rq(rq); +err_destroy_rq: + mlx5e_destroy_rq(rq); + + return err; +} + +static void mlx5e_close_rq(struct mlx5e_rq *rq) +{ + clear_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state); + napi_synchronize(&rq->channel->napi); /* prevent mlx5e_post_rx_wqes */ + + mlx5e_modify_rq(rq, MLX5_RQC_STATE_RDY, MLX5_RQC_STATE_ERR); + while (!mlx5_wq_ll_is_empty(&rq->wq)) + msleep(20); + + /* avoid destroying rq before mlx5e_poll_rx_cq() is done with it */ + napi_synchronize(&rq->channel->napi); + + mlx5e_disable_rq(rq); + mlx5e_destroy_rq(rq); +} + +static void mlx5e_free_sq_db(struct mlx5e_sq *sq) +{ + kfree(sq->dma_fifo); + kfree(sq->skb); +} + +static int mlx5e_alloc_sq_db(struct mlx5e_sq *sq, int numa) +{ + int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); + int df_sz = wq_sz * MLX5_SEND_WQEBB_NUM_DS; + + sq->skb = kzalloc_node(wq_sz * sizeof(*sq->skb), GFP_KERNEL, numa); + sq->dma_fifo = kzalloc_node(df_sz * sizeof(*sq->dma_fifo), GFP_KERNEL, + numa); + + if (!sq->skb || !sq->dma_fifo) { + mlx5e_free_sq_db(sq); + return -ENOMEM; + } + + sq->dma_fifo_mask = df_sz - 1; + + return 0; +} + +static int mlx5e_create_sq(struct mlx5e_channel *c, + int tc, + struct mlx5e_sq_param *param, + struct mlx5e_sq *sq) +{ + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + void *sqc = param->sqc; + void *sqc_wq = MLX5_ADDR_OF(sqc, sqc, wq); + int err; + + err = mlx5_alloc_map_uar(mdev, &sq->uar); + if (err) + return err; + + err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, &sq->wq, + &sq->wq_ctrl); + if (err) + goto err_unmap_free_uar; + + sq->wq.db = &sq->wq.db[MLX5_SND_DBR]; + sq->uar_map = sq->uar.map; + sq->bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2; + + if (mlx5e_alloc_sq_db(sq, cpu_to_node(c->cpu))) + goto err_sq_wq_destroy; + + sq->txq = netdev_get_tx_queue(priv->netdev, + c->ix + tc * priv->params.num_channels); + + sq->pdev = c->pdev; + sq->mkey_be = c->mkey_be; + sq->channel = c; + sq->tc = tc; + + return 0; + +err_sq_wq_destroy: + mlx5_wq_destroy(&sq->wq_ctrl); + +err_unmap_free_uar: + mlx5_unmap_free_uar(mdev, &sq->uar); + + return err; +} + +static void mlx5e_destroy_sq(struct mlx5e_sq *sq) +{ + struct mlx5e_channel *c = sq->channel; + struct mlx5e_priv *priv = c->priv; + + mlx5e_free_sq_db(sq); + mlx5_wq_destroy(&sq->wq_ctrl); + mlx5_unmap_free_uar(priv->mdev, &sq->uar); +} + +static int mlx5e_enable_sq(struct mlx5e_sq *sq, struct mlx5e_sq_param *param) +{ + struct mlx5e_channel *c = sq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + void *in; + void *sqc; + void *wq; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(create_sq_in) + + sizeof(u64) * sq->wq_ctrl.buf.npages; + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + sqc = MLX5_ADDR_OF(create_sq_in, in, ctx); + wq = MLX5_ADDR_OF(sqc, sqc, wq); + + memcpy(sqc, param->sqc, sizeof(param->sqc)); + + MLX5_SET(sqc, sqc, user_index, sq->tc); + MLX5_SET(sqc, sqc, tis_num_0, priv->tisn[sq->tc]); + MLX5_SET(sqc, sqc, cqn, c->sq[sq->tc].cq.mcq.cqn); + MLX5_SET(sqc, sqc, state, MLX5_SQC_STATE_RST); + MLX5_SET(sqc, sqc, tis_lst_sz, 1); + MLX5_SET(sqc, sqc, flush_in_error_en, 1); + + MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_CYCLIC); + MLX5_SET(wq, wq, uar_page, sq->uar.index); + MLX5_SET(wq, wq, log_wq_pg_sz, sq->wq_ctrl.buf.page_shift - + PAGE_SHIFT); + MLX5_SET64(wq, wq, dbr_addr, sq->wq_ctrl.db.dma); + + mlx5_fill_page_array(&sq->wq_ctrl.buf, + (__be64 *)MLX5_ADDR_OF(wq, wq, pas)); + + err = mlx5_create_sq(mdev, in, inlen, &sq->sqn); + + kvfree(in); + + return err; +} + +static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, int next_state) +{ + struct mlx5e_channel *c = sq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + void *in; + void *sqc; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(modify_sq_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + sqc = MLX5_ADDR_OF(modify_sq_in, in, ctx); + + MLX5_SET(modify_sq_in, in, sq_state, curr_state); + MLX5_SET(sqc, sqc, state, next_state); + + err = mlx5_modify_sq(mdev, sq->sqn, in, inlen); + + kvfree(in); + + return err; +} + +static void mlx5e_disable_sq(struct mlx5e_sq *sq) +{ + struct mlx5e_channel *c = sq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + mlx5_destroy_sq(mdev, sq->sqn); +} + +static int mlx5e_open_sq(struct mlx5e_channel *c, + int tc, + struct mlx5e_sq_param *param, + struct mlx5e_sq *sq) +{ + int err; + + err = mlx5e_create_sq(c, tc, param, sq); + if (err) + return err; + + err = mlx5e_enable_sq(sq, param); + if (err) + goto err_destroy_sq; + + err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RST, MLX5_SQC_STATE_RDY); + if (err) + goto err_disable_sq; + + set_bit(MLX5E_SQ_STATE_WAKE_TXQ_ENABLE, &sq->state); + netdev_tx_reset_queue(sq->txq); + netif_tx_start_queue(sq->txq); + + return 0; + +err_disable_sq: + mlx5e_disable_sq(sq); +err_destroy_sq: + mlx5e_destroy_sq(sq); + + return err; +} + +static inline void netif_tx_disable_queue(struct netdev_queue *txq) +{ + __netif_tx_lock_bh(txq); + netif_tx_stop_queue(txq); + __netif_tx_unlock_bh(txq); +} + +static void mlx5e_close_sq(struct mlx5e_sq *sq) +{ + clear_bit(MLX5E_SQ_STATE_WAKE_TXQ_ENABLE, &sq->state); + napi_synchronize(&sq->channel->napi); /* prevent netif_tx_wake_queue */ + netif_tx_disable_queue(sq->txq); + + /* ensure hw is notified of all pending wqes */ + if (mlx5e_sq_has_room_for(sq, 1)) + mlx5e_send_nop(sq); + + mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, MLX5_SQC_STATE_ERR); + while (sq->cc != sq->pc) /* wait till sq is empty */ + msleep(20); + + /* avoid destroying sq before mlx5e_poll_tx_cq() is done with it */ + napi_synchronize(&sq->channel->napi); + + mlx5e_disable_sq(sq); + mlx5e_destroy_sq(sq); +} + +static int mlx5e_create_cq(struct mlx5e_channel *c, + struct mlx5e_cq_param *param, + struct mlx5e_cq *cq) +{ + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_core_cq *mcq = &cq->mcq; + int eqn_not_used; + int irqn; + int err; + u32 i; + + param->wq.numa = cpu_to_node(c->cpu); + param->eq_ix = c->ix; + + err = mlx5_cqwq_create(mdev, ¶m->wq, param->cqc, &cq->wq, + &cq->wq_ctrl); + if (err) + return err; + + mlx5_vector2eqn(mdev, param->eq_ix, &eqn_not_used, &irqn); + + cq->napi = &c->napi; + + mcq->cqe_sz = 64; + mcq->set_ci_db = cq->wq_ctrl.db.db; + mcq->arm_db = cq->wq_ctrl.db.db + 1; + *mcq->set_ci_db = 0; + *mcq->arm_db = 0; + mcq->vector = param->eq_ix; + mcq->comp = mlx5e_completion_event; + mcq->event = mlx5e_cq_error_event; + mcq->irqn = irqn; + mcq->uar = &priv->cq_uar; + + for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) { + struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); + + cqe->op_own = 0xf1; + } + + cq->channel = c; + + return 0; +} + +static void mlx5e_destroy_cq(struct mlx5e_cq *cq) +{ + mlx5_wq_destroy(&cq->wq_ctrl); +} + +static int mlx5e_enable_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) +{ + struct mlx5e_channel *c = cq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_core_cq *mcq = &cq->mcq; + + void *in; + void *cqc; + int inlen; + int irqn_not_used; + int eqn; + int err; + + inlen = MLX5_ST_SZ_BYTES(create_cq_in) + + sizeof(u64) * cq->wq_ctrl.buf.npages; + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + cqc = MLX5_ADDR_OF(create_cq_in, in, cq_context); + + memcpy(cqc, param->cqc, sizeof(param->cqc)); + + mlx5_fill_page_array(&cq->wq_ctrl.buf, + (__be64 *)MLX5_ADDR_OF(create_cq_in, in, pas)); + + mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); + + MLX5_SET(cqc, cqc, c_eqn, eqn); + MLX5_SET(cqc, cqc, uar_page, mcq->uar->index); + MLX5_SET(cqc, cqc, log_page_size, cq->wq_ctrl.buf.page_shift - + PAGE_SHIFT); + MLX5_SET64(cqc, cqc, dbr_addr, cq->wq_ctrl.db.dma); + + err = mlx5_core_create_cq(mdev, mcq, in, inlen); + + kvfree(in); + + if (err) + return err; + + mlx5e_cq_arm(cq); + + return 0; +} + +static void mlx5e_disable_cq(struct mlx5e_cq *cq) +{ + struct mlx5e_channel *c = cq->channel; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + mlx5_core_destroy_cq(mdev, &cq->mcq); +} + +static int mlx5e_open_cq(struct mlx5e_channel *c, + struct mlx5e_cq_param *param, + struct mlx5e_cq *cq, + u16 moderation_usecs, + u16 moderation_frames) +{ + int err; + struct mlx5e_priv *priv = c->priv; + struct mlx5_core_dev *mdev = priv->mdev; + + err = mlx5e_create_cq(c, param, cq); + if (err) + return err; + + err = mlx5e_enable_cq(cq, param); + if (err) + goto err_destroy_cq; + + err = mlx5_core_modify_cq_moderation(mdev, &cq->mcq, + moderation_usecs, + moderation_frames); + if (err) + goto err_destroy_cq; + + return 0; + +err_destroy_cq: + mlx5e_destroy_cq(cq); + + return err; +} + +static void mlx5e_close_cq(struct mlx5e_cq *cq) +{ + mlx5e_disable_cq(cq); + mlx5e_destroy_cq(cq); +} + +static int mlx5e_get_cpu(struct mlx5e_priv *priv, int ix) +{ + return cpumask_first(priv->mdev->priv.irq_info[ix].mask); +} + +static int mlx5e_open_tx_cqs(struct mlx5e_channel *c, + struct mlx5e_channel_param *cparam) +{ + struct mlx5e_priv *priv = c->priv; + int err; + int tc; + + for (tc = 0; tc < c->num_tc; tc++) { + err = mlx5e_open_cq(c, &cparam->tx_cq, &c->sq[tc].cq, + priv->params.tx_cq_moderation_usec, + priv->params.tx_cq_moderation_pkts); + if (err) + goto err_close_tx_cqs; + + c->sq[tc].cq.sqrq = &c->sq[tc]; + } + + return 0; + +err_close_tx_cqs: + for (tc--; tc >= 0; tc--) + mlx5e_close_cq(&c->sq[tc].cq); + + return err; +} + +static void mlx5e_close_tx_cqs(struct mlx5e_channel *c) +{ + int tc; + + for (tc = 0; tc < c->num_tc; tc++) + mlx5e_close_cq(&c->sq[tc].cq); +} + +static int mlx5e_open_sqs(struct mlx5e_channel *c, + struct mlx5e_channel_param *cparam) +{ + int err; + int tc; + + for (tc = 0; tc < c->num_tc; tc++) { + err = mlx5e_open_sq(c, tc, &cparam->sq, &c->sq[tc]); + if (err) + goto err_close_sqs; + } + + return 0; + +err_close_sqs: + for (tc--; tc >= 0; tc--) + mlx5e_close_sq(&c->sq[tc]); + + return err; +} + +static void mlx5e_close_sqs(struct mlx5e_channel *c) +{ + int tc; + + for (tc = 0; tc < c->num_tc; tc++) + mlx5e_close_sq(&c->sq[tc]); +} + +static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, + struct mlx5e_channel_param *cparam, + struct mlx5e_channel **cp) +{ + struct net_device *netdev = priv->netdev; + int cpu = mlx5e_get_cpu(priv, ix); + struct mlx5e_channel *c; + int err; + + c = kzalloc_node(sizeof(*c), GFP_KERNEL, cpu_to_node(cpu)); + if (!c) + return -ENOMEM; + + c->priv = priv; + c->ix = ix; + c->cpu = cpu; + c->pdev = &priv->mdev->pdev->dev; + c->netdev = priv->netdev; + c->mkey_be = cpu_to_be32(priv->mr.key); + c->num_tc = priv->num_tc; + + netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); + + err = mlx5e_open_tx_cqs(c, cparam); + if (err) + goto err_napi_del; + + err = mlx5e_open_cq(c, &cparam->rx_cq, &c->rq.cq, + priv->params.rx_cq_moderation_usec, + priv->params.rx_cq_moderation_pkts); + if (err) + goto err_close_tx_cqs; + c->rq.cq.sqrq = &c->rq; + + napi_enable(&c->napi); + + err = mlx5e_open_sqs(c, cparam); + if (err) + goto err_disable_napi; + + err = mlx5e_open_rq(c, &cparam->rq, &c->rq); + if (err) + goto err_close_sqs; + + netif_set_xps_queue(netdev, get_cpu_mask(c->cpu), ix); + *cp = c; + + return 0; + +err_close_sqs: + mlx5e_close_sqs(c); + +err_disable_napi: + napi_disable(&c->napi); + mlx5e_close_cq(&c->rq.cq); + +err_close_tx_cqs: + mlx5e_close_tx_cqs(c); + +err_napi_del: + netif_napi_del(&c->napi); + kfree(c); + + return err; +} + +static void mlx5e_close_channel(struct mlx5e_channel *c) +{ + mlx5e_close_rq(&c->rq); + mlx5e_close_sqs(c); + napi_disable(&c->napi); + mlx5e_close_cq(&c->rq.cq); + mlx5e_close_tx_cqs(c); + netif_napi_del(&c->napi); + kfree(c); +} + +static void mlx5e_build_rq_param(struct mlx5e_priv *priv, + struct mlx5e_rq_param *param) +{ + void *rqc = param->rqc; + void *wq = MLX5_ADDR_OF(rqc, rqc, wq); + + MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST); + MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); + MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); + MLX5_SET(wq, wq, log_wq_sz, priv->params.log_rq_size); + MLX5_SET(wq, wq, pd, priv->pdn); + + param->wq.numa = dev_to_node(&priv->mdev->pdev->dev); + param->wq.linear = 1; +} + +static void mlx5e_build_sq_param(struct mlx5e_priv *priv, + struct mlx5e_sq_param *param) +{ + void *sqc = param->sqc; + void *wq = MLX5_ADDR_OF(sqc, sqc, wq); + + MLX5_SET(wq, wq, log_wq_sz, priv->params.log_sq_size); + MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); + MLX5_SET(wq, wq, pd, priv->pdn); + + param->wq.numa = dev_to_node(&priv->mdev->pdev->dev); +} + +static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, + struct mlx5e_cq_param *param) +{ + void *cqc = param->cqc; + + MLX5_SET(cqc, cqc, uar_page, priv->cq_uar.index); +} + +static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, + struct mlx5e_cq_param *param) +{ + void *cqc = param->cqc; + + MLX5_SET(cqc, cqc, log_cq_size, priv->params.log_rq_size); + + mlx5e_build_common_cq_param(priv, param); +} + +static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, + struct mlx5e_cq_param *param) +{ + void *cqc = param->cqc; + + MLX5_SET(cqc, cqc, log_cq_size, priv->params.log_sq_size); + + mlx5e_build_common_cq_param(priv, param); +} + +static void mlx5e_build_channel_param(struct mlx5e_priv *priv, + struct mlx5e_channel_param *cparam) +{ + memset(cparam, 0, sizeof(*cparam)); + + mlx5e_build_rq_param(priv, &cparam->rq); + mlx5e_build_sq_param(priv, &cparam->sq); + mlx5e_build_rx_cq_param(priv, &cparam->rx_cq); + mlx5e_build_tx_cq_param(priv, &cparam->tx_cq); +} + +static int mlx5e_open_channels(struct mlx5e_priv *priv) +{ + struct mlx5e_channel_param cparam; + int err; + int i; + int j; + + priv->channel = kcalloc(priv->params.num_channels, + sizeof(struct mlx5e_channel *), GFP_KERNEL); + if (!priv->channel) + return -ENOMEM; + + mlx5e_build_channel_param(priv, &cparam); + for (i = 0; i < priv->params.num_channels; i++) { + err = mlx5e_open_channel(priv, i, &cparam, &priv->channel[i]); + if (err) + goto err_close_channels; + } + + for (j = 0; j < priv->params.num_channels; j++) { + err = mlx5e_wait_for_min_rx_wqes(&priv->channel[j]->rq); + if (err) + goto err_close_channels; + } + + return 0; + +err_close_channels: + for (i--; i >= 0; i--) + mlx5e_close_channel(priv->channel[i]); + + kfree(priv->channel); + + return err; +} + +static void mlx5e_close_channels(struct mlx5e_priv *priv) +{ + int i; + + for (i = 0; i < priv->params.num_channels; i++) + mlx5e_close_channel(priv->channel[i]); + + kfree(priv->channel); +} + +static int mlx5e_open_tis(struct mlx5e_priv *priv, int tc) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u32 in[MLX5_ST_SZ_DW(create_tis_in)]; + void *tisc = MLX5_ADDR_OF(create_tis_in, in, ctx); + + memset(in, 0, sizeof(in)); + + MLX5_SET(tisc, tisc, prio, tc); + + return mlx5_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]); +} + +static void mlx5e_close_tis(struct mlx5e_priv *priv, int tc) +{ + mlx5_destroy_tis(priv->mdev, priv->tisn[tc]); +} + +static int mlx5e_open_tises(struct mlx5e_priv *priv) +{ + int num_tc = priv->num_tc; + int err; + int tc; + + for (tc = 0; tc < num_tc; tc++) { + err = mlx5e_open_tis(priv, tc); + if (err) + goto err_close_tises; + } + + return 0; + +err_close_tises: + for (tc--; tc >= 0; tc--) + mlx5e_close_tis(priv, tc); + + return err; +} + +static void mlx5e_close_tises(struct mlx5e_priv *priv) +{ + int num_tc = priv->num_tc; + int tc; + + for (tc = 0; tc < num_tc; tc++) + mlx5e_close_tis(priv, tc); +} + +static int mlx5e_open_rqt(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u32 *in; + u32 out[MLX5_ST_SZ_DW(create_rqt_out)]; + void *rqtc; + int inlen; + int err; + int sz; + int i; + + sz = 1 << priv->params.rx_hash_log_tbl_sz; + + inlen = MLX5_ST_SZ_BYTES(create_rqt_in) + sizeof(u32) * sz; + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + rqtc = MLX5_ADDR_OF(create_rqt_in, in, rqt_context); + + MLX5_SET(rqtc, rqtc, rqt_actual_size, sz); + MLX5_SET(rqtc, rqtc, rqt_max_size, sz); + + for (i = 0; i < sz; i++) { + int ix = i % priv->params.num_channels; + + MLX5_SET(rqtc, rqtc, rq_num[i], priv->channel[ix]->rq.rqn); + } + + MLX5_SET(create_rqt_in, in, opcode, MLX5_CMD_OP_CREATE_RQT); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(mdev, in, inlen, out, sizeof(out)); + if (!err) + priv->rqtn = MLX5_GET(create_rqt_out, out, rqtn); + + kvfree(in); + + return err; +} + +static void mlx5e_close_rqt(struct mlx5e_priv *priv) +{ + u32 in[MLX5_ST_SZ_DW(destroy_rqt_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_rqt_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(destroy_rqt_in, in, opcode, MLX5_CMD_OP_DESTROY_RQT); + MLX5_SET(destroy_rqt_in, in, rqtn, priv->rqtn); + + mlx5_cmd_exec_check_status(priv->mdev, in, sizeof(in), out, + sizeof(out)); +} + +static void mlx5e_build_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, int tt) +{ + void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); + +#define ROUGH_MAX_L2_L3_HDR_SZ 256 + +#define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ + MLX5_HASH_FIELD_SEL_DST_IP) + +#define MLX5_HASH_ALL (MLX5_HASH_FIELD_SEL_SRC_IP |\ + MLX5_HASH_FIELD_SEL_DST_IP |\ + MLX5_HASH_FIELD_SEL_L4_SPORT |\ + MLX5_HASH_FIELD_SEL_L4_DPORT) + + if (priv->params.lro_en) { + MLX5_SET(tirc, tirc, lro_enable_mask, + MLX5_TIRC_LRO_ENABLE_MASK_IPV4_LRO | + MLX5_TIRC_LRO_ENABLE_MASK_IPV6_LRO); + MLX5_SET(tirc, tirc, lro_max_ip_payload_size, + (priv->params.lro_wqe_sz - + ROUGH_MAX_L2_L3_HDR_SZ) >> 8); + MLX5_SET(tirc, tirc, lro_timeout_period_usecs, + MLX5_CAP_ETH(priv->mdev, + lro_timer_supported_periods[3])); + } + + switch (tt) { + case MLX5E_TT_ANY: + MLX5_SET(tirc, tirc, disp_type, + MLX5_TIRC_DISP_TYPE_DIRECT); + MLX5_SET(tirc, tirc, inline_rqn, + priv->channel[0]->rq.rqn); + break; + default: + MLX5_SET(tirc, tirc, disp_type, + MLX5_TIRC_DISP_TYPE_INDIRECT); + MLX5_SET(tirc, tirc, indirect_table, + priv->rqtn); + MLX5_SET(tirc, tirc, rx_hash_fn, + MLX5_TIRC_RX_HASH_FN_HASH_TOEPLITZ); + MLX5_SET(tirc, tirc, rx_hash_symmetric, 1); + netdev_rss_key_fill(MLX5_ADDR_OF(tirc, tirc, + rx_hash_toeplitz_key), + MLX5_FLD_SZ_BYTES(tirc, + rx_hash_toeplitz_key)); + break; + } + + switch (tt) { + case MLX5E_TT_IPV4_TCP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_TCP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_ALL); + break; + + case MLX5E_TT_IPV6_TCP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_TCP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_ALL); + break; + + case MLX5E_TT_IPV4_UDP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_UDP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_ALL); + break; + + case MLX5E_TT_IPV6_UDP: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, + MLX5_L4_PROT_TYPE_UDP); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_ALL); + break; + + case MLX5E_TT_IPV4: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV4); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP); + break; + + case MLX5E_TT_IPV6: + MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, + MLX5_L3_PROT_TYPE_IPV6); + MLX5_SET(rx_hash_field_select, hfso, selected_fields, + MLX5_HASH_IP); + break; + } +} + +static int mlx5e_open_tir(struct mlx5e_priv *priv, int tt) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u32 *in; + void *tirc; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(create_tir_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); + + mlx5e_build_tir_ctx(priv, tirc, tt); + + err = mlx5_create_tir(mdev, in, inlen, &priv->tirn[tt]); + + kvfree(in); + + return err; +} + +static void mlx5e_close_tir(struct mlx5e_priv *priv, int tt) +{ + mlx5_destroy_tir(priv->mdev, priv->tirn[tt]); +} + +static int mlx5e_open_tirs(struct mlx5e_priv *priv) +{ + int err; + int i; + + for (i = 0; i < MLX5E_NUM_TT; i++) { + err = mlx5e_open_tir(priv, i); + if (err) + goto err_close_tirs; + } + + return 0; + +err_close_tirs: + for (i--; i >= 0; i--) + mlx5e_close_tir(priv, i); + + return err; +} + +static void mlx5e_close_tirs(struct mlx5e_priv *priv) +{ + int i; + + for (i = 0; i < MLX5E_NUM_TT; i++) + mlx5e_close_tir(priv, i); +} + +int mlx5e_open_locked(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + int actual_mtu; + int num_txqs; + int err; + + num_txqs = roundup_pow_of_two(priv->params.num_channels) * + priv->params.num_tc; + netif_set_real_num_tx_queues(netdev, num_txqs); + netif_set_real_num_rx_queues(netdev, priv->params.num_channels); + + err = mlx5_set_port_mtu(mdev, netdev->mtu); + if (err) { + netdev_err(netdev, "%s: mlx5_set_port_mtu failed %d\n", + __func__, err); + return err; + } + + err = mlx5_query_port_oper_mtu(mdev, &actual_mtu); + if (err) { + netdev_err(netdev, "%s: mlx5_query_port_oper_mtu failed %d\n", + __func__, err); + return err; + } + + if (actual_mtu != netdev->mtu) + netdev_warn(netdev, "%s: Failed to set MTU to %d\n", + __func__, netdev->mtu); + + netdev->mtu = actual_mtu; + + err = mlx5e_open_tises(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_open_tises failed, %d\n", + __func__, err); + return err; + } + + err = mlx5e_open_channels(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_open_channels failed, %d\n", + __func__, err); + goto err_close_tises; + } + + err = mlx5e_open_rqt(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_open_rqt failed, %d\n", + __func__, err); + goto err_close_channels; + } + + err = mlx5e_open_tirs(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_open_tir failed, %d\n", + __func__, err); + goto err_close_rqls; + } + + err = mlx5e_open_flow_table(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_open_flow_table failed, %d\n", + __func__, err); + goto err_close_tirs; + } + + err = mlx5e_add_all_vlan_rules(priv); + if (err) { + netdev_err(netdev, "%s: mlx5e_add_all_vlan_rules failed, %d\n", + __func__, err); + goto err_close_flow_table; + } + + mlx5e_init_eth_addr(priv); + + set_bit(MLX5E_STATE_OPENED, &priv->state); + + mlx5e_update_carrier(priv); + mlx5e_set_rx_mode_core(priv); + + schedule_delayed_work(&priv->update_stats_work, 0); + return 0; + +err_close_flow_table: + mlx5e_close_flow_table(priv); + +err_close_tirs: + mlx5e_close_tirs(priv); + +err_close_rqls: + mlx5e_close_rqt(priv); + +err_close_channels: + mlx5e_close_channels(priv); + +err_close_tises: + mlx5e_close_tises(priv); + + return err; +} + +static int mlx5e_open(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + mutex_lock(&priv->state_lock); + err = mlx5e_open_locked(netdev); + mutex_unlock(&priv->state_lock); + + return err; +} + +int mlx5e_close_locked(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + clear_bit(MLX5E_STATE_OPENED, &priv->state); + + mlx5e_set_rx_mode_core(priv); + mlx5e_del_all_vlan_rules(priv); + netif_carrier_off(priv->netdev); + mlx5e_close_flow_table(priv); + mlx5e_close_tirs(priv); + mlx5e_close_rqt(priv); + mlx5e_close_channels(priv); + mlx5e_close_tises(priv); + + return 0; +} + +static int mlx5e_close(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + mutex_lock(&priv->state_lock); + err = mlx5e_close_locked(netdev); + mutex_unlock(&priv->state_lock); + + return err; +} + +int mlx5e_update_priv_params(struct mlx5e_priv *priv, + struct mlx5e_params *new_params) +{ + int err = 0; + int was_opened; + + WARN_ON(!mutex_is_locked(&priv->state_lock)); + + was_opened = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (was_opened) + mlx5e_close_locked(priv->netdev); + + priv->params = *new_params; + + if (was_opened) + err = mlx5e_open_locked(priv->netdev); + + return err; +} + +static struct rtnl_link_stats64 * +mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_vport_stats *vstats = &priv->stats.vport; + + stats->rx_packets = vstats->rx_packets; + stats->rx_bytes = vstats->rx_bytes; + stats->tx_packets = vstats->tx_packets; + stats->tx_bytes = vstats->tx_bytes; + stats->multicast = vstats->rx_multicast_packets + + vstats->tx_multicast_packets; + stats->tx_errors = vstats->tx_error_packets; + stats->rx_errors = vstats->rx_error_packets; + stats->tx_dropped = vstats->tx_queue_dropped; + stats->rx_crc_errors = 0; + stats->rx_length_errors = 0; + + return stats; +} + +static void mlx5e_set_rx_mode(struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + schedule_work(&priv->set_rx_mode_work); +} + +static int mlx5e_set_mac(struct net_device *netdev, void *addr) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct sockaddr *saddr = addr; + + if (!is_valid_ether_addr(saddr->sa_data)) + return -EADDRNOTAVAIL; + + netif_addr_lock_bh(netdev); + ether_addr_copy(netdev->dev_addr, saddr->sa_data); + netif_addr_unlock_bh(netdev); + + schedule_work(&priv->set_rx_mode_work); + + return 0; +} + +static int mlx5e_set_features(struct net_device *netdev, + netdev_features_t features) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + netdev_features_t changes = features ^ netdev->features; + struct mlx5e_params new_params; + bool update_params = false; + + mutex_lock(&priv->state_lock); + new_params = priv->params; + + if (changes & NETIF_F_LRO) { + new_params.lro_en = !!(features & NETIF_F_LRO); + update_params = true; + } + + if (update_params) + mlx5e_update_priv_params(priv, &new_params); + + if (changes & NETIF_F_HW_VLAN_CTAG_FILTER) { + if (features & NETIF_F_HW_VLAN_CTAG_FILTER) + mlx5e_enable_vlan_filter(priv); + else + mlx5e_disable_vlan_filter(priv); + } + + mutex_unlock(&priv->state_lock); + + return 0; +} + +static int mlx5e_change_mtu(struct net_device *netdev, int new_mtu) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + int max_mtu; + int err = 0; + + err = mlx5_query_port_max_mtu(mdev, &max_mtu); + if (err) + return err; + + if (new_mtu > max_mtu || new_mtu < MLX5E_PARAMS_MIN_MTU) { + netdev_err(netdev, "%s: Bad MTU size, mtu must be [%d-%d]\n", + __func__, MLX5E_PARAMS_MIN_MTU, max_mtu); + return -EINVAL; + } + + mutex_lock(&priv->state_lock); + netdev->mtu = new_mtu; + err = mlx5e_update_priv_params(priv, &priv->params); + mutex_unlock(&priv->state_lock); + + return err; +} + +static struct net_device_ops mlx5e_netdev_ops = { + .ndo_open = mlx5e_open, + .ndo_stop = mlx5e_close, + .ndo_start_xmit = mlx5e_xmit, + .ndo_get_stats64 = mlx5e_get_stats, + .ndo_set_rx_mode = mlx5e_set_rx_mode, + .ndo_set_mac_address = mlx5e_set_mac, + .ndo_vlan_rx_add_vid = mlx5e_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = mlx5e_vlan_rx_kill_vid, + .ndo_set_features = mlx5e_set_features, + .ndo_change_mtu = mlx5e_change_mtu, +}; + +static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) +{ + if (MLX5_CAP_GEN(mdev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return -ENOTSUPP; + if (!MLX5_CAP_GEN(mdev, eth_net_offloads) || + !MLX5_CAP_GEN(mdev, nic_flow_table) || + !MLX5_CAP_ETH(mdev, csum_cap) || + !MLX5_CAP_ETH(mdev, max_lso_cap) || + !MLX5_CAP_ETH(mdev, vlan_cap) || + !MLX5_CAP_ETH(mdev, rss_ind_tbl_cap)) { + mlx5_core_warn(mdev, + "Not creating net device, some required device capabilities are missing\n"); + return -ENOTSUPP; + } + return 0; +} + +static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, + struct net_device *netdev, + int num_comp_vectors) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + priv->params.log_sq_size = + MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE; + priv->params.log_rq_size = + MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE; + priv->params.rx_cq_moderation_usec = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; + priv->params.rx_cq_moderation_pkts = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; + priv->params.tx_cq_moderation_usec = + MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC; + priv->params.tx_cq_moderation_pkts = + MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; + priv->params.min_rx_wqes = + MLX5E_PARAMS_DEFAULT_MIN_RX_WQES; + priv->params.rx_hash_log_tbl_sz = + (order_base_2(num_comp_vectors) > + MLX5E_PARAMS_DEFAULT_RX_HASH_LOG_TBL_SZ) ? + order_base_2(num_comp_vectors) : + MLX5E_PARAMS_DEFAULT_RX_HASH_LOG_TBL_SZ; + priv->params.num_tc = 1; + priv->params.default_vlan_prio = 0; + + priv->params.lro_en = false && !!MLX5_CAP_ETH(priv->mdev, lro_cap); + priv->params.lro_wqe_sz = + MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + + priv->mdev = mdev; + priv->netdev = netdev; + priv->params.num_channels = num_comp_vectors; + priv->order_base_2_num_channels = order_base_2(num_comp_vectors); + priv->queue_mapping_channel_mask = + roundup_pow_of_two(num_comp_vectors) - 1; + priv->num_tc = priv->params.num_tc; + priv->default_vlan_prio = priv->params.default_vlan_prio; + + spin_lock_init(&priv->async_events_spinlock); + mutex_init(&priv->state_lock); + + INIT_WORK(&priv->update_carrier_work, mlx5e_update_carrier_work); + INIT_WORK(&priv->set_rx_mode_work, mlx5e_set_rx_mode_work); + INIT_DELAYED_WORK(&priv->update_stats_work, mlx5e_update_stats_work); +} + +static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + mlx5_query_vport_mac_address(priv->mdev, netdev->dev_addr); +} + +static void mlx5e_build_netdev(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + + SET_NETDEV_DEV(netdev, &mdev->pdev->dev); + + if (priv->num_tc > 1) { + mlx5e_netdev_ops.ndo_select_queue = mlx5e_select_queue; + mlx5e_netdev_ops.ndo_start_xmit = mlx5e_xmit_multi_tc; + } + + netdev->netdev_ops = &mlx5e_netdev_ops; + netdev->watchdog_timeo = 15 * HZ; + + netdev->ethtool_ops = &mlx5e_ethtool_ops; + + netdev->vlan_features |= NETIF_F_IP_CSUM; + netdev->vlan_features |= NETIF_F_IPV6_CSUM; + netdev->vlan_features |= NETIF_F_GRO; + netdev->vlan_features |= NETIF_F_TSO; + netdev->vlan_features |= NETIF_F_TSO6; + netdev->vlan_features |= NETIF_F_RXCSUM; + netdev->vlan_features |= NETIF_F_RXHASH; + + if (!!MLX5_CAP_ETH(mdev, lro_cap)) + netdev->vlan_features |= NETIF_F_LRO; + + netdev->hw_features = netdev->vlan_features; + netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX; + netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX; + netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER; + + netdev->features = netdev->hw_features; + if (!priv->params.lro_en) + netdev->features &= ~NETIF_F_LRO; + + netdev->features |= NETIF_F_HIGHDMA; + + netdev->priv_flags |= IFF_UNICAST_FLT; + + mlx5e_set_netdev_dev_addr(netdev); +} + +static int mlx5e_create_mkey(struct mlx5e_priv *priv, u32 pdn, + struct mlx5_core_mr *mr) +{ + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_create_mkey_mbox_in *in; + int err; + + in = mlx5_vzalloc(sizeof(*in)); + if (!in) + return -ENOMEM; + + in->seg.flags = MLX5_PERM_LOCAL_WRITE | + MLX5_PERM_LOCAL_READ | + MLX5_ACCESS_MODE_PA; + in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); + in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); + + err = mlx5_core_create_mkey(mdev, mr, in, sizeof(*in), NULL, NULL, + NULL); + + kvfree(in); + + return err; +} + +static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) +{ + struct net_device *netdev; + struct mlx5e_priv *priv; + int ncv = mdev->priv.eq_table.num_comp_vectors; + int err; + + if (mlx5e_check_required_hca_cap(mdev)) + return NULL; + + netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), + roundup_pow_of_two(ncv) * MLX5E_MAX_NUM_TC, + ncv); + if (!netdev) { + mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); + return NULL; + } + + mlx5e_build_netdev_priv(mdev, netdev, ncv); + mlx5e_build_netdev(netdev); + + netif_carrier_off(netdev); + + priv = netdev_priv(netdev); + + err = mlx5_alloc_map_uar(mdev, &priv->cq_uar); + if (err) { + netdev_err(netdev, "%s: mlx5_alloc_map_uar failed, %d\n", + __func__, err); + goto err_free_netdev; + } + + err = mlx5_core_alloc_pd(mdev, &priv->pdn); + if (err) { + netdev_err(netdev, "%s: mlx5_core_alloc_pd failed, %d\n", + __func__, err); + goto err_unmap_free_uar; + } + + err = mlx5e_create_mkey(priv, priv->pdn, &priv->mr); + if (err) { + netdev_err(netdev, "%s: mlx5e_create_mkey failed, %d\n", + __func__, err); + goto err_dealloc_pd; + } + + err = register_netdev(netdev); + if (err) { + netdev_err(netdev, "%s: register_netdev failed, %d\n", + __func__, err); + goto err_destroy_mkey; + } + + mlx5e_enable_async_events(priv); + + return priv; + +err_destroy_mkey: + mlx5_core_destroy_mkey(mdev, &priv->mr); + +err_dealloc_pd: + mlx5_core_dealloc_pd(mdev, priv->pdn); + +err_unmap_free_uar: + mlx5_unmap_free_uar(mdev, &priv->cq_uar); + +err_free_netdev: + free_netdev(netdev); + + return NULL; +} + +static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) +{ + struct mlx5e_priv *priv = vpriv; + struct net_device *netdev = priv->netdev; + + unregister_netdev(netdev); + mlx5_core_destroy_mkey(priv->mdev, &priv->mr); + mlx5_core_dealloc_pd(priv->mdev, priv->pdn); + mlx5_unmap_free_uar(priv->mdev, &priv->cq_uar); + mlx5e_disable_async_events(priv); + flush_scheduled_work(); + free_netdev(netdev); +} + +static void *mlx5e_get_netdev(void *vpriv) +{ + struct mlx5e_priv *priv = vpriv; + + return priv->netdev; +} + +static struct mlx5_interface mlx5e_interface = { + .add = mlx5e_create_netdev, + .remove = mlx5e_destroy_netdev, + .event = mlx5e_async_event, + .protocol = MLX5_INTERFACE_PROTOCOL_ETH, + .get_dev = mlx5e_get_netdev, +}; + +void mlx5e_init(void) +{ + mlx5_register_interface(&mlx5e_interface); +} + +void mlx5e_cleanup(void) +{ + mlx5_unregister_interface(&mlx5e_interface); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c new file mode 100644 index 000000000000..ce1317cdabd7 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -0,0 +1,249 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/tcp.h> +#include "en.h" + +static inline int mlx5e_alloc_rx_wqe(struct mlx5e_rq *rq, + struct mlx5e_rx_wqe *wqe, u16 ix) +{ + struct sk_buff *skb; + dma_addr_t dma_addr; + + skb = netdev_alloc_skb(rq->netdev, rq->wqe_sz); + if (unlikely(!skb)) + return -ENOMEM; + + skb_reserve(skb, MLX5E_NET_IP_ALIGN); + + dma_addr = dma_map_single(rq->pdev, + /* hw start padding */ + skb->data - MLX5E_NET_IP_ALIGN, + /* hw end padding */ + rq->wqe_sz, + DMA_FROM_DEVICE); + + if (unlikely(dma_mapping_error(rq->pdev, dma_addr))) + goto err_free_skb; + + *((dma_addr_t *)skb->cb) = dma_addr; + wqe->data.addr = cpu_to_be64(dma_addr + MLX5E_NET_IP_ALIGN); + + rq->skb[ix] = skb; + + return 0; + +err_free_skb: + dev_kfree_skb(skb); + + return -ENOMEM; +} + +bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq) +{ + struct mlx5_wq_ll *wq = &rq->wq; + + if (unlikely(!test_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state))) + return false; + + while (!mlx5_wq_ll_is_full(wq)) { + struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(wq, wq->head); + + if (unlikely(mlx5e_alloc_rx_wqe(rq, wqe, wq->head))) + break; + + mlx5_wq_ll_push(wq, be16_to_cpu(wqe->next.next_wqe_index)); + } + + /* ensure wqes are visible to device before updating doorbell record */ + dma_wmb(); + + mlx5_wq_ll_update_db_record(wq); + + return !mlx5_wq_ll_is_full(wq); +} + +static void mlx5e_lro_update_hdr(struct sk_buff *skb, struct mlx5_cqe64 *cqe) +{ + struct ethhdr *eth = (struct ethhdr *)(skb->data); + struct iphdr *ipv4 = (struct iphdr *)(skb->data + ETH_HLEN); + struct ipv6hdr *ipv6 = (struct ipv6hdr *)(skb->data + ETH_HLEN); + struct tcphdr *tcp; + + u8 l4_hdr_type = get_cqe_l4_hdr_type(cqe); + int tcp_ack = ((CQE_L4_HDR_TYPE_TCP_ACK_NO_DATA == l4_hdr_type) || + (CQE_L4_HDR_TYPE_TCP_ACK_AND_DATA == l4_hdr_type)); + + u16 tot_len = be32_to_cpu(cqe->byte_cnt) - ETH_HLEN; + + if (eth->h_proto == htons(ETH_P_IP)) { + tcp = (struct tcphdr *)(skb->data + ETH_HLEN + + sizeof(struct iphdr)); + ipv6 = NULL; + } else { + tcp = (struct tcphdr *)(skb->data + ETH_HLEN + + sizeof(struct ipv6hdr)); + ipv4 = NULL; + } + + if (get_cqe_lro_tcppsh(cqe)) + tcp->psh = 1; + + if (tcp_ack) { + tcp->ack = 1; + tcp->ack_seq = cqe->lro_ack_seq_num; + tcp->window = cqe->lro_tcp_win; + } + + if (ipv4) { + ipv4->ttl = cqe->lro_min_ttl; + ipv4->tot_len = cpu_to_be16(tot_len); + ipv4->check = 0; + ipv4->check = ip_fast_csum((unsigned char *)ipv4, + ipv4->ihl); + } else { + ipv6->hop_limit = cqe->lro_min_ttl; + ipv6->payload_len = cpu_to_be16(tot_len - + sizeof(struct ipv6hdr)); + } +} + +static inline void mlx5e_skb_set_hash(struct mlx5_cqe64 *cqe, + struct sk_buff *skb) +{ + u8 cht = cqe->rss_hash_type; + int ht = (cht & CQE_RSS_HTYPE_L4) ? PKT_HASH_TYPE_L4 : + (cht & CQE_RSS_HTYPE_IP) ? PKT_HASH_TYPE_L3 : + PKT_HASH_TYPE_NONE; + skb_set_hash(skb, be32_to_cpu(cqe->rss_hash_result), ht); +} + +static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe, + struct mlx5e_rq *rq, + struct sk_buff *skb) +{ + struct net_device *netdev = rq->netdev; + u32 cqe_bcnt = be32_to_cpu(cqe->byte_cnt); + int lro_num_seg; + + skb_put(skb, cqe_bcnt); + + lro_num_seg = be32_to_cpu(cqe->srqn) >> 24; + if (lro_num_seg > 1) { + mlx5e_lro_update_hdr(skb, cqe); + skb_shinfo(skb)->gso_size = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + rq->stats.lro_packets++; + rq->stats.lro_bytes += cqe_bcnt; + } + + if (likely(netdev->features & NETIF_F_RXCSUM) && + (cqe->hds_ip_ext & CQE_L2_OK) && + (cqe->hds_ip_ext & CQE_L3_OK) && + (cqe->hds_ip_ext & CQE_L4_OK)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + } else { + skb->ip_summed = CHECKSUM_NONE; + rq->stats.csum_none++; + } + + skb->protocol = eth_type_trans(skb, netdev); + + skb_record_rx_queue(skb, rq->ix); + + if (likely(netdev->features & NETIF_F_RXHASH)) + mlx5e_skb_set_hash(cqe, skb); + + if (cqe_has_vlan(cqe)) + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), + be16_to_cpu(cqe->vlan_info)); +} + +bool mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget) +{ + struct mlx5e_rq *rq = cq->sqrq; + int i; + + /* avoid accessing cq (dma coherent memory) if not needed */ + if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags)) + return false; + + for (i = 0; i < budget; i++) { + struct mlx5e_rx_wqe *wqe; + struct mlx5_cqe64 *cqe; + struct sk_buff *skb; + __be16 wqe_counter_be; + u16 wqe_counter; + + cqe = mlx5e_get_cqe(cq); + if (!cqe) + break; + + wqe_counter_be = cqe->wqe_counter; + wqe_counter = be16_to_cpu(wqe_counter_be); + wqe = mlx5_wq_ll_get_wqe(&rq->wq, wqe_counter); + skb = rq->skb[wqe_counter]; + rq->skb[wqe_counter] = NULL; + + dma_unmap_single(rq->pdev, + *((dma_addr_t *)skb->cb), + skb_end_offset(skb), + DMA_FROM_DEVICE); + + if (unlikely((cqe->op_own >> 4) != MLX5_CQE_RESP_SEND)) { + rq->stats.wqe_err++; + dev_kfree_skb(skb); + goto wq_ll_pop; + } + + mlx5e_build_rx_skb(cqe, rq, skb); + rq->stats.packets++; + napi_gro_receive(cq->napi, skb); + +wq_ll_pop: + mlx5_wq_ll_pop(&rq->wq, wqe_counter_be, + &wqe->next.next_wqe_index); + } + + mlx5_cqwq_update_db_record(&cq->wq); + + /* ensure cq space is freed before enabling more cqes */ + wmb(); + + if (i == budget) { + set_bit(MLX5E_CQ_HAS_CQES, &cq->flags); + return true; + } + + return false; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c new file mode 100644 index 000000000000..8020986cdaf6 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -0,0 +1,344 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/tcp.h> +#include <linux/if_vlan.h> +#include "en.h" + +static void mlx5e_dma_pop_last_pushed(struct mlx5e_sq *sq, dma_addr_t *addr, + u32 *size) +{ + sq->dma_fifo_pc--; + *addr = sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].addr; + *size = sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].size; +} + +static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, struct sk_buff *skb) +{ + dma_addr_t addr; + u32 size; + int i; + + for (i = 0; i < MLX5E_TX_SKB_CB(skb)->num_dma; i++) { + mlx5e_dma_pop_last_pushed(sq, &addr, &size); + dma_unmap_single(sq->pdev, addr, size, DMA_TO_DEVICE); + } +} + +static inline void mlx5e_dma_push(struct mlx5e_sq *sq, dma_addr_t addr, + u32 size) +{ + sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].addr = addr; + sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].size = size; + sq->dma_fifo_pc++; +} + +static inline void mlx5e_dma_get(struct mlx5e_sq *sq, u32 i, dma_addr_t *addr, + u32 *size) +{ + *addr = sq->dma_fifo[i & sq->dma_fifo_mask].addr; + *size = sq->dma_fifo[i & sq->dma_fifo_mask].size; +} + +u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, + void *accel_priv, select_queue_fallback_t fallback) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int channel_ix = fallback(dev, skb); + int up = skb_vlan_tag_present(skb) ? + skb->vlan_tci >> VLAN_PRIO_SHIFT : + priv->default_vlan_prio; + int tc = netdev_get_prio_tc_map(dev, up); + + return (tc << priv->order_base_2_num_channels) | channel_ix; +} + +static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, + struct sk_buff *skb) +{ +#define MLX5E_MIN_INLINE 16 /* eth header with vlan (w/o next ethertype) */ + return MLX5E_MIN_INLINE; +} + +static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs) +{ + struct vlan_ethhdr *vhdr = (struct vlan_ethhdr *)start; + int cpy1_sz = 2 * ETH_ALEN; + int cpy2_sz = ihs - cpy1_sz - VLAN_HLEN; + + skb_copy_from_linear_data(skb, vhdr, cpy1_sz); + skb_pull_inline(skb, cpy1_sz); + vhdr->h_vlan_proto = skb->vlan_proto; + vhdr->h_vlan_TCI = cpu_to_be16(skb_vlan_tag_get(skb)); + skb_copy_from_linear_data(skb, &vhdr->h_vlan_encapsulated_proto, + cpy2_sz); + skb_pull_inline(skb, cpy2_sz); +} + +static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) +{ + struct mlx5_wq_cyc *wq = &sq->wq; + + u16 pi = sq->pc & wq->sz_m1; + struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(wq, pi); + + struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; + struct mlx5_wqe_eth_seg *eseg = &wqe->eth; + struct mlx5_wqe_data_seg *dseg; + + u8 opcode = MLX5_OPCODE_SEND; + dma_addr_t dma_addr = 0; + u16 headlen; + u16 ds_cnt; + u16 ihs; + int i; + + memset(wqe, 0, sizeof(*wqe)); + + if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) + eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM; + else + sq->stats.csum_offload_none++; + + if (skb_is_gso(skb)) { + u32 payload_len; + int num_pkts; + + eseg->mss = cpu_to_be16(skb_shinfo(skb)->gso_size); + opcode = MLX5_OPCODE_LSO; + ihs = skb_transport_offset(skb) + tcp_hdrlen(skb); + payload_len = skb->len - ihs; + num_pkts = (payload_len / skb_shinfo(skb)->gso_size) + + !!(payload_len % skb_shinfo(skb)->gso_size); + MLX5E_TX_SKB_CB(skb)->num_bytes = skb->len + + (num_pkts - 1) * ihs; + sq->stats.tso_packets++; + sq->stats.tso_bytes += payload_len; + } else { + ihs = mlx5e_get_inline_hdr_size(sq, skb); + MLX5E_TX_SKB_CB(skb)->num_bytes = max_t(unsigned int, skb->len, + ETH_ZLEN); + } + + if (skb_vlan_tag_present(skb)) { + mlx5e_insert_vlan(eseg->inline_hdr_start, skb, ihs); + } else { + skb_copy_from_linear_data(skb, eseg->inline_hdr_start, ihs); + skb_pull_inline(skb, ihs); + } + + eseg->inline_hdr_sz = cpu_to_be16(ihs); + + ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; + ds_cnt += DIV_ROUND_UP(ihs - sizeof(eseg->inline_hdr_start), + MLX5_SEND_WQE_DS); + dseg = (struct mlx5_wqe_data_seg *)cseg + ds_cnt; + + MLX5E_TX_SKB_CB(skb)->num_dma = 0; + + headlen = skb_headlen(skb); + if (headlen) { + dma_addr = dma_map_single(sq->pdev, skb->data, headlen, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(sq->pdev, dma_addr))) + goto dma_unmap_wqe_err; + + dseg->addr = cpu_to_be64(dma_addr); + dseg->lkey = sq->mkey_be; + dseg->byte_count = cpu_to_be32(headlen); + + mlx5e_dma_push(sq, dma_addr, headlen); + MLX5E_TX_SKB_CB(skb)->num_dma++; + + dseg++; + } + + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i]; + int fsz = skb_frag_size(frag); + + dma_addr = skb_frag_dma_map(sq->pdev, frag, 0, fsz, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(sq->pdev, dma_addr))) + goto dma_unmap_wqe_err; + + dseg->addr = cpu_to_be64(dma_addr); + dseg->lkey = sq->mkey_be; + dseg->byte_count = cpu_to_be32(fsz); + + mlx5e_dma_push(sq, dma_addr, fsz); + MLX5E_TX_SKB_CB(skb)->num_dma++; + + dseg++; + } + + ds_cnt += MLX5E_TX_SKB_CB(skb)->num_dma; + + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt); + cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; + + sq->skb[pi] = skb; + + MLX5E_TX_SKB_CB(skb)->num_wqebbs = DIV_ROUND_UP(ds_cnt, + MLX5_SEND_WQEBB_NUM_DS); + sq->pc += MLX5E_TX_SKB_CB(skb)->num_wqebbs; + + netdev_tx_sent_queue(sq->txq, MLX5E_TX_SKB_CB(skb)->num_bytes); + + if (unlikely(!mlx5e_sq_has_room_for(sq, MLX5_SEND_WQE_MAX_WQEBBS))) { + netif_tx_stop_queue(sq->txq); + sq->stats.stopped++; + } + + if (!skb->xmit_more || netif_xmit_stopped(sq->txq)) + mlx5e_tx_notify_hw(sq, wqe); + + sq->stats.packets++; + return NETDEV_TX_OK; + +dma_unmap_wqe_err: + sq->stats.dropped++; + mlx5e_dma_unmap_wqe_err(sq, skb); + + dev_kfree_skb_any(skb); + + return NETDEV_TX_OK; +} + +netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int ix = skb->queue_mapping; + int tc = 0; + struct mlx5e_channel *c = priv->channel[ix]; + struct mlx5e_sq *sq = &c->sq[tc]; + + return mlx5e_sq_xmit(sq, skb); +} + +netdev_tx_t mlx5e_xmit_multi_tc(struct sk_buff *skb, struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int ix = skb->queue_mapping & priv->queue_mapping_channel_mask; + int tc = skb->queue_mapping >> priv->order_base_2_num_channels; + struct mlx5e_channel *c = priv->channel[ix]; + struct mlx5e_sq *sq = &c->sq[tc]; + + return mlx5e_sq_xmit(sq, skb); +} + +bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq) +{ + struct mlx5e_sq *sq; + u32 dma_fifo_cc; + u32 nbytes; + u16 npkts; + u16 sqcc; + int i; + + /* avoid accessing cq (dma coherent memory) if not needed */ + if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags)) + return false; + + sq = cq->sqrq; + + npkts = 0; + nbytes = 0; + + /* sq->cc must be updated only after mlx5_cqwq_update_db_record(), + * otherwise a cq overrun may occur + */ + sqcc = sq->cc; + + /* avoid dirtying sq cache line every cqe */ + dma_fifo_cc = sq->dma_fifo_cc; + + for (i = 0; i < MLX5E_TX_CQ_POLL_BUDGET; i++) { + struct mlx5_cqe64 *cqe; + struct sk_buff *skb; + u16 ci; + int j; + + cqe = mlx5e_get_cqe(cq); + if (!cqe) + break; + + ci = sqcc & sq->wq.sz_m1; + skb = sq->skb[ci]; + + if (unlikely(!skb)) { /* nop */ + sq->stats.nop++; + sqcc++; + goto free_skb; + } + + for (j = 0; j < MLX5E_TX_SKB_CB(skb)->num_dma; j++) { + dma_addr_t addr; + u32 size; + + mlx5e_dma_get(sq, dma_fifo_cc, &addr, &size); + dma_fifo_cc++; + dma_unmap_single(sq->pdev, addr, size, DMA_TO_DEVICE); + } + + npkts++; + nbytes += MLX5E_TX_SKB_CB(skb)->num_bytes; + sqcc += MLX5E_TX_SKB_CB(skb)->num_wqebbs; + +free_skb: + dev_kfree_skb(skb); + } + + mlx5_cqwq_update_db_record(&cq->wq); + + /* ensure cq space is freed before enabling more cqes */ + wmb(); + + sq->dma_fifo_cc = dma_fifo_cc; + sq->cc = sqcc; + + netdev_tx_completed_queue(sq->txq, npkts, nbytes); + + if (netif_tx_queue_stopped(sq->txq) && + mlx5e_sq_has_room_for(sq, MLX5_SEND_WQE_MAX_WQEBBS) && + likely(test_bit(MLX5E_SQ_STATE_WAKE_TXQ_ENABLE, &sq->state))) { + netif_tx_wake_queue(sq->txq); + sq->stats.wake++; + } + if (i == MLX5E_TX_CQ_POLL_BUDGET) { + set_bit(MLX5E_CQ_HAS_CQES, &cq->flags); + return true; + } + + return false; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c new file mode 100644 index 000000000000..088bc424157c --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -0,0 +1,107 @@ +/* + * Copyright (c) 2015, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq) +{ + struct mlx5_cqwq *wq = &cq->wq; + u32 ci = mlx5_cqwq_get_ci(wq); + struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(wq, ci); + int cqe_ownership_bit = cqe->op_own & MLX5_CQE_OWNER_MASK; + int sw_ownership_val = mlx5_cqwq_get_wrap_cnt(wq) & 1; + + if (cqe_ownership_bit != sw_ownership_val) + return NULL; + + mlx5_cqwq_pop(wq); + + /* ensure cqe content is read after cqe ownership bit */ + rmb(); + + return cqe; +} + +int mlx5e_napi_poll(struct napi_struct *napi, int budget) +{ + struct mlx5e_channel *c = container_of(napi, struct mlx5e_channel, + napi); + bool busy = false; + int i; + + clear_bit(MLX5E_CHANNEL_NAPI_SCHED, &c->flags); + + for (i = 0; i < c->num_tc; i++) + busy |= mlx5e_poll_tx_cq(&c->sq[i].cq); + + busy |= mlx5e_poll_rx_cq(&c->rq.cq, budget); + + busy |= mlx5e_post_rx_wqes(c->rq.cq.sqrq); + + if (busy) + return budget; + + napi_complete(napi); + + /* avoid losing completion event during/after polling cqs */ + if (test_bit(MLX5E_CHANNEL_NAPI_SCHED, &c->flags)) { + napi_schedule(napi); + return 0; + } + + for (i = 0; i < c->num_tc; i++) + mlx5e_cq_arm(&c->sq[i].cq); + mlx5e_cq_arm(&c->rq.cq); + + return 0; +} + +void mlx5e_completion_event(struct mlx5_core_cq *mcq) +{ + struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq); + + set_bit(MLX5E_CQ_HAS_CQES, &cq->flags); + set_bit(MLX5E_CHANNEL_NAPI_SCHED, &cq->channel->flags); + barrier(); + napi_schedule(cq->napi); +} + +void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event) +{ + struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq); + struct mlx5e_channel *c = cq->channel; + struct mlx5e_priv *priv = c->priv; + struct net_device *netdev = priv->netdev; + + netdev_err(netdev, "%s: cqn=0x%.6x event=0x%.2x\n", + __func__, mcq->cqn, event); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 58800e4f3958..a40b96d4c662 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -339,15 +339,14 @@ static void init_eq_buf(struct mlx5_eq *eq) int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, int nent, u64 mask, const char *name, struct mlx5_uar *uar) { - struct mlx5_eq_table *table = &dev->priv.eq_table; + struct mlx5_priv *priv = &dev->priv; struct mlx5_create_eq_mbox_in *in; struct mlx5_create_eq_mbox_out out; int err; int inlen; eq->nent = roundup_pow_of_two(nent + MLX5_NUM_SPARE_EQE); - err = mlx5_buf_alloc(dev, eq->nent * MLX5_EQE_SIZE, 2 * PAGE_SIZE, - &eq->buf); + err = mlx5_buf_alloc(dev, eq->nent * MLX5_EQE_SIZE, &eq->buf); if (err) return err; @@ -378,14 +377,15 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, goto err_in; } - snprintf(eq->name, MLX5_MAX_EQ_NAME, "%s@pci:%s", + snprintf(priv->irq_info[vecidx].name, MLX5_MAX_IRQ_NAME, "%s@pci:%s", name, pci_name(dev->pdev)); + eq->eqn = out.eq_number; eq->irqn = vecidx; eq->dev = dev; eq->doorbell = uar->map + MLX5_EQ_DOORBEL_OFFSET; - err = request_irq(table->msix_arr[vecidx].vector, mlx5_msix_handler, 0, - eq->name, eq); + err = request_irq(priv->msix_arr[vecidx].vector, mlx5_msix_handler, 0, + priv->irq_info[vecidx].name, eq); if (err) goto err_eq; @@ -401,7 +401,7 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, return 0; err_irq: - free_irq(table->msix_arr[vecidx].vector, eq); + free_irq(priv->msix_arr[vecidx].vector, eq); err_eq: mlx5_cmd_destroy_eq(dev, eq->eqn); @@ -417,16 +417,15 @@ EXPORT_SYMBOL_GPL(mlx5_create_map_eq); int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq) { - struct mlx5_eq_table *table = &dev->priv.eq_table; int err; mlx5_debug_eq_remove(dev, eq); - free_irq(table->msix_arr[eq->irqn].vector, eq); + free_irq(dev->priv.msix_arr[eq->irqn].vector, eq); err = mlx5_cmd_destroy_eq(dev, eq->eqn); if (err) mlx5_core_warn(dev, "failed to destroy a previously created eq: eqn %d\n", eq->eqn); - synchronize_irq(table->msix_arr[eq->irqn].vector); + synchronize_irq(dev->priv.msix_arr[eq->irqn].vector); mlx5_buf_free(dev, &eq->buf); return err; @@ -456,7 +455,7 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev) u32 async_event_mask = MLX5_ASYNC_EVENT_MASK; int err; - if (dev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG) + if (MLX5_CAP_GEN(dev, pg)) async_event_mask |= (1ull << MLX5_EVENT_TYPE_PAGE_FAULT); err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD, @@ -479,7 +478,7 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev) err = mlx5_create_map_eq(dev, &table->pages_eq, MLX5_EQ_VEC_PAGES, - dev->caps.gen.max_vf + 1, + /* TODO: sriov max_vf + */ 1, 1 << MLX5_EVENT_TYPE_PAGE_REQUEST, "mlx5_pages_eq", &dev->priv.uuari.uars[0]); if (err) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c b/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c new file mode 100644 index 000000000000..ca90b9bc3b95 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/flow_table.c @@ -0,0 +1,422 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/export.h> +#include <linux/mlx5/driver.h> +#include <linux/mlx5/flow_table.h> +#include "mlx5_core.h" + +struct mlx5_ftg { + struct mlx5_flow_table_group g; + u32 id; + u32 start_ix; +}; + +struct mlx5_flow_table { + struct mlx5_core_dev *dev; + u8 level; + u8 type; + u32 id; + struct mutex mutex; /* sync bitmap alloc */ + u16 num_groups; + struct mlx5_ftg *group; + unsigned long *bitmap; + u32 size; +}; + +static int mlx5_set_flow_entry_cmd(struct mlx5_flow_table *ft, u32 group_ix, + u32 flow_index, void *flow_context) +{ + u32 out[MLX5_ST_SZ_DW(set_fte_out)]; + u32 *in; + void *in_flow_context; + int fcdls = + MLX5_GET(flow_context, flow_context, destination_list_size) * + MLX5_ST_SZ_BYTES(dest_format_struct); + int inlen = MLX5_ST_SZ_BYTES(set_fte_in) + fcdls; + int err; + + in = mlx5_vzalloc(inlen); + if (!in) { + mlx5_core_warn(ft->dev, "failed to allocate inbox\n"); + return -ENOMEM; + } + + MLX5_SET(set_fte_in, in, table_type, ft->type); + MLX5_SET(set_fte_in, in, table_id, ft->id); + MLX5_SET(set_fte_in, in, flow_index, flow_index); + MLX5_SET(set_fte_in, in, opcode, MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY); + + in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context); + memcpy(in_flow_context, flow_context, + MLX5_ST_SZ_BYTES(flow_context) + fcdls); + + MLX5_SET(flow_context, in_flow_context, group_id, + ft->group[group_ix].id); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out, + sizeof(out)); + kvfree(in); + + return err; +} + +static void mlx5_del_flow_entry_cmd(struct mlx5_flow_table *ft, u32 flow_index) +{ + u32 in[MLX5_ST_SZ_DW(delete_fte_in)]; + u32 out[MLX5_ST_SZ_DW(delete_fte_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + +#define MLX5_SET_DFTEI(p, x, v) MLX5_SET(delete_fte_in, p, x, v) + MLX5_SET_DFTEI(in, table_type, ft->type); + MLX5_SET_DFTEI(in, table_id, ft->id); + MLX5_SET_DFTEI(in, flow_index, flow_index); + MLX5_SET_DFTEI(in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY); + + mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); +} + +static void mlx5_destroy_flow_group_cmd(struct mlx5_flow_table *ft, int i) +{ + u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + +#define MLX5_SET_DFGI(p, x, v) MLX5_SET(destroy_flow_group_in, p, x, v) + MLX5_SET_DFGI(in, table_type, ft->type); + MLX5_SET_DFGI(in, table_id, ft->id); + MLX5_SET_DFGI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_GROUP); + MLX5_SET_DFGI(in, group_id, ft->group[i].id); + mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); +} + +static int mlx5_create_flow_group_cmd(struct mlx5_flow_table *ft, int i) +{ + u32 out[MLX5_ST_SZ_DW(create_flow_group_out)]; + u32 *in; + void *in_match_criteria; + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_table_group *g = &ft->group[i].g; + u32 start_ix = ft->group[i].start_ix; + u32 end_ix = start_ix + (1 << g->log_sz) - 1; + int err; + + in = mlx5_vzalloc(inlen); + if (!in) { + mlx5_core_warn(ft->dev, "failed to allocate inbox\n"); + return -ENOMEM; + } + in_match_criteria = MLX5_ADDR_OF(create_flow_group_in, in, + match_criteria); + + memset(out, 0, sizeof(out)); + +#define MLX5_SET_CFGI(p, x, v) MLX5_SET(create_flow_group_in, p, x, v) + MLX5_SET_CFGI(in, table_type, ft->type); + MLX5_SET_CFGI(in, table_id, ft->id); + MLX5_SET_CFGI(in, opcode, MLX5_CMD_OP_CREATE_FLOW_GROUP); + MLX5_SET_CFGI(in, start_flow_index, start_ix); + MLX5_SET_CFGI(in, end_flow_index, end_ix); + MLX5_SET_CFGI(in, match_criteria_enable, g->match_criteria_enable); + + memcpy(in_match_criteria, g->match_criteria, + MLX5_ST_SZ_BYTES(fte_match_param)); + + err = mlx5_cmd_exec_check_status(ft->dev, in, inlen, out, + sizeof(out)); + if (!err) + ft->group[i].id = MLX5_GET(create_flow_group_out, out, + group_id); + + kvfree(in); + + return err; +} + +static void mlx5_destroy_flow_table_groups(struct mlx5_flow_table *ft) +{ + int i; + + for (i = 0; i < ft->num_groups; i++) + mlx5_destroy_flow_group_cmd(ft, i); +} + +static int mlx5_create_flow_table_groups(struct mlx5_flow_table *ft) +{ + int err; + int i; + + for (i = 0; i < ft->num_groups; i++) { + err = mlx5_create_flow_group_cmd(ft, i); + if (err) + goto err_destroy_flow_table_groups; + } + + return 0; + +err_destroy_flow_table_groups: + for (i--; i >= 0; i--) + mlx5_destroy_flow_group_cmd(ft, i); + + return err; +} + +static int mlx5_create_flow_table_cmd(struct mlx5_flow_table *ft) +{ + u32 in[MLX5_ST_SZ_DW(create_flow_table_in)]; + u32 out[MLX5_ST_SZ_DW(create_flow_table_out)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(create_flow_table_in, in, table_type, ft->type); + MLX5_SET(create_flow_table_in, in, level, ft->level); + MLX5_SET(create_flow_table_in, in, log_size, order_base_2(ft->size)); + + MLX5_SET(create_flow_table_in, in, opcode, + MLX5_CMD_OP_CREATE_FLOW_TABLE); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, + sizeof(out)); + if (err) + return err; + + ft->id = MLX5_GET(create_flow_table_out, out, table_id); + + return 0; +} + +static void mlx5_destroy_flow_table_cmd(struct mlx5_flow_table *ft) +{ + u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + +#define MLX5_SET_DFTI(p, x, v) MLX5_SET(destroy_flow_table_in, p, x, v) + MLX5_SET_DFTI(in, table_type, ft->type); + MLX5_SET_DFTI(in, table_id, ft->id); + MLX5_SET_DFTI(in, opcode, MLX5_CMD_OP_DESTROY_FLOW_TABLE); + + mlx5_cmd_exec_check_status(ft->dev, in, sizeof(in), out, sizeof(out)); +} + +static int mlx5_find_group(struct mlx5_flow_table *ft, u8 match_criteria_enable, + u32 *match_criteria, int *group_ix) +{ + void *mc_outer = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers); + void *mc_misc = MLX5_ADDR_OF(fte_match_param, match_criteria, + misc_parameters); + void *mc_inner = MLX5_ADDR_OF(fte_match_param, match_criteria, + inner_headers); + int mc_outer_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4); + int mc_misc_sz = MLX5_ST_SZ_BYTES(fte_match_set_misc); + int mc_inner_sz = MLX5_ST_SZ_BYTES(fte_match_set_lyr_2_4); + int i; + + for (i = 0; i < ft->num_groups; i++) { + struct mlx5_flow_table_group *g = &ft->group[i].g; + void *gmc_outer = MLX5_ADDR_OF(fte_match_param, + g->match_criteria, + outer_headers); + void *gmc_misc = MLX5_ADDR_OF(fte_match_param, + g->match_criteria, + misc_parameters); + void *gmc_inner = MLX5_ADDR_OF(fte_match_param, + g->match_criteria, + inner_headers); + + if (g->match_criteria_enable != match_criteria_enable) + continue; + + if (match_criteria_enable & MLX5_MATCH_OUTER_HEADERS) + if (memcmp(mc_outer, gmc_outer, mc_outer_sz)) + continue; + + if (match_criteria_enable & MLX5_MATCH_MISC_PARAMETERS) + if (memcmp(mc_misc, gmc_misc, mc_misc_sz)) + continue; + + if (match_criteria_enable & MLX5_MATCH_INNER_HEADERS) + if (memcmp(mc_inner, gmc_inner, mc_inner_sz)) + continue; + + *group_ix = i; + return 0; + } + + return -EINVAL; +} + +static int alloc_flow_index(struct mlx5_flow_table *ft, int group_ix, u32 *ix) +{ + struct mlx5_ftg *g = &ft->group[group_ix]; + int err = 0; + + mutex_lock(&ft->mutex); + + *ix = find_next_zero_bit(ft->bitmap, ft->size, g->start_ix); + if (*ix >= (g->start_ix + (1 << g->g.log_sz))) + err = -ENOSPC; + else + __set_bit(*ix, ft->bitmap); + + mutex_unlock(&ft->mutex); + + return err; +} + +static void mlx5_free_flow_index(struct mlx5_flow_table *ft, u32 ix) +{ + __clear_bit(ix, ft->bitmap); +} + +int mlx5_add_flow_table_entry(void *flow_table, u8 match_criteria_enable, + void *match_criteria, void *flow_context, + u32 *flow_index) +{ + struct mlx5_flow_table *ft = flow_table; + int group_ix; + int err; + + err = mlx5_find_group(ft, match_criteria_enable, match_criteria, + &group_ix); + if (err) { + mlx5_core_warn(ft->dev, "mlx5_find_group failed\n"); + return err; + } + + err = alloc_flow_index(ft, group_ix, flow_index); + if (err) { + mlx5_core_warn(ft->dev, "alloc_flow_index failed\n"); + return err; + } + + return mlx5_set_flow_entry_cmd(ft, group_ix, *flow_index, flow_context); +} +EXPORT_SYMBOL(mlx5_add_flow_table_entry); + +void mlx5_del_flow_table_entry(void *flow_table, u32 flow_index) +{ + struct mlx5_flow_table *ft = flow_table; + + mlx5_del_flow_entry_cmd(ft, flow_index); + mlx5_free_flow_index(ft, flow_index); +} +EXPORT_SYMBOL(mlx5_del_flow_table_entry); + +void *mlx5_create_flow_table(struct mlx5_core_dev *dev, u8 level, u8 table_type, + u16 num_groups, + struct mlx5_flow_table_group *group) +{ + struct mlx5_flow_table *ft; + u32 start_ix = 0; + u32 ft_size = 0; + void *gr; + void *bm; + int err; + int i; + + for (i = 0; i < num_groups; i++) + ft_size += (1 << group[i].log_sz); + + ft = kzalloc(sizeof(*ft), GFP_KERNEL); + gr = kcalloc(num_groups, sizeof(struct mlx5_ftg), GFP_KERNEL); + bm = kcalloc(BITS_TO_LONGS(ft_size), sizeof(uintptr_t), GFP_KERNEL); + if (!ft || !gr || !bm) + goto err_free_ft; + + ft->group = gr; + ft->bitmap = bm; + ft->num_groups = num_groups; + ft->level = level; + ft->type = table_type; + ft->size = ft_size; + ft->dev = dev; + mutex_init(&ft->mutex); + + for (i = 0; i < ft->num_groups; i++) { + memcpy(&ft->group[i].g, &group[i], sizeof(*group)); + ft->group[i].start_ix = start_ix; + start_ix += 1 << group[i].log_sz; + } + + err = mlx5_create_flow_table_cmd(ft); + if (err) + goto err_free_ft; + + err = mlx5_create_flow_table_groups(ft); + if (err) + goto err_destroy_flow_table_cmd; + + return ft; + +err_destroy_flow_table_cmd: + mlx5_destroy_flow_table_cmd(ft); + +err_free_ft: + mlx5_core_warn(dev, "failed to alloc flow table\n"); + kfree(bm); + kfree(gr); + kfree(ft); + + return NULL; +} +EXPORT_SYMBOL(mlx5_create_flow_table); + +void mlx5_destroy_flow_table(void *flow_table) +{ + struct mlx5_flow_table *ft = flow_table; + + mlx5_destroy_flow_table_groups(ft); + mlx5_destroy_flow_table_cmd(ft); + kfree(ft->bitmap); + kfree(ft->group); + kfree(ft); +} +EXPORT_SYMBOL(mlx5_destroy_flow_table); + +u32 mlx5_get_flow_table_id(void *flow_table) +{ + struct mlx5_flow_table *ft = flow_table; + + return ft->id; +} +EXPORT_SYMBOL(mlx5_get_flow_table_id); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 4b4cda3bcc5f..801ccadd709a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -64,50 +64,74 @@ out_out: return err; } -int mlx5_cmd_query_hca_cap(struct mlx5_core_dev *dev, struct mlx5_caps *caps) +int mlx5_query_hca_caps(struct mlx5_core_dev *dev) { - return mlx5_core_get_caps(dev, caps, HCA_CAP_OPMOD_GET_CUR); -} - -int mlx5_query_odp_caps(struct mlx5_core_dev *dev, struct mlx5_odp_caps *caps) -{ - u8 in[MLX5_ST_SZ_BYTES(query_hca_cap_in)]; - int out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out); - void *out; int err; - if (!(dev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG)) - return -ENOTSUPP; + err = mlx5_core_get_caps(dev, MLX5_CAP_GENERAL, HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; - memset(in, 0, sizeof(in)); - out = kzalloc(out_sz, GFP_KERNEL); - if (!out) - return -ENOMEM; - MLX5_SET(query_hca_cap_in, in, opcode, MLX5_CMD_OP_QUERY_HCA_CAP); - MLX5_SET(query_hca_cap_in, in, op_mod, HCA_CAP_OPMOD_GET_ODP_CUR); - err = mlx5_cmd_exec(dev, in, sizeof(in), out, out_sz); + err = mlx5_core_get_caps(dev, MLX5_CAP_GENERAL, HCA_CAP_OPMOD_GET_MAX); if (err) - goto out; + return err; - err = mlx5_cmd_status_to_err_v2(out); - if (err) { - mlx5_core_warn(dev, "query cur hca ODP caps failed, %d\n", err); - goto out; + if (MLX5_CAP_GEN(dev, eth_net_offloads)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ETHERNET_OFFLOADS, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ETHERNET_OFFLOADS, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; } - memcpy(caps, MLX5_ADDR_OF(query_hca_cap_out, out, capability_struct), - sizeof(*caps)); + if (MLX5_CAP_GEN(dev, pg)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ODP, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ODP, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } - mlx5_core_dbg(dev, "on-demand paging capabilities:\nrc: %08x\nuc: %08x\nud: %08x\n", - be32_to_cpu(caps->per_transport_caps.rc_odp_caps), - be32_to_cpu(caps->per_transport_caps.uc_odp_caps), - be32_to_cpu(caps->per_transport_caps.ud_odp_caps)); + if (MLX5_CAP_GEN(dev, atomic)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ATOMIC, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ATOMIC, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } -out: - kfree(out); - return err; + if (MLX5_CAP_GEN(dev, roce)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_ROCE, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_ROCE, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } + + if (MLX5_CAP_GEN(dev, nic_flow_table)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_FLOW_TABLE, + HCA_CAP_OPMOD_GET_CUR); + if (err) + return err; + err = mlx5_core_get_caps(dev, MLX5_CAP_FLOW_TABLE, + HCA_CAP_OPMOD_GET_MAX); + if (err) + return err; + } + return 0; } -EXPORT_SYMBOL(mlx5_query_odp_caps); int mlx5_cmd_init_hca(struct mlx5_core_dev *dev) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 28425e5ea91f..2510fed3494d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -38,6 +38,7 @@ #include <linux/dma-mapping.h> #include <linux/slab.h> #include <linux/io-mapping.h> +#include <linux/interrupt.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/cq.h> #include <linux/mlx5/qp.h> @@ -47,10 +48,6 @@ #include <linux/mlx5/mlx5_ifc.h> #include "mlx5_core.h" -#define DRIVER_NAME "mlx5_core" -#define DRIVER_VERSION "3.0" -#define DRIVER_RELDATE "January 2015" - MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>"); MODULE_DESCRIPTION("Mellanox Connect-IB, ConnectX-4 core driver"); MODULE_LICENSE("Dual BSD/GPL"); @@ -208,24 +205,28 @@ static void release_bar(struct pci_dev *pdev) static int mlx5_enable_msix(struct mlx5_core_dev *dev) { - struct mlx5_eq_table *table = &dev->priv.eq_table; - int num_eqs = 1 << dev->caps.gen.log_max_eq; + struct mlx5_priv *priv = &dev->priv; + struct mlx5_eq_table *table = &priv->eq_table; + int num_eqs = 1 << MLX5_CAP_GEN(dev, log_max_eq); int nvec; int i; - nvec = dev->caps.gen.num_ports * num_online_cpus() + MLX5_EQ_VEC_COMP_BASE; + nvec = MLX5_CAP_GEN(dev, num_ports) * num_online_cpus() + + MLX5_EQ_VEC_COMP_BASE; nvec = min_t(int, nvec, num_eqs); if (nvec <= MLX5_EQ_VEC_COMP_BASE) return -ENOMEM; - table->msix_arr = kzalloc(nvec * sizeof(*table->msix_arr), GFP_KERNEL); - if (!table->msix_arr) - return -ENOMEM; + priv->msix_arr = kcalloc(nvec, sizeof(*priv->msix_arr), GFP_KERNEL); + + priv->irq_info = kcalloc(nvec, sizeof(*priv->irq_info), GFP_KERNEL); + if (!priv->msix_arr || !priv->irq_info) + goto err_free_msix; for (i = 0; i < nvec; i++) - table->msix_arr[i].entry = i; + priv->msix_arr[i].entry = i; - nvec = pci_enable_msix_range(dev->pdev, table->msix_arr, + nvec = pci_enable_msix_range(dev->pdev, priv->msix_arr, MLX5_EQ_VEC_COMP_BASE + 1, nvec); if (nvec < 0) return nvec; @@ -233,14 +234,20 @@ static int mlx5_enable_msix(struct mlx5_core_dev *dev) table->num_comp_vectors = nvec - MLX5_EQ_VEC_COMP_BASE; return 0; + +err_free_msix: + kfree(priv->irq_info); + kfree(priv->msix_arr); + return -ENOMEM; } static void mlx5_disable_msix(struct mlx5_core_dev *dev) { - struct mlx5_eq_table *table = &dev->priv.eq_table; + struct mlx5_priv *priv = &dev->priv; pci_disable_msix(dev->pdev); - kfree(table->msix_arr); + kfree(priv->irq_info); + kfree(priv->msix_arr); } struct mlx5_reg_host_endianess { @@ -277,98 +284,28 @@ static u16 to_fw_pkey_sz(u32 size) } } -/* selectively copy writable fields clearing any reserved area - */ -static void copy_rw_fields(void *to, struct mlx5_caps *from) +static u16 to_sw_pkey_sz(int pkey_sz) { - __be64 *flags_off = (__be64 *)MLX5_ADDR_OF(cmd_hca_cap, to, reserved_22); - u64 v64; - - MLX5_SET(cmd_hca_cap, to, log_max_qp, from->gen.log_max_qp); - MLX5_SET(cmd_hca_cap, to, log_max_ra_req_qp, from->gen.log_max_ra_req_qp); - MLX5_SET(cmd_hca_cap, to, log_max_ra_res_qp, from->gen.log_max_ra_res_qp); - MLX5_SET(cmd_hca_cap, to, pkey_table_size, from->gen.pkey_table_size); - MLX5_SET(cmd_hca_cap, to, pkey_table_size, to_fw_pkey_sz(from->gen.pkey_table_size)); - MLX5_SET(cmd_hca_cap, to, log_uar_page_sz, PAGE_SHIFT - 12); - v64 = from->gen.flags & MLX5_CAP_BITS_RW_MASK; - *flags_off = cpu_to_be64(v64); -} - -static u16 get_pkey_table_size(int pkey) -{ - if (pkey > MLX5_MAX_LOG_PKEY_TABLE) + if (pkey_sz > MLX5_MAX_LOG_PKEY_TABLE) return 0; - return MLX5_MIN_PKEY_TABLE_SIZE << pkey; + return MLX5_MIN_PKEY_TABLE_SIZE << pkey_sz; } -static void fw2drv_caps(struct mlx5_caps *caps, void *out) -{ - struct mlx5_general_caps *gen = &caps->gen; - - gen->max_srq_wqes = 1 << MLX5_GET_PR(cmd_hca_cap, out, log_max_srq_sz); - gen->max_wqes = 1 << MLX5_GET_PR(cmd_hca_cap, out, log_max_qp_sz); - gen->log_max_qp = MLX5_GET_PR(cmd_hca_cap, out, log_max_qp); - gen->log_max_strq = MLX5_GET_PR(cmd_hca_cap, out, log_max_strq_sz); - gen->log_max_srq = MLX5_GET_PR(cmd_hca_cap, out, log_max_srqs); - gen->max_cqes = 1 << MLX5_GET_PR(cmd_hca_cap, out, log_max_cq_sz); - gen->log_max_cq = MLX5_GET_PR(cmd_hca_cap, out, log_max_cq); - gen->max_eqes = 1 << MLX5_GET_PR(cmd_hca_cap, out, log_max_eq_sz); - gen->log_max_mkey = MLX5_GET_PR(cmd_hca_cap, out, log_max_mkey); - gen->log_max_eq = MLX5_GET_PR(cmd_hca_cap, out, log_max_eq); - gen->max_indirection = MLX5_GET_PR(cmd_hca_cap, out, max_indirection); - gen->log_max_mrw_sz = MLX5_GET_PR(cmd_hca_cap, out, log_max_mrw_sz); - gen->log_max_bsf_list_size = MLX5_GET_PR(cmd_hca_cap, out, log_max_bsf_list_size); - gen->log_max_klm_list_size = MLX5_GET_PR(cmd_hca_cap, out, log_max_klm_list_size); - gen->log_max_ra_req_dc = MLX5_GET_PR(cmd_hca_cap, out, log_max_ra_req_dc); - gen->log_max_ra_res_dc = MLX5_GET_PR(cmd_hca_cap, out, log_max_ra_res_dc); - gen->log_max_ra_req_qp = MLX5_GET_PR(cmd_hca_cap, out, log_max_ra_req_qp); - gen->log_max_ra_res_qp = MLX5_GET_PR(cmd_hca_cap, out, log_max_ra_res_qp); - gen->max_qp_counters = MLX5_GET_PR(cmd_hca_cap, out, max_qp_cnt); - gen->pkey_table_size = get_pkey_table_size(MLX5_GET_PR(cmd_hca_cap, out, pkey_table_size)); - gen->local_ca_ack_delay = MLX5_GET_PR(cmd_hca_cap, out, local_ca_ack_delay); - gen->num_ports = MLX5_GET_PR(cmd_hca_cap, out, num_ports); - gen->log_max_msg = MLX5_GET_PR(cmd_hca_cap, out, log_max_msg); - gen->stat_rate_support = MLX5_GET_PR(cmd_hca_cap, out, stat_rate_support); - gen->flags = be64_to_cpu(*(__be64 *)MLX5_ADDR_OF(cmd_hca_cap, out, reserved_22)); - pr_debug("flags = 0x%llx\n", gen->flags); - gen->uar_sz = MLX5_GET_PR(cmd_hca_cap, out, uar_sz); - gen->min_log_pg_sz = MLX5_GET_PR(cmd_hca_cap, out, log_pg_sz); - gen->bf_reg_size = MLX5_GET_PR(cmd_hca_cap, out, bf); - gen->bf_reg_size = 1 << MLX5_GET_PR(cmd_hca_cap, out, log_bf_reg_size); - gen->max_sq_desc_sz = MLX5_GET_PR(cmd_hca_cap, out, max_wqe_sz_sq); - gen->max_rq_desc_sz = MLX5_GET_PR(cmd_hca_cap, out, max_wqe_sz_rq); - gen->max_dc_sq_desc_sz = MLX5_GET_PR(cmd_hca_cap, out, max_wqe_sz_sq_dc); - gen->max_qp_mcg = MLX5_GET_PR(cmd_hca_cap, out, max_qp_mcg); - gen->log_max_pd = MLX5_GET_PR(cmd_hca_cap, out, log_max_pd); - gen->log_max_xrcd = MLX5_GET_PR(cmd_hca_cap, out, log_max_xrcd); - gen->log_uar_page_sz = MLX5_GET_PR(cmd_hca_cap, out, log_uar_page_sz); -} - -static const char *caps_opmod_str(u16 opmod) -{ - switch (opmod) { - case HCA_CAP_OPMOD_GET_MAX: - return "GET_MAX"; - case HCA_CAP_OPMOD_GET_CUR: - return "GET_CUR"; - default: - return "Invalid"; - } -} - -int mlx5_core_get_caps(struct mlx5_core_dev *dev, struct mlx5_caps *caps, - u16 opmod) +int mlx5_core_get_caps(struct mlx5_core_dev *dev, enum mlx5_cap_type cap_type, + enum mlx5_cap_mode cap_mode) { u8 in[MLX5_ST_SZ_BYTES(query_hca_cap_in)]; int out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out); - void *out; + void *out, *hca_caps; + u16 opmod = (cap_type << 1) | (cap_mode & 0x01); int err; memset(in, 0, sizeof(in)); out = kzalloc(out_sz, GFP_KERNEL); if (!out) return -ENOMEM; + MLX5_SET(query_hca_cap_in, in, opcode, MLX5_CMD_OP_QUERY_HCA_CAP); MLX5_SET(query_hca_cap_in, in, op_mod, opmod); err = mlx5_cmd_exec(dev, in, sizeof(in), out, out_sz); @@ -377,12 +314,30 @@ int mlx5_core_get_caps(struct mlx5_core_dev *dev, struct mlx5_caps *caps, err = mlx5_cmd_status_to_err_v2(out); if (err) { - mlx5_core_warn(dev, "query max hca cap failed, %d\n", err); + mlx5_core_warn(dev, + "QUERY_HCA_CAP : type(%x) opmode(%x) Failed(%d)\n", + cap_type, cap_mode, err); goto query_ex; } - mlx5_core_dbg(dev, "%s\n", caps_opmod_str(opmod)); - fw2drv_caps(caps, MLX5_ADDR_OF(query_hca_cap_out, out, capability_struct)); + hca_caps = MLX5_ADDR_OF(query_hca_cap_out, out, capability); + + switch (cap_mode) { + case HCA_CAP_OPMOD_GET_MAX: + memcpy(dev->hca_caps_max[cap_type], hca_caps, + MLX5_UN_SZ_BYTES(hca_cap_union)); + break; + case HCA_CAP_OPMOD_GET_CUR: + memcpy(dev->hca_caps_cur[cap_type], hca_caps, + MLX5_UN_SZ_BYTES(hca_cap_union)); + break; + default: + mlx5_core_warn(dev, + "Tried to query dev cap type(%x) with wrong opmode(%x)\n", + cap_type, cap_mode); + err = -EINVAL; + break; + } query_ex: kfree(out); return err; @@ -409,49 +364,45 @@ static int handle_hca_cap(struct mlx5_core_dev *dev) { void *set_ctx = NULL; struct mlx5_profile *prof = dev->profile; - struct mlx5_caps *cur_caps = NULL; - struct mlx5_caps *max_caps = NULL; int err = -ENOMEM; int set_sz = MLX5_ST_SZ_BYTES(set_hca_cap_in); + void *set_hca_cap; set_ctx = kzalloc(set_sz, GFP_KERNEL); if (!set_ctx) goto query_ex; - max_caps = kzalloc(sizeof(*max_caps), GFP_KERNEL); - if (!max_caps) - goto query_ex; - - cur_caps = kzalloc(sizeof(*cur_caps), GFP_KERNEL); - if (!cur_caps) - goto query_ex; - - err = mlx5_core_get_caps(dev, max_caps, HCA_CAP_OPMOD_GET_MAX); + err = mlx5_core_get_caps(dev, MLX5_CAP_GENERAL, HCA_CAP_OPMOD_GET_MAX); if (err) goto query_ex; - err = mlx5_core_get_caps(dev, cur_caps, HCA_CAP_OPMOD_GET_CUR); + err = mlx5_core_get_caps(dev, MLX5_CAP_GENERAL, HCA_CAP_OPMOD_GET_CUR); if (err) goto query_ex; + set_hca_cap = MLX5_ADDR_OF(set_hca_cap_in, set_ctx, + capability); + memcpy(set_hca_cap, dev->hca_caps_cur[MLX5_CAP_GENERAL], + MLX5_ST_SZ_BYTES(cmd_hca_cap)); + + mlx5_core_dbg(dev, "Current Pkey table size %d Setting new size %d\n", + to_sw_pkey_sz(MLX5_CAP_GEN(dev, pkey_table_size)), + 128); /* we limit the size of the pkey table to 128 entries for now */ - cur_caps->gen.pkey_table_size = 128; + MLX5_SET(cmd_hca_cap, set_hca_cap, pkey_table_size, + to_fw_pkey_sz(128)); if (prof->mask & MLX5_PROF_MASK_QP_SIZE) - cur_caps->gen.log_max_qp = prof->log_max_qp; + MLX5_SET(cmd_hca_cap, set_hca_cap, log_max_qp, + prof->log_max_qp); - /* disable checksum */ - cur_caps->gen.flags &= ~MLX5_DEV_CAP_FLAG_CMDIF_CSUM; + /* disable cmdif checksum */ + MLX5_SET(cmd_hca_cap, set_hca_cap, cmdif_checksum, 0); - copy_rw_fields(MLX5_ADDR_OF(set_hca_cap_in, set_ctx, hca_capability_struct), - cur_caps); err = set_caps(dev, set_ctx, set_sz); query_ex: - kfree(cur_caps); - kfree(max_caps); kfree(set_ctx); - return err; } @@ -507,6 +458,74 @@ static int mlx5_core_disable_hca(struct mlx5_core_dev *dev) return 0; } +static int mlx5_irq_set_affinity_hint(struct mlx5_core_dev *mdev, int i) +{ + struct mlx5_priv *priv = &mdev->priv; + struct msix_entry *msix = priv->msix_arr; + int irq = msix[i + MLX5_EQ_VEC_COMP_BASE].vector; + int numa_node = dev_to_node(&mdev->pdev->dev); + int err; + + if (!zalloc_cpumask_var(&priv->irq_info[i].mask, GFP_KERNEL)) { + mlx5_core_warn(mdev, "zalloc_cpumask_var failed"); + return -ENOMEM; + } + + cpumask_set_cpu(cpumask_local_spread(i, numa_node), + priv->irq_info[i].mask); + + err = irq_set_affinity_hint(irq, priv->irq_info[i].mask); + if (err) { + mlx5_core_warn(mdev, "irq_set_affinity_hint failed,irq 0x%.4x", + irq); + goto err_clear_mask; + } + + return 0; + +err_clear_mask: + free_cpumask_var(priv->irq_info[i].mask); + return err; +} + +static void mlx5_irq_clear_affinity_hint(struct mlx5_core_dev *mdev, int i) +{ + struct mlx5_priv *priv = &mdev->priv; + struct msix_entry *msix = priv->msix_arr; + int irq = msix[i + MLX5_EQ_VEC_COMP_BASE].vector; + + irq_set_affinity_hint(irq, NULL); + free_cpumask_var(priv->irq_info[i].mask); +} + +static int mlx5_irq_set_affinity_hints(struct mlx5_core_dev *mdev) +{ + int err; + int i; + + for (i = 0; i < mdev->priv.eq_table.num_comp_vectors; i++) { + err = mlx5_irq_set_affinity_hint(mdev, i); + if (err) + goto err_out; + } + + return 0; + +err_out: + for (i--; i >= 0; i--) + mlx5_irq_clear_affinity_hint(mdev, i); + + return err; +} + +static void mlx5_irq_clear_affinity_hints(struct mlx5_core_dev *mdev) +{ + int i; + + for (i = 0; i < mdev->priv.eq_table.num_comp_vectors; i++) + mlx5_irq_clear_affinity_hint(mdev, i); +} + int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn, int *irqn) { struct mlx5_eq_table *table = &dev->priv.eq_table; @@ -549,7 +568,7 @@ static void free_comp_eqs(struct mlx5_core_dev *dev) static int alloc_comp_eqs(struct mlx5_core_dev *dev) { struct mlx5_eq_table *table = &dev->priv.eq_table; - char name[MLX5_MAX_EQ_NAME]; + char name[MLX5_MAX_IRQ_NAME]; struct mlx5_eq *eq; int ncomp_vec; int nent; @@ -566,7 +585,7 @@ static int alloc_comp_eqs(struct mlx5_core_dev *dev) goto clean; } - snprintf(name, MLX5_MAX_EQ_NAME, "mlx5_comp%d", i); + snprintf(name, MLX5_MAX_IRQ_NAME, "mlx5_comp%d", i); err = mlx5_create_map_eq(dev, eq, i + MLX5_EQ_VEC_COMP_BASE, nent, 0, name, &dev->priv.uuari.uars[0]); @@ -588,6 +607,61 @@ clean: return err; } +#ifdef CONFIG_MLX5_CORE_EN +static int mlx5_core_set_issi(struct mlx5_core_dev *dev) +{ + u32 query_in[MLX5_ST_SZ_DW(query_issi_in)]; + u32 query_out[MLX5_ST_SZ_DW(query_issi_out)]; + u32 set_in[MLX5_ST_SZ_DW(set_issi_in)]; + u32 set_out[MLX5_ST_SZ_DW(set_issi_out)]; + int err; + u32 sup_issi; + + memset(query_in, 0, sizeof(query_in)); + memset(query_out, 0, sizeof(query_out)); + + MLX5_SET(query_issi_in, query_in, opcode, MLX5_CMD_OP_QUERY_ISSI); + + err = mlx5_cmd_exec_check_status(dev, query_in, sizeof(query_in), + query_out, sizeof(query_out)); + if (err) { + if (((struct mlx5_outbox_hdr *)query_out)->status == + MLX5_CMD_STAT_BAD_OP_ERR) { + pr_debug("Only ISSI 0 is supported\n"); + return 0; + } + + pr_err("failed to query ISSI\n"); + return err; + } + + sup_issi = MLX5_GET(query_issi_out, query_out, supported_issi_dw0); + + if (sup_issi & (1 << 1)) { + memset(set_in, 0, sizeof(set_in)); + memset(set_out, 0, sizeof(set_out)); + + MLX5_SET(set_issi_in, set_in, opcode, MLX5_CMD_OP_SET_ISSI); + MLX5_SET(set_issi_in, set_in, current_issi, 1); + + err = mlx5_cmd_exec_check_status(dev, set_in, sizeof(set_in), + set_out, sizeof(set_out)); + if (err) { + pr_err("failed to set ISSI=1\n"); + return err; + } + + dev->issi = 1; + + return 0; + } else if (sup_issi & (1 << 0)) { + return 0; + } + + return -ENOTSUPP; +} +#endif + static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) { struct mlx5_priv *priv = &dev->priv; @@ -650,6 +724,14 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) goto err_pagealloc_cleanup; } +#ifdef CONFIG_MLX5_CORE_EN + err = mlx5_core_set_issi(dev); + if (err) { + dev_err(&pdev->dev, "failed to set issi\n"); + goto err_disable_hca; + } +#endif + err = mlx5_satisfy_startup_pages(dev, 1); if (err) { dev_err(&pdev->dev, "failed to allocate boot pages\n"); @@ -688,7 +770,7 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) mlx5_start_health_poll(dev); - err = mlx5_cmd_query_hca_cap(dev, &dev->caps); + err = mlx5_query_hca_caps(dev); if (err) { dev_err(&pdev->dev, "query hca failed\n"); goto err_stop_poll; @@ -730,6 +812,12 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) goto err_stop_eqs; } + err = mlx5_irq_set_affinity_hints(dev); + if (err) { + dev_err(&pdev->dev, "Failed to alloc affinity hint cpumask\n"); + goto err_free_comp_eqs; + } + MLX5_INIT_DOORBELL_LOCK(&priv->cq_uar_lock); mlx5_init_cq_table(dev); @@ -739,6 +827,9 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) return 0; +err_free_comp_eqs: + free_comp_eqs(dev); + err_stop_eqs: mlx5_stop_eqs(dev); @@ -793,6 +884,7 @@ static void mlx5_dev_cleanup(struct mlx5_core_dev *dev) mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); mlx5_cleanup_cq_table(dev); + mlx5_irq_clear_affinity_hints(dev); free_comp_eqs(dev); mlx5_stop_eqs(dev); mlx5_free_uuars(dev, &priv->uuari); @@ -1048,6 +1140,10 @@ static int __init init(void) if (err) goto err_health; +#ifdef CONFIG_MLX5_CORE_EN + mlx5e_init(); +#endif + return 0; err_health: @@ -1060,6 +1156,9 @@ err_debug: static void __exit cleanup(void) { +#ifdef CONFIG_MLX5_CORE_EN + mlx5e_cleanup(); +#endif pci_unregister_driver(&mlx5_core_driver); mlx5_health_cleanup(); destroy_workqueue(mlx5_core_wq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mcg.c b/drivers/net/ethernet/mellanox/mlx5/core/mcg.c index d79fd85d1dd5..d5a0c2d61a18 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mcg.c @@ -91,7 +91,7 @@ int mlx5_core_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid, u32 qpn) memset(&in, 0, sizeof(in)); memset(&out, 0, sizeof(out)); - in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_DETACH_FROM_MCG); + in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_DETTACH_FROM_MCG); memcpy(in.gid, mgid, sizeof(*mgid)); in.qpn = cpu_to_be32(qpn); err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out)); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index a051b906afdf..6983c1047255 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -37,6 +37,10 @@ #include <linux/kernel.h> #include <linux/sched.h> +#define DRIVER_NAME "mlx5_core" +#define DRIVER_VERSION "3.0-1" +#define DRIVER_RELDATE "January 2015" + extern int mlx5_core_debug_mask; #define mlx5_core_dbg(dev, format, ...) \ @@ -65,11 +69,20 @@ enum { MLX5_CMD_TIME, /* print command execution time */ }; +static inline int mlx5_cmd_exec_check_status(struct mlx5_core_dev *dev, u32 *in, + int in_size, u32 *out, + int out_size) +{ + mlx5_cmd_exec(dev, in, in_size, out, out_size); + return mlx5_cmd_status_to_err((struct mlx5_outbox_hdr *)out); +} -int mlx5_cmd_query_hca_cap(struct mlx5_core_dev *dev, - struct mlx5_caps *caps); +int mlx5_query_hca_caps(struct mlx5_core_dev *dev); int mlx5_cmd_query_adapter(struct mlx5_core_dev *dev); int mlx5_cmd_init_hca(struct mlx5_core_dev *dev); int mlx5_cmd_teardown_hca(struct mlx5_core_dev *dev); +void mlx5e_init(void); +void mlx5e_cleanup(void); + #endif /* __MLX5_CORE_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 49e90f2612d8..7d3d0f9f328d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -102,3 +102,165 @@ int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps) return err; } EXPORT_SYMBOL_GPL(mlx5_set_port_caps); + +int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys, + int ptys_size, int proto_mask) +{ + u32 in[MLX5_ST_SZ_DW(ptys_reg)]; + int err; + + memset(in, 0, sizeof(in)); + MLX5_SET(ptys_reg, in, local_port, 1); + MLX5_SET(ptys_reg, in, proto_mask, proto_mask); + + err = mlx5_core_access_reg(dev, in, sizeof(in), ptys, + ptys_size, MLX5_REG_PTYS, 0, 0); + + return err; +} +EXPORT_SYMBOL_GPL(mlx5_query_port_ptys); + +int mlx5_query_port_proto_cap(struct mlx5_core_dev *dev, + u32 *proto_cap, int proto_mask) +{ + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + int err; + + err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask); + if (err) + return err; + + if (proto_mask == MLX5_PTYS_EN) + *proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); + else + *proto_cap = MLX5_GET(ptys_reg, out, ib_proto_capability); + + return 0; +} +EXPORT_SYMBOL_GPL(mlx5_query_port_proto_cap); + +int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev, + u32 *proto_admin, int proto_mask) +{ + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + int err; + + err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask); + if (err) + return err; + + if (proto_mask == MLX5_PTYS_EN) + *proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); + else + *proto_admin = MLX5_GET(ptys_reg, out, ib_proto_admin); + + return 0; +} +EXPORT_SYMBOL_GPL(mlx5_query_port_proto_admin); + +int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, + int proto_mask) +{ + u32 in[MLX5_ST_SZ_DW(ptys_reg)]; + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(ptys_reg, in, local_port, 1); + MLX5_SET(ptys_reg, in, proto_mask, proto_mask); + if (proto_mask == MLX5_PTYS_EN) + MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin); + else + MLX5_SET(ptys_reg, in, ib_proto_admin, proto_admin); + + err = mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PTYS, 0, 1); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_set_port_proto); + +int mlx5_set_port_status(struct mlx5_core_dev *dev, + enum mlx5_port_status status) +{ + u32 in[MLX5_ST_SZ_DW(paos_reg)]; + u32 out[MLX5_ST_SZ_DW(paos_reg)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(paos_reg, in, admin_status, status); + MLX5_SET(paos_reg, in, ase, 1); + + return mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PAOS, 0, 1); +} + +int mlx5_query_port_status(struct mlx5_core_dev *dev, u8 *status) +{ + u32 in[MLX5_ST_SZ_DW(paos_reg)]; + u32 out[MLX5_ST_SZ_DW(paos_reg)]; + int err; + + memset(in, 0, sizeof(in)); + + err = mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PAOS, 0, 0); + if (err) + return err; + + *status = MLX5_GET(paos_reg, out, oper_status); + return err; +} + +static int mlx5_query_port_mtu(struct mlx5_core_dev *dev, + int *admin_mtu, int *max_mtu, int *oper_mtu) +{ + u32 in[MLX5_ST_SZ_DW(pmtu_reg)]; + u32 out[MLX5_ST_SZ_DW(pmtu_reg)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(pmtu_reg, in, local_port, 1); + + err = mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PMTU, 0, 0); + if (err) + return err; + + if (max_mtu) + *max_mtu = MLX5_GET(pmtu_reg, out, max_mtu); + if (oper_mtu) + *oper_mtu = MLX5_GET(pmtu_reg, out, oper_mtu); + if (admin_mtu) + *admin_mtu = MLX5_GET(pmtu_reg, out, admin_mtu); + + return 0; +} + +int mlx5_set_port_mtu(struct mlx5_core_dev *dev, int mtu) +{ + u32 in[MLX5_ST_SZ_DW(pmtu_reg)]; + u32 out[MLX5_ST_SZ_DW(pmtu_reg)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(pmtu_reg, in, admin_mtu, mtu); + MLX5_SET(pmtu_reg, in, local_port, 1); + + return mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), + MLX5_REG_PMTU, 0, 1); +} +EXPORT_SYMBOL_GPL(mlx5_set_port_mtu); + +int mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, int *max_mtu) +{ + return mlx5_query_port_mtu(dev, NULL, max_mtu, NULL); +} +EXPORT_SYMBOL_GPL(mlx5_query_port_max_mtu); + +int mlx5_query_port_oper_mtu(struct mlx5_core_dev *dev, int *oper_mtu) +{ + return mlx5_query_port_mtu(dev, NULL, NULL, oper_mtu); +} +EXPORT_SYMBOL_GPL(mlx5_query_port_oper_mtu); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c new file mode 100644 index 000000000000..3c555d708af1 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c @@ -0,0 +1,169 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/mlx5/driver.h> +#include "mlx5_core.h" +#include "transobj.h" + +int mlx5_create_rq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *rqn) +{ + u32 out[MLX5_ST_SZ_DW(create_rq_out)]; + int err; + + MLX5_SET(create_rq_in, in, opcode, MLX5_CMD_OP_CREATE_RQ); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); + if (!err) + *rqn = MLX5_GET(create_rq_out, out, rqn); + + return err; +} + +int mlx5_modify_rq(struct mlx5_core_dev *dev, u32 rqn, u32 *in, int inlen) +{ + u32 out[MLX5_ST_SZ_DW(modify_rq_out)]; + + MLX5_SET(modify_rq_in, in, rqn, rqn); + MLX5_SET(modify_rq_in, in, opcode, MLX5_CMD_OP_MODIFY_RQ); + + memset(out, 0, sizeof(out)); + return mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); +} + +void mlx5_destroy_rq(struct mlx5_core_dev *dev, u32 rqn) +{ + u32 in[MLX5_ST_SZ_DW(destroy_rq_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_rq_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(destroy_rq_in, in, opcode, MLX5_CMD_OP_DESTROY_RQ); + MLX5_SET(destroy_rq_in, in, rqn, rqn); + + mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out)); +} + +int mlx5_create_sq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *sqn) +{ + u32 out[MLX5_ST_SZ_DW(create_sq_out)]; + int err; + + MLX5_SET(create_sq_in, in, opcode, MLX5_CMD_OP_CREATE_SQ); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); + if (!err) + *sqn = MLX5_GET(create_sq_out, out, sqn); + + return err; +} + +int mlx5_modify_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *in, int inlen) +{ + u32 out[MLX5_ST_SZ_DW(modify_sq_out)]; + + MLX5_SET(modify_sq_in, in, sqn, sqn); + MLX5_SET(modify_sq_in, in, opcode, MLX5_CMD_OP_MODIFY_SQ); + + memset(out, 0, sizeof(out)); + return mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); +} + +void mlx5_destroy_sq(struct mlx5_core_dev *dev, u32 sqn) +{ + u32 in[MLX5_ST_SZ_DW(destroy_sq_in)]; + u32 out[MLX5_ST_SZ_DW(destroy_sq_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(destroy_sq_in, in, opcode, MLX5_CMD_OP_DESTROY_SQ); + MLX5_SET(destroy_sq_in, in, sqn, sqn); + + mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out)); +} + +int mlx5_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *tirn) +{ + u32 out[MLX5_ST_SZ_DW(create_tir_out)]; + int err; + + MLX5_SET(create_tir_in, in, opcode, MLX5_CMD_OP_CREATE_TIR); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); + if (!err) + *tirn = MLX5_GET(create_tir_out, out, tirn); + + return err; +} + +void mlx5_destroy_tir(struct mlx5_core_dev *dev, u32 tirn) +{ + u32 in[MLX5_ST_SZ_DW(destroy_tir_out)]; + u32 out[MLX5_ST_SZ_DW(destroy_tir_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(destroy_tir_in, in, opcode, MLX5_CMD_OP_DESTROY_TIR); + MLX5_SET(destroy_tir_in, in, tirn, tirn); + + mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out)); +} + +int mlx5_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *tisn) +{ + u32 out[MLX5_ST_SZ_DW(create_tis_out)]; + int err; + + MLX5_SET(create_tis_in, in, opcode, MLX5_CMD_OP_CREATE_TIS); + + memset(out, 0, sizeof(out)); + err = mlx5_cmd_exec_check_status(dev, in, inlen, out, sizeof(out)); + if (!err) + *tisn = MLX5_GET(create_tis_out, out, tisn); + + return err; +} + +void mlx5_destroy_tis(struct mlx5_core_dev *dev, u32 tisn) +{ + u32 in[MLX5_ST_SZ_DW(destroy_tis_out)]; + u32 out[MLX5_ST_SZ_DW(destroy_tis_out)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(destroy_tis_in, in, opcode, MLX5_CMD_OP_DESTROY_TIS); + MLX5_SET(destroy_tis_in, in, tisn, tisn); + + mlx5_cmd_exec_check_status(dev, in, sizeof(in), out, sizeof(out)); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.h b/drivers/net/ethernet/mellanox/mlx5/core/transobj.h new file mode 100644 index 000000000000..1bc898cc4933 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/transobj.h @@ -0,0 +1,47 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef __TRANSOBJ_H__ +#define __TRANSOBJ_H__ + +int mlx5_create_rq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *rqn); +int mlx5_modify_rq(struct mlx5_core_dev *dev, u32 rqn, u32 *in, int inlen); +void mlx5_destroy_rq(struct mlx5_core_dev *dev, u32 rqn); +int mlx5_create_sq(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *sqn); +int mlx5_modify_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *in, int inlen); +void mlx5_destroy_sq(struct mlx5_core_dev *dev, u32 sqn); +int mlx5_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *tirn); +void mlx5_destroy_tir(struct mlx5_core_dev *dev, u32 tirn); +int mlx5_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *tisn); +void mlx5_destroy_tis(struct mlx5_core_dev *dev, u32 tisn); + +#endif /* __TRANSOBJ_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/uar.c b/drivers/net/ethernet/mellanox/mlx5/core/uar.c index 5a89bb1d678a..9ef85873ceea 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/uar.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/uar.c @@ -175,12 +175,13 @@ int mlx5_alloc_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari) for (i = 0; i < tot_uuars; i++) { bf = &uuari->bfs[i]; - bf->buf_size = dev->caps.gen.bf_reg_size / 2; + bf->buf_size = (1 << MLX5_CAP_GEN(dev, log_bf_reg_size)) / 2; bf->uar = &uuari->uars[i / MLX5_BF_REGS_PER_PAGE]; bf->regreg = uuari->uars[i / MLX5_BF_REGS_PER_PAGE].map; bf->reg = NULL; /* Add WC support */ - bf->offset = (i % MLX5_BF_REGS_PER_PAGE) * dev->caps.gen.bf_reg_size + - MLX5_BF_OFFSET; + bf->offset = (i % MLX5_BF_REGS_PER_PAGE) * + (1 << MLX5_CAP_GEN(dev, log_bf_reg_size)) + + MLX5_BF_OFFSET; bf->need_lock = need_uuar_lock(i); spin_lock_init(&bf->lock); spin_lock_init(&bf->lock32); @@ -223,3 +224,40 @@ int mlx5_free_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari) return 0; } + +int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar) +{ + phys_addr_t pfn; + phys_addr_t uar_bar_start; + int err; + + err = mlx5_cmd_alloc_uar(mdev, &uar->index); + if (err) { + mlx5_core_warn(mdev, "mlx5_cmd_alloc_uar() failed, %d\n", err); + return err; + } + + uar_bar_start = pci_resource_start(mdev->pdev, 0); + pfn = (uar_bar_start >> PAGE_SHIFT) + uar->index; + uar->map = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE); + if (!uar->map) { + mlx5_core_warn(mdev, "ioremap() failed, %d\n", err); + err = -ENOMEM; + goto err_free_uar; + } + + return 0; + +err_free_uar: + mlx5_cmd_free_uar(mdev, uar->index); + + return err; +} +EXPORT_SYMBOL(mlx5_alloc_map_uar); + +void mlx5_unmap_free_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar) +{ + iounmap(uar->map); + mlx5_cmd_free_uar(mdev, uar->index); +} +EXPORT_SYMBOL(mlx5_unmap_free_uar); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c new file mode 100644 index 000000000000..ba374b9a6c87 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -0,0 +1,84 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/export.h> +#include <linux/etherdevice.h> +#include <linux/mlx5/driver.h> +#include "vport.h" +#include "mlx5_core.h" + +u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod) +{ + u32 in[MLX5_ST_SZ_DW(query_vport_state_in)]; + u32 out[MLX5_ST_SZ_DW(query_vport_state_out)]; + int err; + + memset(in, 0, sizeof(in)); + + MLX5_SET(query_vport_state_in, in, opcode, + MLX5_CMD_OP_QUERY_VPORT_STATE); + MLX5_SET(query_vport_state_in, in, op_mod, opmod); + + err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, + sizeof(out)); + if (err) + mlx5_core_warn(mdev, "MLX5_CMD_OP_QUERY_VPORT_STATE failed\n"); + + return MLX5_GET(query_vport_state_out, out, state); +} + +void mlx5_query_vport_mac_address(struct mlx5_core_dev *mdev, u8 *addr) +{ + u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)]; + u32 *out; + int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out); + u8 *out_addr; + + out = mlx5_vzalloc(outlen); + if (!out) + return; + + out_addr = MLX5_ADDR_OF(query_nic_vport_context_out, out, + nic_vport_context.permanent_address); + + memset(in, 0, sizeof(in)); + + MLX5_SET(query_nic_vport_context_in, in, opcode, + MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); + + memset(out, 0, outlen); + mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out, outlen); + + ether_addr_copy(addr, &out_addr[2]); + + kvfree(out); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.h b/drivers/net/ethernet/mellanox/mlx5/core/vport.h new file mode 100644 index 000000000000..c05ca2c3419d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.h @@ -0,0 +1,41 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef __MLX5_VPORT_H__ +#define __MLX5_VPORT_H__ + +#include <linux/mlx5/driver.h> + +u8 mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod); +void mlx5_query_vport_mac_address(struct mlx5_core_dev *mdev, u8 *addr); + +#endif /* __MLX5_VPORT_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c new file mode 100644 index 000000000000..8388411582cf --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c @@ -0,0 +1,183 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/mlx5/driver.h> +#include "wq.h" +#include "mlx5_core.h" + +u32 mlx5_wq_cyc_get_size(struct mlx5_wq_cyc *wq) +{ + return (u32)wq->sz_m1 + 1; +} + +u32 mlx5_cqwq_get_size(struct mlx5_cqwq *wq) +{ + return wq->sz_m1 + 1; +} + +u32 mlx5_wq_ll_get_size(struct mlx5_wq_ll *wq) +{ + return (u32)wq->sz_m1 + 1; +} + +static u32 mlx5_wq_cyc_get_byte_size(struct mlx5_wq_cyc *wq) +{ + return mlx5_wq_cyc_get_size(wq) << wq->log_stride; +} + +static u32 mlx5_cqwq_get_byte_size(struct mlx5_cqwq *wq) +{ + return mlx5_cqwq_get_size(wq) << wq->log_stride; +} + +static u32 mlx5_wq_ll_get_byte_size(struct mlx5_wq_ll *wq) +{ + return mlx5_wq_ll_get_size(wq) << wq->log_stride; +} + +int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *wqc, struct mlx5_wq_cyc *wq, + struct mlx5_wq_ctrl *wq_ctrl) +{ + int err; + + wq->log_stride = MLX5_GET(wq, wqc, log_wq_stride); + wq->sz_m1 = (1 << MLX5_GET(wq, wqc, log_wq_sz)) - 1; + + err = mlx5_db_alloc(mdev, &wq_ctrl->db); + if (err) { + mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + return err; + } + + err = mlx5_buf_alloc(mdev, mlx5_wq_cyc_get_byte_size(wq), &wq_ctrl->buf); + if (err) { + mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + goto err_db_free; + } + + wq->buf = wq_ctrl->buf.direct.buf; + wq->db = wq_ctrl->db.db; + + wq_ctrl->mdev = mdev; + + return 0; + +err_db_free: + mlx5_db_free(mdev, &wq_ctrl->db); + + return err; +} + +int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *cqc, struct mlx5_cqwq *wq, + struct mlx5_wq_ctrl *wq_ctrl) +{ + int err; + + wq->log_stride = 6 + MLX5_GET(cqc, cqc, cqe_sz); + wq->log_sz = MLX5_GET(cqc, cqc, log_cq_size); + wq->sz_m1 = (1 << wq->log_sz) - 1; + + err = mlx5_db_alloc(mdev, &wq_ctrl->db); + if (err) { + mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + return err; + } + + err = mlx5_buf_alloc(mdev, mlx5_cqwq_get_byte_size(wq), &wq_ctrl->buf); + if (err) { + mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + goto err_db_free; + } + + wq->buf = wq_ctrl->buf.direct.buf; + wq->db = wq_ctrl->db.db; + + wq_ctrl->mdev = mdev; + + return 0; + +err_db_free: + mlx5_db_free(mdev, &wq_ctrl->db); + + return err; +} + +int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *wqc, struct mlx5_wq_ll *wq, + struct mlx5_wq_ctrl *wq_ctrl) +{ + struct mlx5_wqe_srq_next_seg *next_seg; + int err; + int i; + + wq->log_stride = MLX5_GET(wq, wqc, log_wq_stride); + wq->sz_m1 = (1 << MLX5_GET(wq, wqc, log_wq_sz)) - 1; + + err = mlx5_db_alloc(mdev, &wq_ctrl->db); + if (err) { + mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + return err; + } + + err = mlx5_buf_alloc(mdev, mlx5_wq_ll_get_byte_size(wq), &wq_ctrl->buf); + if (err) { + mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + goto err_db_free; + } + + wq->buf = wq_ctrl->buf.direct.buf; + wq->db = wq_ctrl->db.db; + + for (i = 0; i < wq->sz_m1; i++) { + next_seg = mlx5_wq_ll_get_wqe(wq, i); + next_seg->next_wqe_index = cpu_to_be16(i + 1); + } + next_seg = mlx5_wq_ll_get_wqe(wq, i); + wq->tail_next = &next_seg->next_wqe_index; + + wq_ctrl->mdev = mdev; + + return 0; + +err_db_free: + mlx5_db_free(mdev, &wq_ctrl->db); + + return err; +} + +void mlx5_wq_destroy(struct mlx5_wq_ctrl *wq_ctrl) +{ + mlx5_buf_free(wq_ctrl->mdev, &wq_ctrl->buf); + mlx5_db_free(wq_ctrl->mdev, &wq_ctrl->db); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.h b/drivers/net/ethernet/mellanox/mlx5/core/wq.h new file mode 100644 index 000000000000..e0ddd69fb429 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.h @@ -0,0 +1,171 @@ +/* + * Copyright (c) 2013-2015, Mellanox Technologies, Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, 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. + * + * - 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. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef __MLX5_WQ_H__ +#define __MLX5_WQ_H__ + +#include <linux/mlx5/mlx5_ifc.h> + +struct mlx5_wq_param { + int linear; + int numa; +}; + +struct mlx5_wq_ctrl { + struct mlx5_core_dev *mdev; + struct mlx5_buf buf; + struct mlx5_db db; +}; + +struct mlx5_wq_cyc { + void *buf; + __be32 *db; + u16 sz_m1; + u8 log_stride; +}; + +struct mlx5_cqwq { + void *buf; + __be32 *db; + u32 sz_m1; + u32 cc; /* consumer counter */ + u8 log_sz; + u8 log_stride; +}; + +struct mlx5_wq_ll { + void *buf; + __be32 *db; + __be16 *tail_next; + u16 sz_m1; + u16 head; + u16 wqe_ctr; + u16 cur_sz; + u8 log_stride; +}; + +int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *wqc, struct mlx5_wq_cyc *wq, + struct mlx5_wq_ctrl *wq_ctrl); +u32 mlx5_wq_cyc_get_size(struct mlx5_wq_cyc *wq); + +int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *cqc, struct mlx5_cqwq *wq, + struct mlx5_wq_ctrl *wq_ctrl); +u32 mlx5_cqwq_get_size(struct mlx5_cqwq *wq); + +int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, + void *wqc, struct mlx5_wq_ll *wq, + struct mlx5_wq_ctrl *wq_ctrl); +u32 mlx5_wq_ll_get_size(struct mlx5_wq_ll *wq); + +void mlx5_wq_destroy(struct mlx5_wq_ctrl *wq_ctrl); + +static inline u16 mlx5_wq_cyc_ctr2ix(struct mlx5_wq_cyc *wq, u16 ctr) +{ + return ctr & wq->sz_m1; +} + +static inline void *mlx5_wq_cyc_get_wqe(struct mlx5_wq_cyc *wq, u16 ix) +{ + return wq->buf + (ix << wq->log_stride); +} + +static inline int mlx5_wq_cyc_cc_bigger(u16 cc1, u16 cc2) +{ + int equal = (cc1 == cc2); + int smaller = 0x8000 & (cc1 - cc2); + + return !equal && !smaller; +} + +static inline u32 mlx5_cqwq_get_ci(struct mlx5_cqwq *wq) +{ + return wq->cc & wq->sz_m1; +} + +static inline void *mlx5_cqwq_get_wqe(struct mlx5_cqwq *wq, u32 ix) +{ + return wq->buf + (ix << wq->log_stride); +} + +static inline u32 mlx5_cqwq_get_wrap_cnt(struct mlx5_cqwq *wq) +{ + return wq->cc >> wq->log_sz; +} + +static inline void mlx5_cqwq_pop(struct mlx5_cqwq *wq) +{ + wq->cc++; +} + +static inline void mlx5_cqwq_update_db_record(struct mlx5_cqwq *wq) +{ + *wq->db = cpu_to_be32(wq->cc & 0xffffff); +} + +static inline int mlx5_wq_ll_is_full(struct mlx5_wq_ll *wq) +{ + return wq->cur_sz == wq->sz_m1; +} + +static inline int mlx5_wq_ll_is_empty(struct mlx5_wq_ll *wq) +{ + return !wq->cur_sz; +} + +static inline void *mlx5_wq_ll_get_wqe(struct mlx5_wq_ll *wq, u16 ix) +{ + return wq->buf + (ix << wq->log_stride); +} + +static inline void mlx5_wq_ll_push(struct mlx5_wq_ll *wq, u16 head_next) +{ + wq->head = head_next; + wq->wqe_ctr++; + wq->cur_sz++; +} + +static inline void mlx5_wq_ll_pop(struct mlx5_wq_ll *wq, __be16 ix, + __be16 *next_tail_next) +{ + *wq->tail_next = ix; + wq->tail_next = next_tail_next; + wq->cur_sz--; +} + +static inline void mlx5_wq_ll_update_db_record(struct mlx5_wq_ll *wq) +{ + *wq->db = cpu_to_be32(wq->wqe_ctr); +} + +#endif /* __MLX5_WQ_H__ */ diff --git a/drivers/net/ethernet/moxa/moxart_ether.c b/drivers/net/ethernet/moxa/moxart_ether.c index 81d0f1c86d6d..becbb5f1f5a7 100644 --- a/drivers/net/ethernet/moxa/moxart_ether.c +++ b/drivers/net/ethernet/moxa/moxart_ether.c @@ -244,7 +244,6 @@ static int moxart_rx_poll(struct napi_struct *napi, int budget) napi_gro_receive(&priv->napi, skb); rx++; - ndev->last_rx = jiffies; priv->stats.rx_packets++; priv->stats.rx_bytes += len; if (desc0 & RX_DESC0_MULTICAST) diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c index 1e0f72b65459..c28111749e1f 100644 --- a/drivers/net/ethernet/neterion/s2io.c +++ b/drivers/net/ethernet/neterion/s2io.c @@ -5308,7 +5308,8 @@ static int do_s2io_prog_unicast(struct net_device *dev, u8 *addr) /** * s2io_ethtool_sset - Sets different link parameters. - * @sp : private member of the device structure, which is a pointer to the * s2io_nic structure. + * @sp : private member of the device structure, which is a pointer to the + * s2io_nic structure. * @info: pointer to the structure with parameters given by ethtool to set * link information. * Description: @@ -5793,7 +5794,8 @@ static void s2io_vpd_read(struct s2io_nic *nic) /** * s2io_ethtool_geeprom - reads the value stored in the Eeprom. - * @sp : private member of the device structure, which is a pointer to the * s2io_nic structure. + * @sp : private member of the device structure, which is a pointer to the + * s2io_nic structure. * @eeprom : pointer to the user level structure provided by ethtool, * containing all relevant information. * @data_buf : user defined value to be written into Eeprom. diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index f221126a5c4e..055f3763e577 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -1326,9 +1326,6 @@ struct qlcnic_eswitch { }; -/* Return codes for Error handling */ -#define QL_STATUS_INVALID_PARAM -1 - #define MAX_BW 100 /* % of link speed */ #define MIN_BW 1 /* % of link speed */ #define MAX_VLAN_ID 4095 diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index 367f3976df56..2f6cc423ab1d 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -1031,7 +1031,7 @@ int qlcnic_init_pci_info(struct qlcnic_adapter *adapter) pfn = pci_info[i].id; if (pfn >= ahw->max_vnic_func) { - ret = QL_STATUS_INVALID_PARAM; + ret = -EINVAL; dev_err(&adapter->pdev->dev, "%s: Invalid function 0x%x, max 0x%x\n", __func__, pfn, ahw->max_vnic_func); goto err_eswitch; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c index 59a721fba018..05c28f2c6df7 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sysfs.c @@ -24,8 +24,6 @@ #include <linux/hwmon-sysfs.h> #endif -#define QLC_STATUS_UNSUPPORTED_CMD -2 - int qlcnicvf_config_bridged_mode(struct qlcnic_adapter *adapter, u32 enable) { return -EOPNOTSUPP; @@ -166,7 +164,7 @@ static int qlcnic_82xx_store_beacon(struct qlcnic_adapter *adapter, u8 b_state, b_rate; if (len != sizeof(u16)) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; memcpy(&beacon, buf, sizeof(u16)); err = qlcnic_validate_beacon(adapter, beacon, &b_state, &b_rate); @@ -383,17 +381,17 @@ static int validate_pm_config(struct qlcnic_adapter *adapter, dest_pci_func = pm_cfg[i].dest_npar; src_index = qlcnic_is_valid_nic_func(adapter, src_pci_func); if (src_index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; dest_index = qlcnic_is_valid_nic_func(adapter, dest_pci_func); if (dest_index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; s_esw_id = adapter->npars[src_index].phy_port; d_esw_id = adapter->npars[dest_index].phy_port; if (s_esw_id != d_esw_id) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; } return 0; @@ -414,7 +412,7 @@ static ssize_t qlcnic_sysfs_write_pm_config(struct file *filp, count = size / sizeof(struct qlcnic_pm_func_cfg); rem = size % sizeof(struct qlcnic_pm_func_cfg); if (rem) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32)); pm_cfg = (struct qlcnic_pm_func_cfg *)buf; @@ -427,7 +425,7 @@ static ssize_t qlcnic_sysfs_write_pm_config(struct file *filp, action = !!pm_cfg[i].action; index = qlcnic_is_valid_nic_func(adapter, pci_func); if (index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; id = adapter->npars[index].phy_port; ret = qlcnic_config_port_mirroring(adapter, id, @@ -440,7 +438,7 @@ static ssize_t qlcnic_sysfs_write_pm_config(struct file *filp, pci_func = pm_cfg[i].pci_func; index = qlcnic_is_valid_nic_func(adapter, pci_func); if (index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; id = adapter->npars[index].phy_port; adapter->npars[index].enable_pm = !!pm_cfg[i].action; adapter->npars[index].dest_npar = id; @@ -499,11 +497,11 @@ static int validate_esw_config(struct qlcnic_adapter *adapter, for (i = 0; i < count; i++) { pci_func = esw_cfg[i].pci_func; if (pci_func >= ahw->max_vnic_func) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (adapter->ahw->op_mode == QLCNIC_MGMT_FUNC) if (qlcnic_is_valid_nic_func(adapter, pci_func) < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; switch (esw_cfg[i].op_mode) { case QLCNIC_PORT_DEFAULTS: @@ -517,25 +515,25 @@ static int validate_esw_config(struct qlcnic_adapter *adapter, if (ret != QLCNIC_NON_PRIV_FUNC) { if (esw_cfg[i].mac_anti_spoof != 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (esw_cfg[i].mac_override != 1) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (esw_cfg[i].promisc_mode != 1) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; } break; case QLCNIC_ADD_VLAN: if (!IS_VALID_VLAN(esw_cfg[i].vlan_id)) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (!esw_cfg[i].op_type) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; break; case QLCNIC_DEL_VLAN: if (!esw_cfg[i].op_type) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; break; default: - return QL_STATUS_INVALID_PARAM; + return -EINVAL; } } @@ -559,7 +557,7 @@ static ssize_t qlcnic_sysfs_write_esw_config(struct file *file, count = size / sizeof(struct qlcnic_esw_func_cfg); rem = size % sizeof(struct qlcnic_esw_func_cfg); if (rem) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32)); esw_cfg = (struct qlcnic_esw_func_cfg *)buf; @@ -570,7 +568,7 @@ static ssize_t qlcnic_sysfs_write_esw_config(struct file *file, for (i = 0; i < count; i++) { if (adapter->ahw->op_mode == QLCNIC_MGMT_FUNC) if (qlcnic_config_switch_port(adapter, &esw_cfg[i])) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (adapter->ahw->pci_func != esw_cfg[i].pci_func) continue; @@ -604,7 +602,7 @@ static ssize_t qlcnic_sysfs_write_esw_config(struct file *file, pci_func = esw_cfg[i].pci_func; index = qlcnic_is_valid_nic_func(adapter, pci_func); if (index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; npar = &adapter->npars[index]; switch (esw_cfg[i].op_mode) { case QLCNIC_PORT_DEFAULTS: @@ -654,7 +652,7 @@ static ssize_t qlcnic_sysfs_read_esw_config(struct file *file, esw_cfg[pci_func].pci_func = pci_func; if (qlcnic_get_eswitch_port_config(adapter, &esw_cfg[pci_func])) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; } qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32)); return size; @@ -669,11 +667,11 @@ static int validate_npar_config(struct qlcnic_adapter *adapter, for (i = 0; i < count; i++) { pci_func = np_cfg[i].pci_func; if (qlcnic_is_valid_nic_func(adapter, pci_func) < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (!IS_VALID_BW(np_cfg[i].min_bw) || !IS_VALID_BW(np_cfg[i].max_bw)) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; } return 0; } @@ -694,7 +692,7 @@ static ssize_t qlcnic_sysfs_write_npar_config(struct file *file, count = size / sizeof(struct qlcnic_npar_func_cfg); rem = size % sizeof(struct qlcnic_npar_func_cfg); if (rem) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32)); np_cfg = (struct qlcnic_npar_func_cfg *)buf; @@ -717,7 +715,7 @@ static ssize_t qlcnic_sysfs_write_npar_config(struct file *file, return ret; index = qlcnic_is_valid_nic_func(adapter, pci_func); if (index < 0) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; adapter->npars[index].min_bw = nic_info.min_tx_bw; adapter->npars[index].max_bw = nic_info.max_tx_bw; } @@ -784,13 +782,13 @@ static ssize_t qlcnic_sysfs_get_port_stats(struct file *file, int ret; if (qlcnic_83xx_check(adapter)) - return QLC_STATUS_UNSUPPORTED_CMD; + return -EOPNOTSUPP; if (size != sizeof(struct qlcnic_esw_statistics)) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (offset >= adapter->ahw->max_vnic_func) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; memset(&port_stats, 0, size); ret = qlcnic_get_port_stats(adapter, offset, QLCNIC_QUERY_RX_COUNTER, @@ -819,13 +817,13 @@ static ssize_t qlcnic_sysfs_get_esw_stats(struct file *file, int ret; if (qlcnic_83xx_check(adapter)) - return QLC_STATUS_UNSUPPORTED_CMD; + return -EOPNOTSUPP; if (size != sizeof(struct qlcnic_esw_statistics)) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; if (offset >= QLCNIC_NIU_MAX_XG_PORTS) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; memset(&esw_stats, 0, size); ret = qlcnic_get_eswitch_stats(adapter, offset, QLCNIC_QUERY_RX_COUNTER, @@ -853,10 +851,10 @@ static ssize_t qlcnic_sysfs_clear_esw_stats(struct file *file, int ret; if (qlcnic_83xx_check(adapter)) - return QLC_STATUS_UNSUPPORTED_CMD; + return -EOPNOTSUPP; if (offset >= QLCNIC_NIU_MAX_XG_PORTS) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; ret = qlcnic_clear_esw_stats(adapter, QLCNIC_STATS_ESWITCH, offset, QLCNIC_QUERY_RX_COUNTER); @@ -883,10 +881,10 @@ static ssize_t qlcnic_sysfs_clear_port_stats(struct file *file, int ret; if (qlcnic_83xx_check(adapter)) - return QLC_STATUS_UNSUPPORTED_CMD; + return -EOPNOTSUPP; if (offset >= adapter->ahw->max_vnic_func) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; ret = qlcnic_clear_esw_stats(adapter, QLCNIC_STATS_PORT, offset, QLCNIC_QUERY_RX_COUNTER); @@ -953,9 +951,7 @@ static ssize_t qlcnic_83xx_sysfs_flash_read_handler(struct file *filp, struct qlcnic_adapter *adapter = dev_get_drvdata(dev); if (!size) - return QL_STATUS_INVALID_PARAM; - if (!buf) - return QL_STATUS_INVALID_PARAM; + return -EINVAL; count = size / sizeof(u32); @@ -1132,9 +1128,6 @@ static ssize_t qlcnic_83xx_sysfs_flash_write_handler(struct file *filp, struct device *dev = container_of(kobj, struct device, kobj); struct qlcnic_adapter *adapter = dev_get_drvdata(dev); - if (!buf) - return QL_STATUS_INVALID_PARAM; - ret = kstrtoul(buf, 16, &data); switch (data) { diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c index 25800a1dedcb..02b7115b6aaa 100644 --- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c +++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c @@ -3871,9 +3871,6 @@ static int ql_adapter_reset(struct ql_adapter *qdev) return status; } - end_jiffies = jiffies + - max((unsigned long)1, usecs_to_jiffies(30)); - /* Check if bit is set then skip the mailbox command and * clear the bit, else we are in normal reset process. */ @@ -3888,6 +3885,7 @@ static int ql_adapter_reset(struct ql_adapter *qdev) ql_write32(qdev, RST_FO, (RST_FO_FR << 16) | RST_FO_FR); + end_jiffies = jiffies + usecs_to_jiffies(30); do { value = ql_read32(qdev, RST_FO); if ((value & RST_FO_FR) == 0) diff --git a/drivers/net/ethernet/qualcomm/qca_spi.c b/drivers/net/ethernet/qualcomm/qca_spi.c index 6af028d5f9bc..2f87909f5186 100644 --- a/drivers/net/ethernet/qualcomm/qca_spi.c +++ b/drivers/net/ethernet/qualcomm/qca_spi.c @@ -839,7 +839,7 @@ static const struct of_device_id qca_spi_of_match[] = { MODULE_DEVICE_TABLE(of, qca_spi_of_match); static int -qca_spi_probe(struct spi_device *spi_device) +qca_spi_probe(struct spi_device *spi) { struct qcaspi *qca = NULL; struct net_device *qcaspi_devs = NULL; @@ -847,52 +847,52 @@ qca_spi_probe(struct spi_device *spi_device) u16 signature; const char *mac; - if (!spi_device->dev.of_node) { - dev_err(&spi_device->dev, "Missing device tree\n"); + if (!spi->dev.of_node) { + dev_err(&spi->dev, "Missing device tree\n"); return -EINVAL; } - legacy_mode = of_property_read_bool(spi_device->dev.of_node, + legacy_mode = of_property_read_bool(spi->dev.of_node, "qca,legacy-mode"); if (qcaspi_clkspeed == 0) { - if (spi_device->max_speed_hz) - qcaspi_clkspeed = spi_device->max_speed_hz; + if (spi->max_speed_hz) + qcaspi_clkspeed = spi->max_speed_hz; else qcaspi_clkspeed = QCASPI_CLK_SPEED; } if ((qcaspi_clkspeed < QCASPI_CLK_SPEED_MIN) || (qcaspi_clkspeed > QCASPI_CLK_SPEED_MAX)) { - dev_info(&spi_device->dev, "Invalid clkspeed: %d\n", + dev_info(&spi->dev, "Invalid clkspeed: %d\n", qcaspi_clkspeed); return -EINVAL; } if ((qcaspi_burst_len < QCASPI_BURST_LEN_MIN) || (qcaspi_burst_len > QCASPI_BURST_LEN_MAX)) { - dev_info(&spi_device->dev, "Invalid burst len: %d\n", + dev_info(&spi->dev, "Invalid burst len: %d\n", qcaspi_burst_len); return -EINVAL; } if ((qcaspi_pluggable < QCASPI_PLUGGABLE_MIN) || (qcaspi_pluggable > QCASPI_PLUGGABLE_MAX)) { - dev_info(&spi_device->dev, "Invalid pluggable: %d\n", + dev_info(&spi->dev, "Invalid pluggable: %d\n", qcaspi_pluggable); return -EINVAL; } - dev_info(&spi_device->dev, "ver=%s, clkspeed=%d, burst_len=%d, pluggable=%d\n", + dev_info(&spi->dev, "ver=%s, clkspeed=%d, burst_len=%d, pluggable=%d\n", QCASPI_DRV_VERSION, qcaspi_clkspeed, qcaspi_burst_len, qcaspi_pluggable); - spi_device->mode = SPI_MODE_3; - spi_device->max_speed_hz = qcaspi_clkspeed; - if (spi_setup(spi_device) < 0) { - dev_err(&spi_device->dev, "Unable to setup SPI device\n"); + spi->mode = SPI_MODE_3; + spi->max_speed_hz = qcaspi_clkspeed; + if (spi_setup(spi) < 0) { + dev_err(&spi->dev, "Unable to setup SPI device\n"); return -EFAULT; } @@ -905,23 +905,23 @@ qca_spi_probe(struct spi_device *spi_device) qca = netdev_priv(qcaspi_devs); if (!qca) { free_netdev(qcaspi_devs); - dev_err(&spi_device->dev, "Fail to retrieve private structure\n"); + dev_err(&spi->dev, "Fail to retrieve private structure\n"); return -ENOMEM; } qca->net_dev = qcaspi_devs; - qca->spi_dev = spi_device; + qca->spi_dev = spi; qca->legacy_mode = legacy_mode; - spi_set_drvdata(spi_device, qcaspi_devs); + spi_set_drvdata(spi, qcaspi_devs); - mac = of_get_mac_address(spi_device->dev.of_node); + mac = of_get_mac_address(spi->dev.of_node); if (mac) ether_addr_copy(qca->net_dev->dev_addr, mac); if (!is_valid_ether_addr(qca->net_dev->dev_addr)) { eth_hw_addr_random(qca->net_dev); - dev_info(&spi_device->dev, "Using random MAC address: %pM\n", + dev_info(&spi->dev, "Using random MAC address: %pM\n", qca->net_dev->dev_addr); } @@ -932,7 +932,7 @@ qca_spi_probe(struct spi_device *spi_device) qcaspi_read_register(qca, SPI_REG_SIGNATURE, &signature); if (signature != QCASPI_GOOD_SIGNATURE) { - dev_err(&spi_device->dev, "Invalid signature (0x%04X)\n", + dev_err(&spi->dev, "Invalid signature (0x%04X)\n", signature); free_netdev(qcaspi_devs); return -EFAULT; @@ -940,7 +940,7 @@ qca_spi_probe(struct spi_device *spi_device) } if (register_netdev(qcaspi_devs)) { - dev_info(&spi_device->dev, "Unable to register net device %s\n", + dev_info(&spi->dev, "Unable to register net device %s\n", qcaspi_devs->name); free_netdev(qcaspi_devs); return -EFAULT; @@ -952,9 +952,9 @@ qca_spi_probe(struct spi_device *spi_device) } static int -qca_spi_remove(struct spi_device *spi_device) +qca_spi_remove(struct spi_device *spi) { - struct net_device *qcaspi_devs = spi_get_drvdata(spi_device); + struct net_device *qcaspi_devs = spi_get_drvdata(spi); struct qcaspi *qca = netdev_priv(qcaspi_devs); qcaspi_remove_device_debugfs(qca); diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c index cf98cc9bbc8d..819289e5be4f 100644 --- a/drivers/net/ethernet/rocker/rocker.c +++ b/drivers/net/ethernet/rocker/rocker.c @@ -181,7 +181,7 @@ struct rocker_desc_info { size_t data_size; size_t tlv_size; struct rocker_desc *desc; - DEFINE_DMA_UNMAP_ADDR(mapaddr); + dma_addr_t mapaddr; }; struct rocker_dma_ring_info { @@ -225,6 +225,7 @@ struct rocker_port { struct napi_struct napi_rx; struct rocker_dma_ring_info tx_ring; struct rocker_dma_ring_info rx_ring; + struct list_head trans_mem; }; struct rocker { @@ -236,21 +237,21 @@ struct rocker { struct { u64 id; } hw; - spinlock_t cmd_ring_lock; + spinlock_t cmd_ring_lock; /* for cmd ring accesses */ struct rocker_dma_ring_info cmd_ring; struct rocker_dma_ring_info event_ring; DECLARE_HASHTABLE(flow_tbl, 16); - spinlock_t flow_tbl_lock; + spinlock_t flow_tbl_lock; /* for flow tbl accesses */ u64 flow_tbl_next_cookie; DECLARE_HASHTABLE(group_tbl, 16); - spinlock_t group_tbl_lock; + spinlock_t group_tbl_lock; /* for group tbl accesses */ DECLARE_HASHTABLE(fdb_tbl, 16); - spinlock_t fdb_tbl_lock; + spinlock_t fdb_tbl_lock; /* for fdb tbl accesses */ unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN]; DECLARE_HASHTABLE(internal_vlan_tbl, 8); - spinlock_t internal_vlan_tbl_lock; + spinlock_t internal_vlan_tbl_lock; /* for vlan tbl accesses */ DECLARE_HASHTABLE(neigh_tbl, 16); - spinlock_t neigh_tbl_lock; + spinlock_t neigh_tbl_lock; /* for neigh tbl accesses */ u32 neigh_tbl_next_index; }; @@ -294,7 +295,7 @@ static bool rocker_vlan_id_is_internal(__be16 vlan_id) return (_vlan_id >= start && _vlan_id <= end); } -static __be16 rocker_port_vid_to_vlan(struct rocker_port *rocker_port, +static __be16 rocker_port_vid_to_vlan(const struct rocker_port *rocker_port, u16 vid, bool *pop_vlan) { __be16 vlan_id; @@ -311,7 +312,7 @@ static __be16 rocker_port_vid_to_vlan(struct rocker_port *rocker_port, return vlan_id; } -static u16 rocker_port_vlan_to_vid(struct rocker_port *rocker_port, +static u16 rocker_port_vlan_to_vid(const struct rocker_port *rocker_port, __be16 vlan_id) { if (rocker_vlan_id_is_internal(vlan_id)) @@ -320,21 +321,87 @@ static u16 rocker_port_vlan_to_vid(struct rocker_port *rocker_port, return ntohs(vlan_id); } -static bool rocker_port_is_bridged(struct rocker_port *rocker_port) +static bool rocker_port_is_bridged(const struct rocker_port *rocker_port) { return !!rocker_port->bridge_dev; } +static void *__rocker_port_mem_alloc(struct rocker_port *rocker_port, + enum switchdev_trans trans, size_t size) +{ + struct list_head *elem = NULL; + + /* If in transaction prepare phase, allocate the memory + * and enqueue it on a per-port list. If in transaction + * commit phase, dequeue the memory from the per-port list + * rather than re-allocating the memory. The idea is the + * driver code paths for prepare and commit are identical + * so the memory allocated in the prepare phase is the + * memory used in the commit phase. + */ + + switch (trans) { + case SWITCHDEV_TRANS_PREPARE: + elem = kzalloc(size + sizeof(*elem), GFP_KERNEL); + if (!elem) + return NULL; + list_add_tail(elem, &rocker_port->trans_mem); + break; + case SWITCHDEV_TRANS_COMMIT: + BUG_ON(list_empty(&rocker_port->trans_mem)); + elem = rocker_port->trans_mem.next; + list_del_init(elem); + break; + case SWITCHDEV_TRANS_NONE: + elem = kzalloc(size + sizeof(*elem), GFP_KERNEL); + if (elem) + INIT_LIST_HEAD(elem); + break; + default: + break; + } + + return elem ? elem + 1 : NULL; +} + +static void *rocker_port_kzalloc(struct rocker_port *rocker_port, + enum switchdev_trans trans, size_t size) +{ + return __rocker_port_mem_alloc(rocker_port, trans, size); +} + +static void *rocker_port_kcalloc(struct rocker_port *rocker_port, + enum switchdev_trans trans, size_t n, + size_t size) +{ + return __rocker_port_mem_alloc(rocker_port, trans, n * size); +} + +static void rocker_port_kfree(enum switchdev_trans trans, const void *mem) +{ + struct list_head *elem; + + /* Frees are ignored if in transaction prepare phase. The + * memory remains on the per-port list until freed in the + * commit phase. + */ + + if (trans == SWITCHDEV_TRANS_PREPARE) + return; + + elem = (struct list_head *)mem - 1; + BUG_ON(!list_empty(elem)); + kfree(elem); +} + struct rocker_wait { wait_queue_head_t wait; bool done; - bool nowait; }; static void rocker_wait_reset(struct rocker_wait *wait) { wait->done = false; - wait->nowait = false; } static void rocker_wait_init(struct rocker_wait *wait) @@ -343,20 +410,22 @@ static void rocker_wait_init(struct rocker_wait *wait) rocker_wait_reset(wait); } -static struct rocker_wait *rocker_wait_create(gfp_t gfp) +static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port, + enum switchdev_trans trans) { struct rocker_wait *wait; - wait = kmalloc(sizeof(*wait), gfp); + wait = rocker_port_kzalloc(rocker_port, trans, sizeof(*wait)); if (!wait) return NULL; rocker_wait_init(wait); return wait; } -static void rocker_wait_destroy(struct rocker_wait *work) +static void rocker_wait_destroy(enum switchdev_trans trans, + struct rocker_wait *wait) { - kfree(work); + rocker_port_kfree(trans, wait); } static bool rocker_wait_event_timeout(struct rocker_wait *wait, @@ -374,18 +443,18 @@ static void rocker_wait_wake_up(struct rocker_wait *wait) wake_up(&wait->wait); } -static u32 rocker_msix_vector(struct rocker *rocker, unsigned int vector) +static u32 rocker_msix_vector(const struct rocker *rocker, unsigned int vector) { return rocker->msix_entries[vector].vector; } -static u32 rocker_msix_tx_vector(struct rocker_port *rocker_port) +static u32 rocker_msix_tx_vector(const struct rocker_port *rocker_port) { return rocker_msix_vector(rocker_port->rocker, ROCKER_MSIX_VEC_TX(rocker_port->port_number)); } -static u32 rocker_msix_rx_vector(struct rocker_port *rocker_port) +static u32 rocker_msix_rx_vector(const struct rocker_port *rocker_port) { return rocker_msix_vector(rocker_port->rocker, ROCKER_MSIX_VEC_RX(rocker_port->port_number)); @@ -404,9 +473,9 @@ static u32 rocker_msix_rx_vector(struct rocker_port *rocker_port) * HW basic testing functions *****************************/ -static int rocker_reg_test(struct rocker *rocker) +static int rocker_reg_test(const struct rocker *rocker) { - struct pci_dev *pdev = rocker->pdev; + const struct pci_dev *pdev = rocker->pdev; u64 test_reg; u64 rnd; @@ -434,12 +503,12 @@ static int rocker_reg_test(struct rocker *rocker) return 0; } -static int rocker_dma_test_one(struct rocker *rocker, struct rocker_wait *wait, - u32 test_type, dma_addr_t dma_handle, - unsigned char *buf, unsigned char *expect, - size_t size) +static int rocker_dma_test_one(const struct rocker *rocker, + struct rocker_wait *wait, u32 test_type, + dma_addr_t dma_handle, const unsigned char *buf, + const unsigned char *expect, size_t size) { - struct pci_dev *pdev = rocker->pdev; + const struct pci_dev *pdev = rocker->pdev; int i; rocker_wait_reset(wait); @@ -463,7 +532,7 @@ static int rocker_dma_test_one(struct rocker *rocker, struct rocker_wait *wait, #define ROCKER_TEST_DMA_BUF_SIZE (PAGE_SIZE * 4) #define ROCKER_TEST_DMA_FILL_PATTERN 0x96 -static int rocker_dma_test_offset(struct rocker *rocker, +static int rocker_dma_test_offset(const struct rocker *rocker, struct rocker_wait *wait, int offset) { struct pci_dev *pdev = rocker->pdev; @@ -523,7 +592,8 @@ free_alloc: return err; } -static int rocker_dma_test(struct rocker *rocker, struct rocker_wait *wait) +static int rocker_dma_test(const struct rocker *rocker, + struct rocker_wait *wait) { int i; int err; @@ -545,9 +615,9 @@ static irqreturn_t rocker_test_irq_handler(int irq, void *dev_id) return IRQ_HANDLED; } -static int rocker_basic_hw_test(struct rocker *rocker) +static int rocker_basic_hw_test(const struct rocker *rocker) { - struct pci_dev *pdev = rocker->pdev; + const struct pci_dev *pdev = rocker->pdev; struct rocker_wait wait; int err; @@ -680,7 +750,7 @@ static u64 rocker_tlv_get_u64(const struct rocker_tlv *tlv) return *(u64 *) rocker_tlv_data(tlv); } -static void rocker_tlv_parse(struct rocker_tlv **tb, int maxtype, +static void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype, const char *buf, int buf_len) { const struct rocker_tlv *tlv; @@ -693,19 +763,19 @@ static void rocker_tlv_parse(struct rocker_tlv **tb, int maxtype, u32 type = rocker_tlv_type(tlv); if (type > 0 && type <= maxtype) - tb[type] = (struct rocker_tlv *) tlv; + tb[type] = tlv; } } -static void rocker_tlv_parse_nested(struct rocker_tlv **tb, int maxtype, +static void rocker_tlv_parse_nested(const struct rocker_tlv **tb, int maxtype, const struct rocker_tlv *tlv) { rocker_tlv_parse(tb, maxtype, rocker_tlv_data(tlv), rocker_tlv_len(tlv)); } -static void rocker_tlv_parse_desc(struct rocker_tlv **tb, int maxtype, - struct rocker_desc_info *desc_info) +static void rocker_tlv_parse_desc(const struct rocker_tlv **tb, int maxtype, + const struct rocker_desc_info *desc_info) { rocker_tlv_parse(tb, maxtype, desc_info->data, desc_info->desc->tlv_size); @@ -790,9 +860,9 @@ static void rocker_tlv_nest_end(struct rocker_desc_info *desc_info, } static void rocker_tlv_nest_cancel(struct rocker_desc_info *desc_info, - struct rocker_tlv *start) + const struct rocker_tlv *start) { - desc_info->tlv_size = (char *) start - desc_info->data; + desc_info->tlv_size = (const char *) start - desc_info->data; } /****************************************** @@ -804,7 +874,7 @@ static u32 __pos_inc(u32 pos, size_t limit) return ++pos == limit ? 0 : pos; } -static int rocker_desc_err(struct rocker_desc_info *desc_info) +static int rocker_desc_err(const struct rocker_desc_info *desc_info) { int err = desc_info->desc->comp_err & ~ROCKER_DMA_DESC_COMP_ERR_GEN; @@ -832,31 +902,31 @@ static int rocker_desc_err(struct rocker_desc_info *desc_info) return -EINVAL; } -static void rocker_desc_gen_clear(struct rocker_desc_info *desc_info) +static void rocker_desc_gen_clear(const struct rocker_desc_info *desc_info) { desc_info->desc->comp_err &= ~ROCKER_DMA_DESC_COMP_ERR_GEN; } -static bool rocker_desc_gen(struct rocker_desc_info *desc_info) +static bool rocker_desc_gen(const struct rocker_desc_info *desc_info) { u32 comp_err = desc_info->desc->comp_err; return comp_err & ROCKER_DMA_DESC_COMP_ERR_GEN ? true : false; } -static void *rocker_desc_cookie_ptr_get(struct rocker_desc_info *desc_info) +static void *rocker_desc_cookie_ptr_get(const struct rocker_desc_info *desc_info) { return (void *)(uintptr_t)desc_info->desc->cookie; } -static void rocker_desc_cookie_ptr_set(struct rocker_desc_info *desc_info, +static void rocker_desc_cookie_ptr_set(const struct rocker_desc_info *desc_info, void *ptr) { desc_info->desc->cookie = (uintptr_t) ptr; } static struct rocker_desc_info * -rocker_desc_head_get(struct rocker_dma_ring_info *info) +rocker_desc_head_get(const struct rocker_dma_ring_info *info) { static struct rocker_desc_info *desc_info; u32 head = __pos_inc(info->head, info->size); @@ -868,15 +938,15 @@ rocker_desc_head_get(struct rocker_dma_ring_info *info) return desc_info; } -static void rocker_desc_commit(struct rocker_desc_info *desc_info) +static void rocker_desc_commit(const struct rocker_desc_info *desc_info) { desc_info->desc->buf_size = desc_info->data_size; desc_info->desc->tlv_size = desc_info->tlv_size; } -static void rocker_desc_head_set(struct rocker *rocker, +static void rocker_desc_head_set(const struct rocker *rocker, struct rocker_dma_ring_info *info, - struct rocker_desc_info *desc_info) + const struct rocker_desc_info *desc_info) { u32 head = __pos_inc(info->head, info->size); @@ -901,8 +971,8 @@ rocker_desc_tail_get(struct rocker_dma_ring_info *info) return desc_info; } -static void rocker_dma_ring_credits_set(struct rocker *rocker, - struct rocker_dma_ring_info *info, +static void rocker_dma_ring_credits_set(const struct rocker *rocker, + const struct rocker_dma_ring_info *info, u32 credits) { if (credits) @@ -915,7 +985,7 @@ static unsigned long rocker_dma_ring_size_fix(size_t size) min(roundup_pow_of_two(size), ROCKER_DMA_SIZE_MAX)); } -static int rocker_dma_ring_create(struct rocker *rocker, +static int rocker_dma_ring_create(const struct rocker *rocker, unsigned int type, size_t size, struct rocker_dma_ring_info *info) @@ -951,8 +1021,8 @@ static int rocker_dma_ring_create(struct rocker *rocker, return 0; } -static void rocker_dma_ring_destroy(struct rocker *rocker, - struct rocker_dma_ring_info *info) +static void rocker_dma_ring_destroy(const struct rocker *rocker, + const struct rocker_dma_ring_info *info) { rocker_write64(rocker, DMA_DESC_ADDR(info->type), 0); @@ -962,7 +1032,7 @@ static void rocker_dma_ring_destroy(struct rocker *rocker, kfree(info->desc_info); } -static void rocker_dma_ring_pass_to_producer(struct rocker *rocker, +static void rocker_dma_ring_pass_to_producer(const struct rocker *rocker, struct rocker_dma_ring_info *info) { int i; @@ -977,8 +1047,8 @@ static void rocker_dma_ring_pass_to_producer(struct rocker *rocker, rocker_desc_commit(&info->desc_info[i]); } -static int rocker_dma_ring_bufs_alloc(struct rocker *rocker, - struct rocker_dma_ring_info *info, +static int rocker_dma_ring_bufs_alloc(const struct rocker *rocker, + const struct rocker_dma_ring_info *info, int direction, size_t buf_size) { struct pci_dev *pdev = rocker->pdev; @@ -1015,7 +1085,7 @@ static int rocker_dma_ring_bufs_alloc(struct rocker *rocker, rollback: for (i--; i >= 0; i--) { - struct rocker_desc_info *desc_info = &info->desc_info[i]; + const struct rocker_desc_info *desc_info = &info->desc_info[i]; pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr), desc_info->data_size, direction); @@ -1024,15 +1094,15 @@ rollback: return err; } -static void rocker_dma_ring_bufs_free(struct rocker *rocker, - struct rocker_dma_ring_info *info, +static void rocker_dma_ring_bufs_free(const struct rocker *rocker, + const struct rocker_dma_ring_info *info, int direction) { struct pci_dev *pdev = rocker->pdev; int i; for (i = 0; i < info->size; i++) { - struct rocker_desc_info *desc_info = &info->desc_info[i]; + const struct rocker_desc_info *desc_info = &info->desc_info[i]; struct rocker_desc *desc = &info->desc[i]; desc->buf_addr = 0; @@ -1045,7 +1115,7 @@ static void rocker_dma_ring_bufs_free(struct rocker *rocker, static int rocker_dma_rings_init(struct rocker *rocker) { - struct pci_dev *pdev = rocker->pdev; + const struct pci_dev *pdev = rocker->pdev; int err; err = rocker_dma_ring_create(rocker, ROCKER_DMA_CMD, @@ -1102,11 +1172,11 @@ static void rocker_dma_rings_fini(struct rocker *rocker) rocker_dma_ring_destroy(rocker, &rocker->cmd_ring); } -static int rocker_dma_rx_ring_skb_map(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_dma_rx_ring_skb_map(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, struct sk_buff *skb, size_t buf_len) { + const struct rocker *rocker = rocker_port->rocker; struct pci_dev *pdev = rocker->pdev; dma_addr_t dma_handle; @@ -1126,13 +1196,12 @@ tlv_put_failure: return -EMSGSIZE; } -static size_t rocker_port_rx_buf_len(struct rocker_port *rocker_port) +static size_t rocker_port_rx_buf_len(const struct rocker_port *rocker_port) { return rocker_port->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN; } -static int rocker_dma_rx_ring_skb_alloc(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_dma_rx_ring_skb_alloc(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info) { struct net_device *dev = rocker_port->dev; @@ -1149,8 +1218,7 @@ static int rocker_dma_rx_ring_skb_alloc(struct rocker *rocker, skb = netdev_alloc_skb_ip_align(dev, buf_len); if (!skb) return -ENOMEM; - err = rocker_dma_rx_ring_skb_map(rocker, rocker_port, desc_info, - skb, buf_len); + err = rocker_dma_rx_ring_skb_map(rocker_port, desc_info, skb, buf_len); if (err) { dev_kfree_skb_any(skb); return err; @@ -1159,8 +1227,8 @@ static int rocker_dma_rx_ring_skb_alloc(struct rocker *rocker, return 0; } -static void rocker_dma_rx_ring_skb_unmap(struct rocker *rocker, - struct rocker_tlv **attrs) +static void rocker_dma_rx_ring_skb_unmap(const struct rocker *rocker, + const struct rocker_tlv **attrs) { struct pci_dev *pdev = rocker->pdev; dma_addr_t dma_handle; @@ -1174,10 +1242,10 @@ static void rocker_dma_rx_ring_skb_unmap(struct rocker *rocker, pci_unmap_single(pdev, dma_handle, len, PCI_DMA_FROMDEVICE); } -static void rocker_dma_rx_ring_skb_free(struct rocker *rocker, - struct rocker_desc_info *desc_info) +static void rocker_dma_rx_ring_skb_free(const struct rocker *rocker, + const struct rocker_desc_info *desc_info) { - struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1]; struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info); if (!skb) @@ -1187,15 +1255,15 @@ static void rocker_dma_rx_ring_skb_free(struct rocker *rocker, dev_kfree_skb_any(skb); } -static int rocker_dma_rx_ring_skbs_alloc(struct rocker *rocker, - struct rocker_port *rocker_port) +static int rocker_dma_rx_ring_skbs_alloc(const struct rocker_port *rocker_port) { - struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring; + const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring; + const struct rocker *rocker = rocker_port->rocker; int i; int err; for (i = 0; i < rx_ring->size; i++) { - err = rocker_dma_rx_ring_skb_alloc(rocker, rocker_port, + err = rocker_dma_rx_ring_skb_alloc(rocker_port, &rx_ring->desc_info[i]); if (err) goto rollback; @@ -1208,10 +1276,10 @@ rollback: return err; } -static void rocker_dma_rx_ring_skbs_free(struct rocker *rocker, - struct rocker_port *rocker_port) +static void rocker_dma_rx_ring_skbs_free(const struct rocker_port *rocker_port) { - struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring; + const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring; + const struct rocker *rocker = rocker_port->rocker; int i; for (i = 0; i < rx_ring->size; i++) @@ -1257,7 +1325,7 @@ static int rocker_port_dma_rings_init(struct rocker_port *rocker_port) goto err_dma_rx_ring_bufs_alloc; } - err = rocker_dma_rx_ring_skbs_alloc(rocker, rocker_port); + err = rocker_dma_rx_ring_skbs_alloc(rocker_port); if (err) { netdev_err(rocker_port->dev, "failed to alloc rx dma ring skbs\n"); goto err_dma_rx_ring_skbs_alloc; @@ -1283,7 +1351,7 @@ static void rocker_port_dma_rings_fini(struct rocker_port *rocker_port) { struct rocker *rocker = rocker_port->rocker; - rocker_dma_rx_ring_skbs_free(rocker, rocker_port); + rocker_dma_rx_ring_skbs_free(rocker_port); rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring, PCI_DMA_BIDIRECTIONAL); rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring); @@ -1292,7 +1360,8 @@ static void rocker_port_dma_rings_fini(struct rocker_port *rocker_port) rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring); } -static void rocker_port_set_enable(struct rocker_port *rocker_port, bool enable) +static void rocker_port_set_enable(const struct rocker_port *rocker_port, + bool enable) { u64 val = rocker_read64(rocker_port->rocker, PORT_PHYS_ENABLE); @@ -1310,19 +1379,14 @@ static void rocker_port_set_enable(struct rocker_port *rocker_port, bool enable) static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id) { struct rocker *rocker = dev_id; - struct rocker_desc_info *desc_info; + const struct rocker_desc_info *desc_info; struct rocker_wait *wait; u32 credits = 0; spin_lock(&rocker->cmd_ring_lock); while ((desc_info = rocker_desc_tail_get(&rocker->cmd_ring))) { wait = rocker_desc_cookie_ptr_get(desc_info); - if (wait->nowait) { - rocker_desc_gen_clear(desc_info); - rocker_wait_destroy(wait); - } else { - rocker_wait_wake_up(wait); - } + rocker_wait_wake_up(wait); credits++; } spin_unlock(&rocker->cmd_ring_lock); @@ -1331,22 +1395,22 @@ static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id) return IRQ_HANDLED; } -static void rocker_port_link_up(struct rocker_port *rocker_port) +static void rocker_port_link_up(const struct rocker_port *rocker_port) { netif_carrier_on(rocker_port->dev); netdev_info(rocker_port->dev, "Link is up\n"); } -static void rocker_port_link_down(struct rocker_port *rocker_port) +static void rocker_port_link_down(const struct rocker_port *rocker_port) { netif_carrier_off(rocker_port->dev); netdev_info(rocker_port->dev, "Link is down\n"); } -static int rocker_event_link_change(struct rocker *rocker, +static int rocker_event_link_change(const struct rocker *rocker, const struct rocker_tlv *info) { - struct rocker_tlv *attrs[ROCKER_TLV_EVENT_LINK_CHANGED_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_LINK_CHANGED_MAX + 1]; unsigned int port_number; bool link_up; struct rocker_port *rocker_port; @@ -1374,22 +1438,44 @@ static int rocker_event_link_change(struct rocker *rocker, } #define ROCKER_OP_FLAG_REMOVE BIT(0) -#define ROCKER_OP_FLAG_NOWAIT BIT(1) -#define ROCKER_OP_FLAG_LEARNED BIT(2) -#define ROCKER_OP_FLAG_REFRESH BIT(3) +#define ROCKER_OP_FLAG_LEARNED BIT(1) +#define ROCKER_OP_FLAG_REFRESH BIT(2) static int rocker_port_fdb(struct rocker_port *rocker_port, + enum switchdev_trans trans, const unsigned char *addr, __be16 vlan_id, int flags); -static int rocker_event_mac_vlan_seen(struct rocker *rocker, +struct rocker_mac_vlan_seen_work { + struct work_struct work; + struct rocker_port *rocker_port; + int flags; + unsigned char addr[ETH_ALEN]; + __be16 vlan_id; +}; + +static void rocker_event_mac_vlan_seen_work(struct work_struct *work) +{ + const struct rocker_mac_vlan_seen_work *sw = + container_of(work, struct rocker_mac_vlan_seen_work, work); + + rtnl_lock(); + rocker_port_fdb(sw->rocker_port, SWITCHDEV_TRANS_NONE, + sw->addr, sw->vlan_id, sw->flags); + rtnl_unlock(); + + kfree(work); +} + +static int rocker_event_mac_vlan_seen(const struct rocker *rocker, const struct rocker_tlv *info) { - struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAX + 1]; + struct rocker_mac_vlan_seen_work *sw; + const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAX + 1]; unsigned int port_number; struct rocker_port *rocker_port; - unsigned char *addr; - int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_LEARNED; + const unsigned char *addr; + int flags = ROCKER_OP_FLAG_LEARNED; __be16 vlan_id; rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info); @@ -1411,14 +1497,27 @@ static int rocker_event_mac_vlan_seen(struct rocker *rocker, rocker_port->stp_state != BR_STATE_FORWARDING) return 0; - return rocker_port_fdb(rocker_port, addr, vlan_id, flags); + sw = kmalloc(sizeof(*sw), GFP_ATOMIC); + if (!sw) + return -ENOMEM; + + INIT_WORK(&sw->work, rocker_event_mac_vlan_seen_work); + + sw->rocker_port = rocker_port; + sw->flags = flags; + ether_addr_copy(sw->addr, addr); + sw->vlan_id = vlan_id; + + schedule_work(&sw->work); + + return 0; } -static int rocker_event_process(struct rocker *rocker, - struct rocker_desc_info *desc_info) +static int rocker_event_process(const struct rocker *rocker, + const struct rocker_desc_info *desc_info) { - struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAX + 1]; - struct rocker_tlv *info; + const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAX + 1]; + const struct rocker_tlv *info; u16 type; rocker_tlv_parse_desc(attrs, ROCKER_TLV_EVENT_MAX, desc_info); @@ -1442,8 +1541,8 @@ static int rocker_event_process(struct rocker *rocker, static irqreturn_t rocker_event_irq_handler(int irq, void *dev_id) { struct rocker *rocker = dev_id; - struct pci_dev *pdev = rocker->pdev; - struct rocker_desc_info *desc_info; + const struct pci_dev *pdev = rocker->pdev; + const struct rocker_desc_info *desc_info; u32 credits = 0; int err; @@ -1487,65 +1586,70 @@ static irqreturn_t rocker_rx_irq_handler(int irq, void *dev_id) * Command interface ********************/ -typedef int (*rocker_cmd_cb_t)(struct rocker *rocker, - struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info, - void *priv); +typedef int (*rocker_cmd_prep_cb_t)(const struct rocker_port *rocker_port, + struct rocker_desc_info *desc_info, + void *priv); + +typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info, + void *priv); -static int rocker_cmd_exec(struct rocker *rocker, - struct rocker_port *rocker_port, - rocker_cmd_cb_t prepare, void *prepare_priv, - rocker_cmd_cb_t process, void *process_priv, - bool nowait) +static int rocker_cmd_exec(struct rocker_port *rocker_port, + enum switchdev_trans trans, + rocker_cmd_prep_cb_t prepare, void *prepare_priv, + rocker_cmd_proc_cb_t process, void *process_priv) { + struct rocker *rocker = rocker_port->rocker; struct rocker_desc_info *desc_info; struct rocker_wait *wait; unsigned long flags; int err; - wait = rocker_wait_create(nowait ? GFP_ATOMIC : GFP_KERNEL); + wait = rocker_wait_create(rocker_port, trans); if (!wait) return -ENOMEM; - wait->nowait = nowait; spin_lock_irqsave(&rocker->cmd_ring_lock, flags); + desc_info = rocker_desc_head_get(&rocker->cmd_ring); if (!desc_info) { spin_unlock_irqrestore(&rocker->cmd_ring_lock, flags); err = -EAGAIN; goto out; } - err = prepare(rocker, rocker_port, desc_info, prepare_priv); + + err = prepare(rocker_port, desc_info, prepare_priv); if (err) { spin_unlock_irqrestore(&rocker->cmd_ring_lock, flags); goto out; } + rocker_desc_cookie_ptr_set(desc_info, wait); - rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info); - spin_unlock_irqrestore(&rocker->cmd_ring_lock, flags); - if (nowait) - return 0; + if (trans != SWITCHDEV_TRANS_PREPARE) + rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info); - if (!rocker_wait_event_timeout(wait, HZ / 10)) - return -EIO; + spin_unlock_irqrestore(&rocker->cmd_ring_lock, flags); + + if (trans != SWITCHDEV_TRANS_PREPARE) + if (!rocker_wait_event_timeout(wait, HZ / 10)) + return -EIO; err = rocker_desc_err(desc_info); if (err) return err; if (process) - err = process(rocker, rocker_port, desc_info, process_priv); + err = process(rocker_port, desc_info, process_priv); rocker_desc_gen_clear(desc_info); out: - rocker_wait_destroy(wait); + rocker_wait_destroy(trans, wait); return err; } static int -rocker_cmd_get_port_settings_prep(struct rocker *rocker, - struct rocker_port *rocker_port, +rocker_cmd_get_port_settings_prep(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -1565,14 +1669,13 @@ rocker_cmd_get_port_settings_prep(struct rocker *rocker, } static int -rocker_cmd_get_port_settings_ethtool_proc(struct rocker *rocker, - struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info, +rocker_cmd_get_port_settings_ethtool_proc(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info, void *priv) { struct ethtool_cmd *ecmd = priv; - struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; - struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; + const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; u32 speed; u8 duplex; u8 autoneg; @@ -1604,15 +1707,14 @@ rocker_cmd_get_port_settings_ethtool_proc(struct rocker *rocker, } static int -rocker_cmd_get_port_settings_macaddr_proc(struct rocker *rocker, - struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info, +rocker_cmd_get_port_settings_macaddr_proc(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info, void *priv) { unsigned char *macaddr = priv; - struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; - struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; - struct rocker_tlv *attr; + const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; + const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; + const struct rocker_tlv *attr; rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info); if (!attrs[ROCKER_TLV_CMD_INFO]) @@ -1637,17 +1739,16 @@ struct port_name { }; static int -rocker_cmd_get_port_settings_phys_name_proc(struct rocker *rocker, - struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info, +rocker_cmd_get_port_settings_phys_name_proc(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info, void *priv) { - struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; - struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; + const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; struct port_name *name = priv; - struct rocker_tlv *attr; + const struct rocker_tlv *attr; size_t i, j, len; - char *str; + const char *str; rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info); if (!attrs[ROCKER_TLV_CMD_INFO]) @@ -1679,8 +1780,7 @@ rocker_cmd_get_port_settings_phys_name_proc(struct rocker *rocker, } static int -rocker_cmd_set_port_settings_ethtool_prep(struct rocker *rocker, - struct rocker_port *rocker_port, +rocker_cmd_set_port_settings_ethtool_prep(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -1710,12 +1810,11 @@ rocker_cmd_set_port_settings_ethtool_prep(struct rocker *rocker, } static int -rocker_cmd_set_port_settings_macaddr_prep(struct rocker *rocker, - struct rocker_port *rocker_port, +rocker_cmd_set_port_settings_macaddr_prep(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { - unsigned char *macaddr = priv; + const unsigned char *macaddr = priv; struct rocker_tlv *cmd_info; if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, @@ -1735,8 +1834,7 @@ rocker_cmd_set_port_settings_macaddr_prep(struct rocker *rocker, } static int -rocker_cmd_set_port_learning_prep(struct rocker *rocker, - struct rocker_port *rocker_port, +rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -1761,46 +1859,48 @@ rocker_cmd_set_port_learning_prep(struct rocker *rocker, static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port, struct ethtool_cmd *ecmd) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, rocker_cmd_get_port_settings_prep, NULL, rocker_cmd_get_port_settings_ethtool_proc, - ecmd, false); + ecmd); } static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port, unsigned char *macaddr) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, rocker_cmd_get_port_settings_prep, NULL, rocker_cmd_get_port_settings_macaddr_proc, - macaddr, false); + macaddr); } static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port, struct ethtool_cmd *ecmd) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, rocker_cmd_set_port_settings_ethtool_prep, - ecmd, NULL, NULL, false); + ecmd, NULL, NULL); } static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port, unsigned char *macaddr) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, rocker_cmd_set_port_settings_macaddr_prep, - macaddr, NULL, NULL, false); + macaddr, NULL, NULL); } -static int rocker_port_set_learning(struct rocker_port *rocker_port) +static int rocker_port_set_learning(struct rocker_port *rocker_port, + enum switchdev_trans trans) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, trans, rocker_cmd_set_port_learning_prep, - NULL, NULL, NULL, false); + NULL, NULL, NULL); } -static int rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) +static int +rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info, + const struct rocker_flow_tbl_entry *entry) { if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT, entry->key.ig_port.in_pport)) @@ -1815,8 +1915,9 @@ static int rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) +static int +rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info, + const struct rocker_flow_tbl_entry *entry) { if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT, entry->key.vlan.in_pport)) @@ -1838,8 +1939,9 @@ static int rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) +static int +rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info, + const struct rocker_flow_tbl_entry *entry) { if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT, entry->key.term_mac.in_pport)) @@ -1875,7 +1977,7 @@ static int rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info, static int rocker_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) + const struct rocker_flow_tbl_entry *entry) { if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE, entry->key.ucast_routing.eth_type)) @@ -1896,8 +1998,9 @@ rocker_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) +static int +rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info, + const struct rocker_flow_tbl_entry *entry) { if (entry->key.bridge.has_eth_dst && rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC, @@ -1929,8 +2032,9 @@ static int rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info, - struct rocker_flow_tbl_entry *entry) +static int +rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info, + const struct rocker_flow_tbl_entry *entry) { if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT, entry->key.acl.in_pport)) @@ -1995,12 +2099,11 @@ static int rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_flow_tbl_add(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_cmd_flow_tbl_add(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { - struct rocker_flow_tbl_entry *entry = priv; + const struct rocker_flow_tbl_entry *entry = priv; struct rocker_tlv *cmd_info; int err = 0; @@ -2053,8 +2156,7 @@ static int rocker_cmd_flow_tbl_add(struct rocker *rocker, return 0; } -static int rocker_cmd_flow_tbl_del(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_cmd_flow_tbl_del(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -2090,7 +2192,7 @@ rocker_cmd_group_tbl_add_l2_interface(struct rocker_desc_info *desc_info, static int rocker_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info, - struct rocker_group_tbl_entry *entry) + const struct rocker_group_tbl_entry *entry) { if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER, entry->l2_rewrite.group_id)) @@ -2113,7 +2215,7 @@ rocker_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info, static int rocker_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info, - struct rocker_group_tbl_entry *entry) + const struct rocker_group_tbl_entry *entry) { int i; struct rocker_tlv *group_ids; @@ -2139,7 +2241,7 @@ rocker_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info, static int rocker_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info, - struct rocker_group_tbl_entry *entry) + const struct rocker_group_tbl_entry *entry) { if (!is_zero_ether_addr(entry->l3_unicast.eth_src) && rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC, @@ -2163,8 +2265,7 @@ rocker_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info, return 0; } -static int rocker_cmd_group_tbl_add(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_cmd_group_tbl_add(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -2209,8 +2310,7 @@ static int rocker_cmd_group_tbl_add(struct rocker *rocker, return 0; } -static int rocker_cmd_group_tbl_del(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_cmd_group_tbl_del(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -2293,7 +2393,8 @@ static void rocker_free_tbls(struct rocker *rocker) } static struct rocker_flow_tbl_entry * -rocker_flow_tbl_find(struct rocker *rocker, struct rocker_flow_tbl_entry *match) +rocker_flow_tbl_find(const struct rocker *rocker, + const struct rocker_flow_tbl_entry *match) { struct rocker_flow_tbl_entry *found; size_t key_len = match->key_len ? match->key_len : sizeof(found->key); @@ -2308,8 +2409,8 @@ rocker_flow_tbl_find(struct rocker *rocker, struct rocker_flow_tbl_entry *match) } static int rocker_flow_tbl_add(struct rocker_port *rocker_port, - struct rocker_flow_tbl_entry *match, - bool nowait) + enum switchdev_trans trans, + struct rocker_flow_tbl_entry *match) { struct rocker *rocker = rocker_port->rocker; struct rocker_flow_tbl_entry *found; @@ -2324,8 +2425,9 @@ static int rocker_flow_tbl_add(struct rocker_port *rocker_port, if (found) { match->cookie = found->cookie; - hash_del(&found->entry); - kfree(found); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); + rocker_port_kfree(trans, found); found = match; found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD; } else { @@ -2334,18 +2436,18 @@ static int rocker_flow_tbl_add(struct rocker_port *rocker_port, found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD; } - hash_add(rocker->flow_tbl, &found->entry, found->key_crc32); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_add(rocker->flow_tbl, &found->entry, found->key_crc32); spin_unlock_irqrestore(&rocker->flow_tbl_lock, flags); - return rocker_cmd_exec(rocker, rocker_port, - rocker_cmd_flow_tbl_add, - found, NULL, NULL, nowait); + return rocker_cmd_exec(rocker_port, trans, rocker_cmd_flow_tbl_add, + found, NULL, NULL); } static int rocker_flow_tbl_del(struct rocker_port *rocker_port, - struct rocker_flow_tbl_entry *match, - bool nowait) + enum switchdev_trans trans, + struct rocker_flow_tbl_entry *match) { struct rocker *rocker = rocker_port->rocker; struct rocker_flow_tbl_entry *found; @@ -2360,47 +2462,43 @@ static int rocker_flow_tbl_del(struct rocker_port *rocker_port, found = rocker_flow_tbl_find(rocker, match); if (found) { - hash_del(&found->entry); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL; } spin_unlock_irqrestore(&rocker->flow_tbl_lock, flags); - kfree(match); + rocker_port_kfree(trans, match); if (found) { - err = rocker_cmd_exec(rocker, rocker_port, + err = rocker_cmd_exec(rocker_port, trans, rocker_cmd_flow_tbl_del, - found, NULL, NULL, nowait); - kfree(found); + found, NULL, NULL); + rocker_port_kfree(trans, found); } return err; } -static gfp_t rocker_op_flags_gfp(int flags) -{ - return flags & ROCKER_OP_FLAG_NOWAIT ? GFP_ATOMIC : GFP_KERNEL; -} - static int rocker_flow_tbl_do(struct rocker_port *rocker_port, - int flags, struct rocker_flow_tbl_entry *entry) + enum switchdev_trans trans, int flags, + struct rocker_flow_tbl_entry *entry) { - bool nowait = flags & ROCKER_OP_FLAG_NOWAIT; - if (flags & ROCKER_OP_FLAG_REMOVE) - return rocker_flow_tbl_del(rocker_port, entry, nowait); + return rocker_flow_tbl_del(rocker_port, trans, entry); else - return rocker_flow_tbl_add(rocker_port, entry, nowait); + return rocker_flow_tbl_add(rocker_port, trans, entry); } static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port, - int flags, u32 in_pport, u32 in_pport_mask, + enum switchdev_trans trans, int flags, + u32 in_pport, u32 in_pport_mask, enum rocker_of_dpa_table_id goto_tbl) { struct rocker_flow_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2410,18 +2508,19 @@ static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port, entry->key.ig_port.in_pport_mask = in_pport_mask; entry->key.ig_port.goto_tbl = goto_tbl; - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port, - int flags, u32 in_pport, - __be16 vlan_id, __be16 vlan_id_mask, + enum switchdev_trans trans, int flags, + u32 in_pport, __be16 vlan_id, + __be16 vlan_id_mask, enum rocker_of_dpa_table_id goto_tbl, bool untagged, __be16 new_vlan_id) { struct rocker_flow_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2435,10 +2534,11 @@ static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port, entry->key.vlan.untagged = untagged; entry->key.vlan.new_vlan_id = new_vlan_id; - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port, + enum switchdev_trans trans, u32 in_pport, u32 in_pport_mask, __be16 eth_type, const u8 *eth_dst, const u8 *eth_dst_mask, __be16 vlan_id, @@ -2447,7 +2547,7 @@ static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port, { struct rocker_flow_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2471,11 +2571,11 @@ static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port, entry->key.term_mac.vlan_id_mask = vlan_id_mask; entry->key.term_mac.copy_to_cpu = copy_to_cpu; - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port, - int flags, + enum switchdev_trans trans, int flags, const u8 *eth_dst, const u8 *eth_dst_mask, __be16 vlan_id, u32 tunnel_id, enum rocker_of_dpa_table_id goto_tbl, @@ -2487,7 +2587,7 @@ static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port, bool dflt = !eth_dst || (eth_dst && eth_dst_mask); bool wild = false; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2500,7 +2600,7 @@ static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port, if (eth_dst_mask) { entry->key.bridge.has_eth_dst_mask = 1; ether_addr_copy(entry->key.bridge.eth_dst_mask, eth_dst_mask); - if (memcmp(eth_dst_mask, ff_mac, ETH_ALEN)) + if (!ether_addr_equal(eth_dst_mask, ff_mac)) wild = true; } @@ -2525,10 +2625,11 @@ static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port, entry->key.bridge.group_id = group_id; entry->key.bridge.copy_to_cpu = copy_to_cpu; - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port, + enum switchdev_trans trans, __be16 eth_type, __be32 dst, __be32 dst_mask, u32 priority, enum rocker_of_dpa_table_id goto_tbl, @@ -2536,7 +2637,7 @@ static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port, { struct rocker_flow_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2550,30 +2651,29 @@ static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port, entry->key_len = offsetof(struct rocker_flow_tbl_key, ucast_routing.group_id); - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static int rocker_flow_tbl_acl(struct rocker_port *rocker_port, - int flags, u32 in_pport, - u32 in_pport_mask, + enum switchdev_trans trans, int flags, + u32 in_pport, u32 in_pport_mask, const u8 *eth_src, const u8 *eth_src_mask, const u8 *eth_dst, const u8 *eth_dst_mask, - __be16 eth_type, - __be16 vlan_id, __be16 vlan_id_mask, - u8 ip_proto, u8 ip_proto_mask, - u8 ip_tos, u8 ip_tos_mask, + __be16 eth_type, __be16 vlan_id, + __be16 vlan_id_mask, u8 ip_proto, + u8 ip_proto_mask, u8 ip_tos, u8 ip_tos_mask, u32 group_id) { u32 priority; struct rocker_flow_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; priority = ROCKER_PRIORITY_ACL_NORMAL; if (eth_dst && eth_dst_mask) { - if (memcmp(eth_dst_mask, mcast_mac, ETH_ALEN) == 0) + if (ether_addr_equal(eth_dst_mask, mcast_mac)) priority = ROCKER_PRIORITY_ACL_DFLT; else if (is_link_local_ether_addr(eth_dst)) priority = ROCKER_PRIORITY_ACL_CTRL; @@ -2602,12 +2702,12 @@ static int rocker_flow_tbl_acl(struct rocker_port *rocker_port, entry->key.acl.ip_tos_mask = ip_tos_mask; entry->key.acl.group_id = group_id; - return rocker_flow_tbl_do(rocker_port, flags, entry); + return rocker_flow_tbl_do(rocker_port, trans, flags, entry); } static struct rocker_group_tbl_entry * -rocker_group_tbl_find(struct rocker *rocker, - struct rocker_group_tbl_entry *match) +rocker_group_tbl_find(const struct rocker *rocker, + const struct rocker_group_tbl_entry *match) { struct rocker_group_tbl_entry *found; @@ -2620,22 +2720,23 @@ rocker_group_tbl_find(struct rocker *rocker, return NULL; } -static void rocker_group_tbl_entry_free(struct rocker_group_tbl_entry *entry) +static void rocker_group_tbl_entry_free(enum switchdev_trans trans, + struct rocker_group_tbl_entry *entry) { switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) { case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD: case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST: - kfree(entry->group_ids); + rocker_port_kfree(trans, entry->group_ids); break; default: break; } - kfree(entry); + rocker_port_kfree(trans, entry); } static int rocker_group_tbl_add(struct rocker_port *rocker_port, - struct rocker_group_tbl_entry *match, - bool nowait) + enum switchdev_trans trans, + struct rocker_group_tbl_entry *match) { struct rocker *rocker = rocker_port->rocker; struct rocker_group_tbl_entry *found; @@ -2646,8 +2747,9 @@ static int rocker_group_tbl_add(struct rocker_port *rocker_port, found = rocker_group_tbl_find(rocker, match); if (found) { - hash_del(&found->entry); - rocker_group_tbl_entry_free(found); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); + rocker_group_tbl_entry_free(trans, found); found = match; found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD; } else { @@ -2655,18 +2757,18 @@ static int rocker_group_tbl_add(struct rocker_port *rocker_port, found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD; } - hash_add(rocker->group_tbl, &found->entry, found->group_id); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_add(rocker->group_tbl, &found->entry, found->group_id); spin_unlock_irqrestore(&rocker->group_tbl_lock, flags); - return rocker_cmd_exec(rocker, rocker_port, - rocker_cmd_group_tbl_add, - found, NULL, NULL, nowait); + return rocker_cmd_exec(rocker_port, trans, rocker_cmd_group_tbl_add, + found, NULL, NULL); } static int rocker_group_tbl_del(struct rocker_port *rocker_port, - struct rocker_group_tbl_entry *match, - bool nowait) + enum switchdev_trans trans, + struct rocker_group_tbl_entry *match) { struct rocker *rocker = rocker_port->rocker; struct rocker_group_tbl_entry *found; @@ -2678,93 +2780,95 @@ static int rocker_group_tbl_del(struct rocker_port *rocker_port, found = rocker_group_tbl_find(rocker, match); if (found) { - hash_del(&found->entry); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL; } spin_unlock_irqrestore(&rocker->group_tbl_lock, flags); - rocker_group_tbl_entry_free(match); + rocker_group_tbl_entry_free(trans, match); if (found) { - err = rocker_cmd_exec(rocker, rocker_port, + err = rocker_cmd_exec(rocker_port, trans, rocker_cmd_group_tbl_del, - found, NULL, NULL, nowait); - rocker_group_tbl_entry_free(found); + found, NULL, NULL); + rocker_group_tbl_entry_free(trans, found); } return err; } static int rocker_group_tbl_do(struct rocker_port *rocker_port, - int flags, struct rocker_group_tbl_entry *entry) + enum switchdev_trans trans, int flags, + struct rocker_group_tbl_entry *entry) { - bool nowait = flags & ROCKER_OP_FLAG_NOWAIT; - if (flags & ROCKER_OP_FLAG_REMOVE) - return rocker_group_tbl_del(rocker_port, entry, nowait); + return rocker_group_tbl_del(rocker_port, trans, entry); else - return rocker_group_tbl_add(rocker_port, entry, nowait); + return rocker_group_tbl_add(rocker_port, trans, entry); } static int rocker_group_l2_interface(struct rocker_port *rocker_port, - int flags, __be16 vlan_id, - u32 out_pport, int pop_vlan) + enum switchdev_trans trans, int flags, + __be16 vlan_id, u32 out_pport, + int pop_vlan) { struct rocker_group_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; entry->group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport); entry->l2_interface.pop_vlan = pop_vlan; - return rocker_group_tbl_do(rocker_port, flags, entry); + return rocker_group_tbl_do(rocker_port, trans, flags, entry); } static int rocker_group_l2_fan_out(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, u8 group_count, - u32 *group_ids, u32 group_id) + const u32 *group_ids, u32 group_id) { struct rocker_group_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; entry->group_id = group_id; entry->group_count = group_count; - entry->group_ids = kcalloc(group_count, sizeof(u32), - rocker_op_flags_gfp(flags)); + entry->group_ids = rocker_port_kcalloc(rocker_port, trans, group_count, + sizeof(u32)); if (!entry->group_ids) { - kfree(entry); + rocker_port_kfree(trans, entry); return -ENOMEM; } memcpy(entry->group_ids, group_ids, group_count * sizeof(u32)); - return rocker_group_tbl_do(rocker_port, flags, entry); + return rocker_group_tbl_do(rocker_port, trans, flags, entry); } static int rocker_group_l2_flood(struct rocker_port *rocker_port, - int flags, __be16 vlan_id, - u8 group_count, u32 *group_ids, - u32 group_id) + enum switchdev_trans trans, int flags, + __be16 vlan_id, u8 group_count, + const u32 *group_ids, u32 group_id) { - return rocker_group_l2_fan_out(rocker_port, flags, + return rocker_group_l2_fan_out(rocker_port, trans, flags, group_count, group_ids, group_id); } static int rocker_group_l3_unicast(struct rocker_port *rocker_port, - int flags, u32 index, u8 *src_mac, - u8 *dst_mac, __be16 vlan_id, - bool ttl_check, u32 pport) + enum switchdev_trans trans, int flags, + u32 index, const u8 *src_mac, const u8 *dst_mac, + __be16 vlan_id, bool ttl_check, u32 pport) { struct rocker_group_tbl_entry *entry; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2777,11 +2881,11 @@ static int rocker_group_l3_unicast(struct rocker_port *rocker_port, entry->l3_unicast.ttl_check = ttl_check; entry->l3_unicast.group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, pport); - return rocker_group_tbl_do(rocker_port, flags, entry); + return rocker_group_tbl_do(rocker_port, trans, flags, entry); } static struct rocker_neigh_tbl_entry * - rocker_neigh_tbl_find(struct rocker *rocker, __be32 ip_addr) +rocker_neigh_tbl_find(const struct rocker *rocker, __be32 ip_addr) { struct rocker_neigh_tbl_entry *found; @@ -2794,37 +2898,44 @@ static struct rocker_neigh_tbl_entry * } static void _rocker_neigh_add(struct rocker *rocker, + enum switchdev_trans trans, struct rocker_neigh_tbl_entry *entry) { - entry->index = rocker->neigh_tbl_next_index++; + entry->index = rocker->neigh_tbl_next_index; + if (trans == SWITCHDEV_TRANS_PREPARE) + return; + rocker->neigh_tbl_next_index++; entry->ref_count++; hash_add(rocker->neigh_tbl, &entry->entry, be32_to_cpu(entry->ip_addr)); } -static void _rocker_neigh_del(struct rocker *rocker, +static void _rocker_neigh_del(enum switchdev_trans trans, struct rocker_neigh_tbl_entry *entry) { + if (trans == SWITCHDEV_TRANS_PREPARE) + return; if (--entry->ref_count == 0) { hash_del(&entry->entry); - kfree(entry); + rocker_port_kfree(trans, entry); } } -static void _rocker_neigh_update(struct rocker *rocker, - struct rocker_neigh_tbl_entry *entry, - u8 *eth_dst, bool ttl_check) +static void _rocker_neigh_update(struct rocker_neigh_tbl_entry *entry, + enum switchdev_trans trans, + const u8 *eth_dst, bool ttl_check) { if (eth_dst) { ether_addr_copy(entry->eth_dst, eth_dst); entry->ttl_check = ttl_check; - } else { + } else if (trans != SWITCHDEV_TRANS_PREPARE) { entry->ref_count++; } } static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, - int flags, __be32 ip_addr, u8 *eth_dst) + enum switchdev_trans trans, + int flags, __be32 ip_addr, const u8 *eth_dst) { struct rocker *rocker = rocker_port->rocker; struct rocker_neigh_tbl_entry *entry; @@ -2840,7 +2951,7 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, bool removing; int err = 0; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2857,12 +2968,12 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, entry->dev = rocker_port->dev; ether_addr_copy(entry->eth_dst, eth_dst); entry->ttl_check = true; - _rocker_neigh_add(rocker, entry); + _rocker_neigh_add(rocker, trans, entry); } else if (removing) { memcpy(entry, found, sizeof(*entry)); - _rocker_neigh_del(rocker, found); + _rocker_neigh_del(trans, found); } else if (updating) { - _rocker_neigh_update(rocker, found, eth_dst, true); + _rocker_neigh_update(found, trans, eth_dst, true); memcpy(entry, found, sizeof(*entry)); } else { err = -ENOENT; @@ -2879,7 +2990,7 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, * other routes' nexthops. */ - err = rocker_group_l3_unicast(rocker_port, flags, + err = rocker_group_l3_unicast(rocker_port, trans, flags, entry->index, rocker_port->dev->dev_addr, entry->eth_dst, @@ -2895,7 +3006,7 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, if (adding || removing) { group_id = ROCKER_GROUP_L3_UNICAST(entry->index); - err = rocker_flow_tbl_ucast4_routing(rocker_port, + err = rocker_flow_tbl_ucast4_routing(rocker_port, trans, eth_type, ip_addr, inet_make_mask(32), priority, goto_tbl, @@ -2909,13 +3020,13 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, err_out: if (!adding) - kfree(entry); + rocker_port_kfree(trans, entry); return err; } static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port, - __be32 ip_addr) + enum switchdev_trans trans, __be32 ip_addr) { struct net_device *dev = rocker_port->dev; struct neighbour *n = __ipv4_neigh_lookup(dev, (__force u32)ip_addr); @@ -2933,7 +3044,8 @@ static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port, */ if (n->nud_state & NUD_VALID) - err = rocker_port_ipv4_neigh(rocker_port, 0, ip_addr, n->ha); + err = rocker_port_ipv4_neigh(rocker_port, trans, 0, + ip_addr, n->ha); else neigh_event_send(n, NULL); @@ -2941,7 +3053,8 @@ static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port, return err; } -static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, +static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, __be32 ip_addr, u32 *index) { struct rocker *rocker = rocker_port->rocker; @@ -2954,7 +3067,7 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, bool resolved = true; int err = 0; - entry = kzalloc(sizeof(*entry), rocker_op_flags_gfp(flags)); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return -ENOMEM; @@ -2971,13 +3084,13 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, if (adding) { entry->ip_addr = ip_addr; entry->dev = rocker_port->dev; - _rocker_neigh_add(rocker, entry); + _rocker_neigh_add(rocker, trans, entry); *index = entry->index; resolved = false; } else if (removing) { - _rocker_neigh_del(rocker, found); + _rocker_neigh_del(trans, found); } else if (updating) { - _rocker_neigh_update(rocker, found, NULL, false); + _rocker_neigh_update(found, trans, NULL, false); resolved = !is_zero_ether_addr(found->eth_dst); } else { err = -ENOENT; @@ -2986,7 +3099,7 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags); if (!adding) - kfree(entry); + rocker_port_kfree(trans, entry); if (err) return err; @@ -2994,24 +3107,25 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, /* Resolved means neigh ip_addr is resolved to neigh mac. */ if (!resolved) - err = rocker_port_ipv4_resolve(rocker_port, ip_addr); + err = rocker_port_ipv4_resolve(rocker_port, trans, ip_addr); return err; } static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, __be16 vlan_id) { struct rocker_port *p; - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0); u32 *group_ids; u8 group_count = 0; int err = 0; int i; - group_ids = kcalloc(rocker->port_count, sizeof(u32), - rocker_op_flags_gfp(flags)); + group_ids = rocker_port_kcalloc(rocker_port, trans, rocker->port_count, + sizeof(u32)); if (!group_ids) return -ENOMEM; @@ -3022,6 +3136,8 @@ static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port, for (i = 0; i < rocker->port_count; i++) { p = rocker->ports[i]; + if (!p) + continue; if (!rocker_port_is_bridged(p)) continue; if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) { @@ -3034,23 +3150,22 @@ static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port, if (group_count == 0) goto no_ports_in_vlan; - err = rocker_group_l2_flood(rocker_port, flags, vlan_id, - group_count, group_ids, - group_id); + err = rocker_group_l2_flood(rocker_port, trans, flags, vlan_id, + group_count, group_ids, group_id); if (err) netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 flood group\n", err); no_ports_in_vlan: - kfree(group_ids); + rocker_port_kfree(trans, group_ids); return err; } static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port, - int flags, __be16 vlan_id, - bool pop_vlan) + enum switchdev_trans trans, int flags, + __be16 vlan_id, bool pop_vlan) { - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; struct rocker_port *p; bool adding = !(flags & ROCKER_OP_FLAG_REMOVE); u32 out_pport; @@ -3065,9 +3180,8 @@ static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port, if (rocker_port->stp_state == BR_STATE_LEARNING || rocker_port->stp_state == BR_STATE_FORWARDING) { out_pport = rocker_port->pport; - err = rocker_group_l2_interface(rocker_port, flags, - vlan_id, out_pport, - pop_vlan); + err = rocker_group_l2_interface(rocker_port, trans, flags, + vlan_id, out_pport, pop_vlan); if (err) { netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 group for pport %d\n", @@ -3083,7 +3197,7 @@ static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port, for (i = 0; i < rocker->port_count; i++) { p = rocker->ports[i]; - if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) + if (p && test_bit(ntohs(vlan_id), p->vlan_bitmap)) ref++; } @@ -3091,9 +3205,8 @@ static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port, return 0; out_pport = 0; - err = rocker_group_l2_interface(rocker_port, flags, - vlan_id, out_pport, - pop_vlan); + err = rocker_group_l2_interface(rocker_port, trans, flags, + vlan_id, out_pport, pop_vlan); if (err) { netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 group for CPU port\n", err); @@ -3149,14 +3262,14 @@ static struct rocker_ctrl { }; static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port, - int flags, struct rocker_ctrl *ctrl, - __be16 vlan_id) + enum switchdev_trans trans, int flags, + const struct rocker_ctrl *ctrl, __be16 vlan_id) { u32 in_pport = rocker_port->pport; u32 in_pport_mask = 0xffffffff; u32 out_pport = 0; - u8 *eth_src = NULL; - u8 *eth_src_mask = NULL; + const u8 *eth_src = NULL; + const u8 *eth_src_mask = NULL; __be16 vlan_id_mask = htons(0xffff); u8 ip_proto = 0; u8 ip_proto_mask = 0; @@ -3165,7 +3278,7 @@ static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port, u32 group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport); int err; - err = rocker_flow_tbl_acl(rocker_port, flags, + err = rocker_flow_tbl_acl(rocker_port, trans, flags, in_pport, in_pport_mask, eth_src, eth_src_mask, ctrl->eth_dst, ctrl->eth_dst_mask, @@ -3182,7 +3295,8 @@ static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port, } static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port, - int flags, struct rocker_ctrl *ctrl, + enum switchdev_trans trans, int flags, + const struct rocker_ctrl *ctrl, __be16 vlan_id) { enum rocker_of_dpa_table_id goto_tbl = @@ -3194,7 +3308,7 @@ static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port, if (!rocker_port_is_bridged(rocker_port)) return 0; - err = rocker_flow_tbl_bridge(rocker_port, flags, + err = rocker_flow_tbl_bridge(rocker_port, trans, flags, ctrl->eth_dst, ctrl->eth_dst_mask, vlan_id, tunnel_id, goto_tbl, group_id, ctrl->copy_to_cpu); @@ -3206,8 +3320,8 @@ static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port, } static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port, - int flags, struct rocker_ctrl *ctrl, - __be16 vlan_id) + enum switchdev_trans trans, int flags, + const struct rocker_ctrl *ctrl, __be16 vlan_id) { u32 in_pport_mask = 0xffffffff; __be16 vlan_id_mask = htons(0xffff); @@ -3216,7 +3330,7 @@ static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port, if (ntohs(vlan_id) == 0) vlan_id = rocker_port->internal_vlan_id; - err = rocker_flow_tbl_term_mac(rocker_port, + err = rocker_flow_tbl_term_mac(rocker_port, trans, rocker_port->pport, in_pport_mask, ctrl->eth_type, ctrl->eth_dst, ctrl->eth_dst_mask, vlan_id, @@ -3229,32 +3343,34 @@ static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port, return err; } -static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port, int flags, - struct rocker_ctrl *ctrl, __be16 vlan_id) +static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, + const struct rocker_ctrl *ctrl, __be16 vlan_id) { if (ctrl->acl) - return rocker_port_ctrl_vlan_acl(rocker_port, flags, + return rocker_port_ctrl_vlan_acl(rocker_port, trans, flags, ctrl, vlan_id); if (ctrl->bridge) - return rocker_port_ctrl_vlan_bridge(rocker_port, flags, + return rocker_port_ctrl_vlan_bridge(rocker_port, trans, flags, ctrl, vlan_id); if (ctrl->term) - return rocker_port_ctrl_vlan_term(rocker_port, flags, + return rocker_port_ctrl_vlan_term(rocker_port, trans, flags, ctrl, vlan_id); return -EOPNOTSUPP; } static int rocker_port_ctrl_vlan_add(struct rocker_port *rocker_port, - int flags, __be16 vlan_id) + enum switchdev_trans trans, int flags, + __be16 vlan_id) { int err = 0; int i; for (i = 0; i < ROCKER_CTRL_MAX; i++) { if (rocker_port->ctrls[i]) { - err = rocker_port_ctrl_vlan(rocker_port, flags, + err = rocker_port_ctrl_vlan(rocker_port, trans, flags, &rocker_ctrls[i], vlan_id); if (err) return err; @@ -3264,8 +3380,9 @@ static int rocker_port_ctrl_vlan_add(struct rocker_port *rocker_port, return err; } -static int rocker_port_ctrl(struct rocker_port *rocker_port, int flags, - struct rocker_ctrl *ctrl) +static int rocker_port_ctrl(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, + const struct rocker_ctrl *ctrl) { u16 vid; int err = 0; @@ -3273,7 +3390,7 @@ static int rocker_port_ctrl(struct rocker_port *rocker_port, int flags, for (vid = 1; vid < VLAN_N_VID; vid++) { if (!test_bit(vid, rocker_port->vlan_bitmap)) continue; - err = rocker_port_ctrl_vlan(rocker_port, flags, + err = rocker_port_ctrl_vlan(rocker_port, trans, flags, ctrl, htons(vid)); if (err) break; @@ -3282,8 +3399,8 @@ static int rocker_port_ctrl(struct rocker_port *rocker_port, int flags, return err; } -static int rocker_port_vlan(struct rocker_port *rocker_port, int flags, - u16 vid) +static int rocker_port_vlan(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, u16 vid) { enum rocker_of_dpa_table_id goto_tbl = ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC; @@ -3297,50 +3414,57 @@ static int rocker_port_vlan(struct rocker_port *rocker_port, int flags, internal_vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, &untagged); - if (adding && test_and_set_bit(ntohs(internal_vlan_id), - rocker_port->vlan_bitmap)) + if (adding && test_bit(ntohs(internal_vlan_id), + rocker_port->vlan_bitmap)) return 0; /* already added */ - else if (!adding && !test_and_clear_bit(ntohs(internal_vlan_id), - rocker_port->vlan_bitmap)) + else if (!adding && !test_bit(ntohs(internal_vlan_id), + rocker_port->vlan_bitmap)) return 0; /* already removed */ + change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap); + if (adding) { - err = rocker_port_ctrl_vlan_add(rocker_port, flags, + err = rocker_port_ctrl_vlan_add(rocker_port, trans, flags, internal_vlan_id); if (err) { netdev_err(rocker_port->dev, "Error (%d) port ctrl vlan add\n", err); - return err; + goto err_out; } } - err = rocker_port_vlan_l2_groups(rocker_port, flags, + err = rocker_port_vlan_l2_groups(rocker_port, trans, flags, internal_vlan_id, untagged); if (err) { netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 groups\n", err); - return err; + goto err_out; } - err = rocker_port_vlan_flood_group(rocker_port, flags, + err = rocker_port_vlan_flood_group(rocker_port, trans, flags, internal_vlan_id); if (err) { netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 flood group\n", err); - return err; + goto err_out; } - err = rocker_flow_tbl_vlan(rocker_port, flags, + err = rocker_flow_tbl_vlan(rocker_port, trans, flags, in_pport, vlan_id, vlan_id_mask, goto_tbl, untagged, internal_vlan_id); if (err) netdev_err(rocker_port->dev, "Error (%d) port VLAN table\n", err); +err_out: + if (trans == SWITCHDEV_TRANS_PREPARE) + change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap); + return err; } -static int rocker_port_ig_tbl(struct rocker_port *rocker_port, int flags) +static int rocker_port_ig_tbl(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags) { enum rocker_of_dpa_table_id goto_tbl; u32 in_pport; @@ -3355,7 +3479,7 @@ static int rocker_port_ig_tbl(struct rocker_port *rocker_port, int flags) in_pport_mask = 0xffff0000; goto_tbl = ROCKER_OF_DPA_TABLE_ID_VLAN; - err = rocker_flow_tbl_ig_port(rocker_port, flags, + err = rocker_flow_tbl_ig_port(rocker_port, trans, flags, in_pport, in_pport_mask, goto_tbl); if (err) @@ -3367,7 +3491,8 @@ static int rocker_port_ig_tbl(struct rocker_port *rocker_port, int flags) struct rocker_fdb_learn_work { struct work_struct work; - struct net_device *dev; + struct rocker_port *rocker_port; + enum switchdev_trans trans; int flags; u8 addr[ETH_ALEN]; u16 vid; @@ -3375,27 +3500,28 @@ struct rocker_fdb_learn_work { static void rocker_port_fdb_learn_work(struct work_struct *work) { - struct rocker_fdb_learn_work *lw = + const struct rocker_fdb_learn_work *lw = container_of(work, struct rocker_fdb_learn_work, work); bool removing = (lw->flags & ROCKER_OP_FLAG_REMOVE); bool learned = (lw->flags & ROCKER_OP_FLAG_LEARNED); - struct netdev_switch_notifier_fdb_info info; + struct switchdev_notifier_fdb_info info; info.addr = lw->addr; info.vid = lw->vid; if (learned && removing) - call_netdev_switch_notifiers(NETDEV_SWITCH_FDB_DEL, - lw->dev, &info.info); + call_switchdev_notifiers(SWITCHDEV_FDB_DEL, + lw->rocker_port->dev, &info.info); else if (learned && !removing) - call_netdev_switch_notifiers(NETDEV_SWITCH_FDB_ADD, - lw->dev, &info.info); + call_switchdev_notifiers(SWITCHDEV_FDB_ADD, + lw->rocker_port->dev, &info.info); - kfree(work); + rocker_port_kfree(lw->trans, work); } static int rocker_port_fdb_learn(struct rocker_port *rocker_port, - int flags, const u8 *addr, __be16 vlan_id) + enum switchdev_trans trans, int flags, + const u8 *addr, __be16 vlan_id) { struct rocker_fdb_learn_work *lw; enum rocker_of_dpa_table_id goto_tbl = @@ -3411,8 +3537,8 @@ static int rocker_port_fdb_learn(struct rocker_port *rocker_port, group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport); if (!(flags & ROCKER_OP_FLAG_REFRESH)) { - err = rocker_flow_tbl_bridge(rocker_port, flags, addr, NULL, - vlan_id, tunnel_id, goto_tbl, + err = rocker_flow_tbl_bridge(rocker_port, trans, flags, addr, + NULL, vlan_id, tunnel_id, goto_tbl, group_id, copy_to_cpu); if (err) return err; @@ -3424,24 +3550,29 @@ static int rocker_port_fdb_learn(struct rocker_port *rocker_port, if (!rocker_port_is_bridged(rocker_port)) return 0; - lw = kmalloc(sizeof(*lw), rocker_op_flags_gfp(flags)); + lw = rocker_port_kzalloc(rocker_port, trans, sizeof(*lw)); if (!lw) return -ENOMEM; INIT_WORK(&lw->work, rocker_port_fdb_learn_work); - lw->dev = rocker_port->dev; + lw->rocker_port = rocker_port; + lw->trans = trans; lw->flags = flags; ether_addr_copy(lw->addr, addr); lw->vid = rocker_port_vlan_to_vid(rocker_port, vlan_id); - schedule_work(&lw->work); + if (trans == SWITCHDEV_TRANS_PREPARE) + rocker_port_kfree(trans, lw); + else + schedule_work(&lw->work); return 0; } static struct rocker_fdb_tbl_entry * -rocker_fdb_tbl_find(struct rocker *rocker, struct rocker_fdb_tbl_entry *match) +rocker_fdb_tbl_find(const struct rocker *rocker, + const struct rocker_fdb_tbl_entry *match) { struct rocker_fdb_tbl_entry *found; @@ -3453,6 +3584,7 @@ rocker_fdb_tbl_find(struct rocker *rocker, struct rocker_fdb_tbl_entry *match) } static int rocker_port_fdb(struct rocker_port *rocker_port, + enum switchdev_trans trans, const unsigned char *addr, __be16 vlan_id, int flags) { @@ -3462,7 +3594,7 @@ static int rocker_port_fdb(struct rocker_port *rocker_port, bool removing = (flags & ROCKER_OP_FLAG_REMOVE); unsigned long lock_flags; - fdb = kzalloc(sizeof(*fdb), rocker_op_flags_gfp(flags)); + fdb = rocker_port_kzalloc(rocker_port, trans, sizeof(*fdb)); if (!fdb) return -ENOMEM; @@ -3477,32 +3609,35 @@ static int rocker_port_fdb(struct rocker_port *rocker_port, found = rocker_fdb_tbl_find(rocker, fdb); if (removing && found) { - kfree(fdb); - hash_del(&found->entry); + rocker_port_kfree(trans, fdb); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); } else if (!removing && !found) { - hash_add(rocker->fdb_tbl, &fdb->entry, fdb->key_crc32); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_add(rocker->fdb_tbl, &fdb->entry, fdb->key_crc32); } spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags); /* Check if adding and already exists, or removing and can't find */ if (!found != !removing) { - kfree(fdb); + rocker_port_kfree(trans, fdb); if (!found && removing) return 0; /* Refreshing existing to update aging timers */ flags |= ROCKER_OP_FLAG_REFRESH; } - return rocker_port_fdb_learn(rocker_port, flags, addr, vlan_id); + return rocker_port_fdb_learn(rocker_port, trans, flags, addr, vlan_id); } -static int rocker_port_fdb_flush(struct rocker_port *rocker_port) +static int rocker_port_fdb_flush(struct rocker_port *rocker_port, + enum switchdev_trans trans) { struct rocker *rocker = rocker_port->rocker; struct rocker_fdb_tbl_entry *found; unsigned long lock_flags; - int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE; + int flags = ROCKER_OP_FLAG_REMOVE; struct hlist_node *tmp; int bkt; int err = 0; @@ -3518,12 +3653,13 @@ static int rocker_port_fdb_flush(struct rocker_port *rocker_port) continue; if (!found->learned) continue; - err = rocker_port_fdb_learn(rocker_port, flags, + err = rocker_port_fdb_learn(rocker_port, trans, flags, found->key.addr, found->key.vlan_id); if (err) goto err_out; - hash_del(&found->entry); + if (trans != SWITCHDEV_TRANS_PREPARE) + hash_del(&found->entry); } err_out: @@ -3533,7 +3669,8 @@ err_out: } static int rocker_port_router_mac(struct rocker_port *rocker_port, - int flags, __be16 vlan_id) + enum switchdev_trans trans, int flags, + __be16 vlan_id) { u32 in_pport_mask = 0xffffffff; __be16 eth_type; @@ -3546,7 +3683,7 @@ static int rocker_port_router_mac(struct rocker_port *rocker_port, vlan_id = rocker_port->internal_vlan_id; eth_type = htons(ETH_P_IP); - err = rocker_flow_tbl_term_mac(rocker_port, + err = rocker_flow_tbl_term_mac(rocker_port, trans, rocker_port->pport, in_pport_mask, eth_type, rocker_port->dev->dev_addr, dst_mac_mask, vlan_id, vlan_id_mask, @@ -3555,7 +3692,7 @@ static int rocker_port_router_mac(struct rocker_port *rocker_port, return err; eth_type = htons(ETH_P_IPV6); - err = rocker_flow_tbl_term_mac(rocker_port, + err = rocker_flow_tbl_term_mac(rocker_port, trans, rocker_port->pport, in_pport_mask, eth_type, rocker_port->dev->dev_addr, dst_mac_mask, vlan_id, vlan_id_mask, @@ -3564,13 +3701,14 @@ static int rocker_port_router_mac(struct rocker_port *rocker_port, return err; } -static int rocker_port_fwding(struct rocker_port *rocker_port) +static int rocker_port_fwding(struct rocker_port *rocker_port, + enum switchdev_trans trans) { bool pop_vlan; u32 out_pport; __be16 vlan_id; u16 vid; - int flags = ROCKER_OP_FLAG_NOWAIT; + int flags = 0; int err; /* Port will be forwarding-enabled if its STP state is LEARNING @@ -3590,9 +3728,8 @@ static int rocker_port_fwding(struct rocker_port *rocker_port) continue; vlan_id = htons(vid); pop_vlan = rocker_vlan_id_is_internal(vlan_id); - err = rocker_group_l2_interface(rocker_port, flags, - vlan_id, out_pport, - pop_vlan); + err = rocker_group_l2_interface(rocker_port, trans, flags, + vlan_id, out_pport, pop_vlan); if (err) { netdev_err(rocker_port->dev, "Error (%d) port VLAN l2 group for pport %d\n", @@ -3604,13 +3741,21 @@ static int rocker_port_fwding(struct rocker_port *rocker_port) return 0; } -static int rocker_port_stp_update(struct rocker_port *rocker_port, u8 state) +static int rocker_port_stp_update(struct rocker_port *rocker_port, + enum switchdev_trans trans, u8 state) { bool want[ROCKER_CTRL_MAX] = { 0, }; + bool prev_ctrls[ROCKER_CTRL_MAX]; + u8 prev_state; int flags; int err; int i; + if (trans == SWITCHDEV_TRANS_PREPARE) { + memcpy(prev_ctrls, rocker_port->ctrls, sizeof(prev_ctrls)); + prev_state = rocker_port->stp_state; + } + if (rocker_port->stp_state == state) return 0; @@ -3638,45 +3783,54 @@ static int rocker_port_stp_update(struct rocker_port *rocker_port, u8 state) for (i = 0; i < ROCKER_CTRL_MAX; i++) { if (want[i] != rocker_port->ctrls[i]) { - flags = ROCKER_OP_FLAG_NOWAIT | - (want[i] ? 0 : ROCKER_OP_FLAG_REMOVE); - err = rocker_port_ctrl(rocker_port, flags, + flags = (want[i] ? 0 : ROCKER_OP_FLAG_REMOVE); + err = rocker_port_ctrl(rocker_port, trans, flags, &rocker_ctrls[i]); if (err) - return err; + goto err_out; rocker_port->ctrls[i] = want[i]; } } - err = rocker_port_fdb_flush(rocker_port); + err = rocker_port_fdb_flush(rocker_port, trans); if (err) - return err; + goto err_out; + + err = rocker_port_fwding(rocker_port, trans); - return rocker_port_fwding(rocker_port); +err_out: + if (trans == SWITCHDEV_TRANS_PREPARE) { + memcpy(rocker_port->ctrls, prev_ctrls, sizeof(prev_ctrls)); + rocker_port->stp_state = prev_state; + } + + return err; } -static int rocker_port_fwd_enable(struct rocker_port *rocker_port) +static int rocker_port_fwd_enable(struct rocker_port *rocker_port, + enum switchdev_trans trans) { if (rocker_port_is_bridged(rocker_port)) /* bridge STP will enable port */ return 0; /* port is not bridged, so simulate going to FORWARDING state */ - return rocker_port_stp_update(rocker_port, BR_STATE_FORWARDING); + return rocker_port_stp_update(rocker_port, trans, BR_STATE_FORWARDING); } -static int rocker_port_fwd_disable(struct rocker_port *rocker_port) +static int rocker_port_fwd_disable(struct rocker_port *rocker_port, + enum switchdev_trans trans) { if (rocker_port_is_bridged(rocker_port)) /* bridge STP will disable port */ return 0; /* port is not bridged, so simulate going to DISABLED state */ - return rocker_port_stp_update(rocker_port, BR_STATE_DISABLED); + return rocker_port_stp_update(rocker_port, trans, BR_STATE_DISABLED); } static struct rocker_internal_vlan_tbl_entry * -rocker_internal_vlan_tbl_find(struct rocker *rocker, int ifindex) +rocker_internal_vlan_tbl_find(const struct rocker *rocker, int ifindex) { struct rocker_internal_vlan_tbl_entry *found; @@ -3731,8 +3885,9 @@ found: return found->vlan_id; } -static void rocker_port_internal_vlan_id_put(struct rocker_port *rocker_port, - int ifindex) +static void +rocker_port_internal_vlan_id_put(const struct rocker_port *rocker_port, + int ifindex) { struct rocker *rocker = rocker_port->rocker; struct rocker_internal_vlan_tbl_entry *found; @@ -3760,11 +3915,12 @@ not_found: spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags); } -static int rocker_port_fib_ipv4(struct rocker_port *rocker_port, __be32 dst, - int dst_len, struct fib_info *fi, u32 tb_id, - int flags) +static int rocker_port_fib_ipv4(struct rocker_port *rocker_port, + enum switchdev_trans trans, __be32 dst, + int dst_len, const struct fib_info *fi, + u32 tb_id, int flags) { - struct fib_nh *nh; + const struct fib_nh *nh; __be16 eth_type = htons(ETH_P_IP); __be32 dst_mask = inet_make_mask(dst_len); __be16 internal_vlan_id = rocker_port->internal_vlan_id; @@ -3784,7 +3940,7 @@ static int rocker_port_fib_ipv4(struct rocker_port *rocker_port, __be32 dst, has_gw = !!nh->nh_gw; if (has_gw && nh_on_port) { - err = rocker_port_ipv4_nh(rocker_port, flags, + err = rocker_port_ipv4_nh(rocker_port, trans, flags, nh->nh_gw, &index); if (err) return err; @@ -3795,7 +3951,7 @@ static int rocker_port_fib_ipv4(struct rocker_port *rocker_port, __be32 dst, group_id = ROCKER_GROUP_L2_INTERFACE(internal_vlan_id, 0); } - err = rocker_flow_tbl_ucast4_routing(rocker_port, eth_type, dst, + err = rocker_flow_tbl_ucast4_routing(rocker_port, trans, eth_type, dst, dst_mask, priority, goto_tbl, group_id, flags); if (err) @@ -3834,7 +3990,7 @@ static int rocker_port_open(struct net_device *dev) goto err_request_rx_irq; } - err = rocker_port_fwd_enable(rocker_port); + err = rocker_port_fwd_enable(rocker_port, SWITCHDEV_TRANS_NONE); if (err) goto err_fwd_enable; @@ -3861,7 +4017,7 @@ static int rocker_port_stop(struct net_device *dev) rocker_port_set_enable(rocker_port, false); napi_disable(&rocker_port->napi_rx); napi_disable(&rocker_port->napi_tx); - rocker_port_fwd_disable(rocker_port); + rocker_port_fwd_disable(rocker_port, SWITCHDEV_TRANS_NONE); free_irq(rocker_msix_rx_vector(rocker_port), rocker_port); free_irq(rocker_msix_tx_vector(rocker_port), rocker_port); rocker_port_dma_rings_fini(rocker_port); @@ -3869,12 +4025,12 @@ static int rocker_port_stop(struct net_device *dev) return 0; } -static void rocker_tx_desc_frags_unmap(struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info) +static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info) { - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; struct pci_dev *pdev = rocker->pdev; - struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1]; struct rocker_tlv *attr; int rem; @@ -3882,7 +4038,7 @@ static void rocker_tx_desc_frags_unmap(struct rocker_port *rocker_port, if (!attrs[ROCKER_TLV_TX_FRAGS]) return; rocker_tlv_for_each_nested(attr, attrs[ROCKER_TLV_TX_FRAGS], rem) { - struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1]; + const struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1]; dma_addr_t dma_handle; size_t len; @@ -3899,11 +4055,11 @@ static void rocker_tx_desc_frags_unmap(struct rocker_port *rocker_port, } } -static int rocker_tx_desc_frag_map_put(struct rocker_port *rocker_port, +static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, char *buf, size_t buf_len) { - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; struct pci_dev *pdev = rocker->pdev; dma_addr_t dma_handle; struct rocker_tlv *frag; @@ -4008,269 +4164,333 @@ static int rocker_port_set_mac_address(struct net_device *dev, void *p) return 0; } -static int rocker_port_vlan_rx_add_vid(struct net_device *dev, - __be16 proto, u16 vid) +static int rocker_port_get_phys_port_name(struct net_device *dev, + char *buf, size_t len) { struct rocker_port *rocker_port = netdev_priv(dev); + struct port_name name = { .buf = buf, .len = len }; int err; - err = rocker_port_vlan(rocker_port, 0, vid); - if (err) - return err; + err = rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, + rocker_cmd_get_port_settings_prep, NULL, + rocker_cmd_get_port_settings_phys_name_proc, + &name); - return rocker_port_router_mac(rocker_port, 0, htons(vid)); + return err ? -EOPNOTSUPP : 0; } -static int rocker_port_vlan_rx_kill_vid(struct net_device *dev, - __be16 proto, u16 vid) +static const struct net_device_ops rocker_port_netdev_ops = { + .ndo_open = rocker_port_open, + .ndo_stop = rocker_port_stop, + .ndo_start_xmit = rocker_port_xmit, + .ndo_set_mac_address = rocker_port_set_mac_address, + .ndo_bridge_getlink = switchdev_port_bridge_getlink, + .ndo_bridge_setlink = switchdev_port_bridge_setlink, + .ndo_bridge_dellink = switchdev_port_bridge_dellink, + .ndo_fdb_add = switchdev_port_fdb_add, + .ndo_fdb_del = switchdev_port_fdb_del, + .ndo_fdb_dump = switchdev_port_fdb_dump, + .ndo_get_phys_port_name = rocker_port_get_phys_port_name, +}; + +/******************** + * swdev interface + ********************/ + +static int rocker_port_attr_get(struct net_device *dev, + struct switchdev_attr *attr) { - struct rocker_port *rocker_port = netdev_priv(dev); - int err; + const struct rocker_port *rocker_port = netdev_priv(dev); + const struct rocker *rocker = rocker_port->rocker; - err = rocker_port_router_mac(rocker_port, ROCKER_OP_FLAG_REMOVE, - htons(vid)); - if (err) - return err; + switch (attr->id) { + case SWITCHDEV_ATTR_PORT_PARENT_ID: + attr->u.ppid.id_len = sizeof(rocker->hw.id); + memcpy(&attr->u.ppid.id, &rocker->hw.id, attr->u.ppid.id_len); + break; + case SWITCHDEV_ATTR_PORT_BRIDGE_FLAGS: + attr->u.brport_flags = rocker_port->brport_flags; + break; + default: + return -EOPNOTSUPP; + } - return rocker_port_vlan(rocker_port, ROCKER_OP_FLAG_REMOVE, vid); + return 0; } -static int rocker_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, - const unsigned char *addr, u16 vid, - u16 nlm_flags) +static void rocker_port_trans_abort(const struct rocker_port *rocker_port) { - struct rocker_port *rocker_port = netdev_priv(dev); - __be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, NULL); - int flags = 0; + struct list_head *mem, *tmp; - if (!rocker_port_is_bridged(rocker_port)) - return -EINVAL; - - return rocker_port_fdb(rocker_port, addr, vlan_id, flags); + list_for_each_safe(mem, tmp, &rocker_port->trans_mem) { + list_del(mem); + kfree(mem); + } } -static int rocker_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, - const unsigned char *addr, u16 vid) +static int rocker_port_brport_flags_set(struct rocker_port *rocker_port, + enum switchdev_trans trans, + unsigned long brport_flags) { - struct rocker_port *rocker_port = netdev_priv(dev); - __be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, NULL); - int flags = ROCKER_OP_FLAG_REMOVE; + unsigned long orig_flags; + int err = 0; - if (!rocker_port_is_bridged(rocker_port)) - return -EINVAL; + orig_flags = rocker_port->brport_flags; + rocker_port->brport_flags = brport_flags; + if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING) + err = rocker_port_set_learning(rocker_port, trans); - return rocker_port_fdb(rocker_port, addr, vlan_id, flags); + if (trans == SWITCHDEV_TRANS_PREPARE) + rocker_port->brport_flags = orig_flags; + + return err; } -static int rocker_fdb_fill_info(struct sk_buff *skb, - struct rocker_port *rocker_port, - const unsigned char *addr, u16 vid, - u32 portid, u32 seq, int type, - unsigned int flags) +static int rocker_port_attr_set(struct net_device *dev, + struct switchdev_attr *attr) { - struct nlmsghdr *nlh; - struct ndmsg *ndm; + struct rocker_port *rocker_port = netdev_priv(dev); + int err = 0; - nlh = nlmsg_put(skb, portid, seq, type, sizeof(*ndm), flags); - if (!nlh) - return -EMSGSIZE; + switch (attr->trans) { + case SWITCHDEV_TRANS_PREPARE: + BUG_ON(!list_empty(&rocker_port->trans_mem)); + break; + case SWITCHDEV_TRANS_ABORT: + rocker_port_trans_abort(rocker_port); + return 0; + default: + break; + } - ndm = nlmsg_data(nlh); - ndm->ndm_family = AF_BRIDGE; - ndm->ndm_pad1 = 0; - ndm->ndm_pad2 = 0; - ndm->ndm_flags = NTF_SELF; - ndm->ndm_type = 0; - ndm->ndm_ifindex = rocker_port->dev->ifindex; - ndm->ndm_state = NUD_REACHABLE; + switch (attr->id) { + case SWITCHDEV_ATTR_PORT_STP_STATE: + err = rocker_port_stp_update(rocker_port, attr->trans, + attr->u.stp_state); + break; + case SWITCHDEV_ATTR_PORT_BRIDGE_FLAGS: + err = rocker_port_brport_flags_set(rocker_port, attr->trans, + attr->u.brport_flags); + break; + default: + err = -EOPNOTSUPP; + break; + } - if (nla_put(skb, NDA_LLADDR, ETH_ALEN, addr)) - goto nla_put_failure; + return err; +} - if (vid && nla_put_u16(skb, NDA_VLAN, vid)) - goto nla_put_failure; +static int rocker_port_vlan_add(struct rocker_port *rocker_port, + enum switchdev_trans trans, u16 vid, u16 flags) +{ + int err; - nlmsg_end(skb, nlh); - return 0; + /* XXX deal with flags for PVID and untagged */ -nla_put_failure: - nlmsg_cancel(skb, nlh); - return -EMSGSIZE; + err = rocker_port_vlan(rocker_port, trans, 0, vid); + if (err) + return err; + + err = rocker_port_router_mac(rocker_port, trans, 0, htons(vid)); + if (err) + rocker_port_vlan(rocker_port, trans, + ROCKER_OP_FLAG_REMOVE, vid); + + return err; } -static int rocker_port_fdb_dump(struct sk_buff *skb, - struct netlink_callback *cb, - struct net_device *dev, - struct net_device *filter_dev, - int idx) +static int rocker_port_vlans_add(struct rocker_port *rocker_port, + enum switchdev_trans trans, + const struct switchdev_obj_vlan *vlan) { - struct rocker_port *rocker_port = netdev_priv(dev); - struct rocker *rocker = rocker_port->rocker; - struct rocker_fdb_tbl_entry *found; - struct hlist_node *tmp; - int bkt; - unsigned long lock_flags; - const unsigned char *addr; u16 vid; int err; - spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags); - hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) { - if (found->key.pport != rocker_port->pport) - continue; - if (idx < cb->args[0]) - goto skip; - addr = found->key.addr; - vid = rocker_port_vlan_to_vid(rocker_port, found->key.vlan_id); - err = rocker_fdb_fill_info(skb, rocker_port, addr, vid, - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, - RTM_NEWNEIGH, NLM_F_MULTI); - if (err < 0) - break; -skip: - ++idx; + for (vid = vlan->vid_start; vid <= vlan->vid_end; vid++) { + err = rocker_port_vlan_add(rocker_port, trans, + vid, vlan->flags); + if (err) + return err; } - spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags); - return idx; + + return 0; } -static int rocker_port_bridge_setlink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) +static int rocker_port_fdb_add(struct rocker_port *rocker_port, + enum switchdev_trans trans, + const struct switchdev_obj_fdb *fdb) { - struct rocker_port *rocker_port = netdev_priv(dev); - struct nlattr *protinfo; - struct nlattr *attr; - int err; + __be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL); + int flags = 0; - protinfo = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_PROTINFO); - if (protinfo) { - attr = nla_find_nested(protinfo, IFLA_BRPORT_LEARNING); - if (attr) { - if (nla_len(attr) < sizeof(u8)) - return -EINVAL; - - if (nla_get_u8(attr)) - rocker_port->brport_flags |= BR_LEARNING; - else - rocker_port->brport_flags &= ~BR_LEARNING; - err = rocker_port_set_learning(rocker_port); - if (err) - return err; - } - attr = nla_find_nested(protinfo, IFLA_BRPORT_LEARNING_SYNC); - if (attr) { - if (nla_len(attr) < sizeof(u8)) - return -EINVAL; - - if (nla_get_u8(attr)) - rocker_port->brport_flags |= BR_LEARNING_SYNC; - else - rocker_port->brport_flags &= ~BR_LEARNING_SYNC; - } - } + if (!rocker_port_is_bridged(rocker_port)) + return -EINVAL; - return 0; + return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags); } -static int rocker_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, - u32 filter_mask, int nlflags) +static int rocker_port_obj_add(struct net_device *dev, + struct switchdev_obj *obj) { struct rocker_port *rocker_port = netdev_priv(dev); - u16 mode = BRIDGE_MODE_UNDEF; - u32 mask = BR_LEARNING | BR_LEARNING_SYNC; + const struct switchdev_obj_ipv4_fib *fib4; + int err = 0; + + switch (obj->trans) { + case SWITCHDEV_TRANS_PREPARE: + BUG_ON(!list_empty(&rocker_port->trans_mem)); + break; + case SWITCHDEV_TRANS_ABORT: + rocker_port_trans_abort(rocker_port); + return 0; + default: + break; + } + + switch (obj->id) { + case SWITCHDEV_OBJ_PORT_VLAN: + err = rocker_port_vlans_add(rocker_port, obj->trans, + &obj->u.vlan); + break; + case SWITCHDEV_OBJ_IPV4_FIB: + fib4 = &obj->u.ipv4_fib; + err = rocker_port_fib_ipv4(rocker_port, obj->trans, + htonl(fib4->dst), fib4->dst_len, + fib4->fi, fib4->tb_id, 0); + break; + case SWITCHDEV_OBJ_PORT_FDB: + err = rocker_port_fdb_add(rocker_port, obj->trans, &obj->u.fdb); + break; + default: + err = -EOPNOTSUPP; + break; + } - return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, - rocker_port->brport_flags, mask, - nlflags); + return err; } -static int rocker_port_get_phys_port_name(struct net_device *dev, - char *buf, size_t len) +static int rocker_port_vlan_del(struct rocker_port *rocker_port, + u16 vid, u16 flags) { - struct rocker_port *rocker_port = netdev_priv(dev); - struct port_name name = { .buf = buf, .len = len }; int err; - err = rocker_cmd_exec(rocker_port->rocker, rocker_port, - rocker_cmd_get_port_settings_prep, NULL, - rocker_cmd_get_port_settings_phys_name_proc, - &name, false); + err = rocker_port_router_mac(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, htons(vid)); + if (err) + return err; - return err ? -EOPNOTSUPP : 0; + return rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, vid); } -static const struct net_device_ops rocker_port_netdev_ops = { - .ndo_open = rocker_port_open, - .ndo_stop = rocker_port_stop, - .ndo_start_xmit = rocker_port_xmit, - .ndo_set_mac_address = rocker_port_set_mac_address, - .ndo_vlan_rx_add_vid = rocker_port_vlan_rx_add_vid, - .ndo_vlan_rx_kill_vid = rocker_port_vlan_rx_kill_vid, - .ndo_fdb_add = rocker_port_fdb_add, - .ndo_fdb_del = rocker_port_fdb_del, - .ndo_fdb_dump = rocker_port_fdb_dump, - .ndo_bridge_setlink = rocker_port_bridge_setlink, - .ndo_bridge_getlink = rocker_port_bridge_getlink, - .ndo_get_phys_port_name = rocker_port_get_phys_port_name, -}; +static int rocker_port_vlans_del(struct rocker_port *rocker_port, + const struct switchdev_obj_vlan *vlan) +{ + u16 vid; + int err; -/******************** - * swdev interface - ********************/ + for (vid = vlan->vid_start; vid <= vlan->vid_end; vid++) { + err = rocker_port_vlan_del(rocker_port, vid, vlan->flags); + if (err) + return err; + } + + return 0; +} -static int rocker_port_swdev_parent_id_get(struct net_device *dev, - struct netdev_phys_item_id *psid) +static int rocker_port_fdb_del(struct rocker_port *rocker_port, + enum switchdev_trans trans, + const struct switchdev_obj_fdb *fdb) { - struct rocker_port *rocker_port = netdev_priv(dev); - struct rocker *rocker = rocker_port->rocker; + __be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL); + int flags = ROCKER_OP_FLAG_REMOVE; - psid->id_len = sizeof(rocker->hw.id); - memcpy(&psid->id, &rocker->hw.id, psid->id_len); - return 0; + if (!rocker_port_is_bridged(rocker_port)) + return -EINVAL; + + return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags); } -static int rocker_port_swdev_port_stp_update(struct net_device *dev, u8 state) +static int rocker_port_obj_del(struct net_device *dev, + struct switchdev_obj *obj) { struct rocker_port *rocker_port = netdev_priv(dev); + const struct switchdev_obj_ipv4_fib *fib4; + int err = 0; - return rocker_port_stp_update(rocker_port, state); + switch (obj->id) { + case SWITCHDEV_OBJ_PORT_VLAN: + err = rocker_port_vlans_del(rocker_port, &obj->u.vlan); + break; + case SWITCHDEV_OBJ_IPV4_FIB: + fib4 = &obj->u.ipv4_fib; + err = rocker_port_fib_ipv4(rocker_port, SWITCHDEV_TRANS_NONE, + htonl(fib4->dst), fib4->dst_len, + fib4->fi, fib4->tb_id, + ROCKER_OP_FLAG_REMOVE); + break; + case SWITCHDEV_OBJ_PORT_FDB: + err = rocker_port_fdb_del(rocker_port, obj->trans, &obj->u.fdb); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; } -static int rocker_port_swdev_fib_ipv4_add(struct net_device *dev, - __be32 dst, int dst_len, - struct fib_info *fi, - u8 tos, u8 type, - u32 nlflags, u32 tb_id) +static int rocker_port_fdb_dump(const struct rocker_port *rocker_port, + struct switchdev_obj *obj) { - struct rocker_port *rocker_port = netdev_priv(dev); - int flags = 0; + struct rocker *rocker = rocker_port->rocker; + struct switchdev_obj_fdb *fdb = &obj->u.fdb; + struct rocker_fdb_tbl_entry *found; + struct hlist_node *tmp; + unsigned long lock_flags; + int bkt; + int err = 0; + + spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags); + hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) { + if (found->key.pport != rocker_port->pport) + continue; + fdb->addr = found->key.addr; + fdb->vid = rocker_port_vlan_to_vid(rocker_port, + found->key.vlan_id); + err = obj->cb(rocker_port->dev, obj); + if (err) + break; + } + spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags); - return rocker_port_fib_ipv4(rocker_port, dst, dst_len, - fi, tb_id, flags); + return err; } -static int rocker_port_swdev_fib_ipv4_del(struct net_device *dev, - __be32 dst, int dst_len, - struct fib_info *fi, - u8 tos, u8 type, u32 tb_id) +static int rocker_port_obj_dump(struct net_device *dev, + struct switchdev_obj *obj) { - struct rocker_port *rocker_port = netdev_priv(dev); - int flags = ROCKER_OP_FLAG_REMOVE; + const struct rocker_port *rocker_port = netdev_priv(dev); + int err = 0; - return rocker_port_fib_ipv4(rocker_port, dst, dst_len, - fi, tb_id, flags); + switch (obj->id) { + case SWITCHDEV_OBJ_PORT_FDB: + err = rocker_port_fdb_dump(rocker_port, obj); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; } -static const struct swdev_ops rocker_port_swdev_ops = { - .swdev_parent_id_get = rocker_port_swdev_parent_id_get, - .swdev_port_stp_update = rocker_port_swdev_port_stp_update, - .swdev_fib_ipv4_add = rocker_port_swdev_fib_ipv4_add, - .swdev_fib_ipv4_del = rocker_port_swdev_fib_ipv4_del, +static const struct switchdev_ops rocker_port_switchdev_ops = { + .switchdev_port_attr_get = rocker_port_attr_get, + .switchdev_port_attr_set = rocker_port_attr_set, + .switchdev_port_obj_add = rocker_port_obj_add, + .switchdev_port_obj_del = rocker_port_obj_del, + .switchdev_port_obj_dump = rocker_port_obj_dump, }; /******************** @@ -4334,8 +4554,7 @@ static void rocker_port_get_strings(struct net_device *netdev, u32 stringset, } static int -rocker_cmd_get_port_stats_prep(struct rocker *rocker, - struct rocker_port *rocker_port, +rocker_cmd_get_port_stats_prep(const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info, void *priv) { @@ -4359,14 +4578,13 @@ rocker_cmd_get_port_stats_prep(struct rocker *rocker, } static int -rocker_cmd_get_port_stats_ethtool_proc(struct rocker *rocker, - struct rocker_port *rocker_port, - struct rocker_desc_info *desc_info, +rocker_cmd_get_port_stats_ethtool_proc(const struct rocker_port *rocker_port, + const struct rocker_desc_info *desc_info, void *priv) { - struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; - struct rocker_tlv *stats_attrs[ROCKER_TLV_CMD_PORT_STATS_MAX + 1]; - struct rocker_tlv *pattr; + const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1]; + const struct rocker_tlv *stats_attrs[ROCKER_TLV_CMD_PORT_STATS_MAX + 1]; + const struct rocker_tlv *pattr; u32 pport; u64 *data = priv; int i; @@ -4400,10 +4618,10 @@ rocker_cmd_get_port_stats_ethtool_proc(struct rocker *rocker, static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port, void *priv) { - return rocker_cmd_exec(rocker_port->rocker, rocker_port, + return rocker_cmd_exec(rocker_port, SWITCHDEV_TRANS_NONE, rocker_cmd_get_port_stats_prep, NULL, rocker_cmd_get_port_stats_ethtool_proc, - priv, false); + priv); } static void rocker_port_get_stats(struct net_device *dev, @@ -4417,8 +4635,6 @@ static void rocker_port_get_stats(struct net_device *dev, for (i = 0; i < ARRAY_SIZE(rocker_port_stats); ++i) data[i] = 0; } - - return; } static int rocker_port_get_sset_count(struct net_device *netdev, int sset) @@ -4453,8 +4669,8 @@ static struct rocker_port *rocker_port_napi_tx_get(struct napi_struct *napi) static int rocker_port_poll_tx(struct napi_struct *napi, int budget) { struct rocker_port *rocker_port = rocker_port_napi_tx_get(napi); - struct rocker *rocker = rocker_port->rocker; - struct rocker_desc_info *desc_info; + const struct rocker *rocker = rocker_port->rocker; + const struct rocker_desc_info *desc_info; u32 credits = 0; int err; @@ -4472,8 +4688,9 @@ static int rocker_port_poll_tx(struct napi_struct *napi, int budget) if (err == 0) { rocker_port->dev->stats.tx_packets++; rocker_port->dev->stats.tx_bytes += skb->len; - } else + } else { rocker_port->dev->stats.tx_errors++; + } dev_kfree_skb_any(skb); credits++; @@ -4488,11 +4705,11 @@ static int rocker_port_poll_tx(struct napi_struct *napi, int budget) return 0; } -static int rocker_port_rx_proc(struct rocker *rocker, - struct rocker_port *rocker_port, +static int rocker_port_rx_proc(const struct rocker *rocker, + const struct rocker_port *rocker_port, struct rocker_desc_info *desc_info) { - struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1]; + const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1]; struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info); size_t rx_len; @@ -4514,7 +4731,7 @@ static int rocker_port_rx_proc(struct rocker *rocker, netif_receive_skb(skb); - return rocker_dma_rx_ring_skb_alloc(rocker, rocker_port, desc_info); + return rocker_dma_rx_ring_skb_alloc(rocker_port, desc_info); } static struct rocker_port *rocker_port_napi_rx_get(struct napi_struct *napi) @@ -4525,7 +4742,7 @@ static struct rocker_port *rocker_port_napi_rx_get(struct napi_struct *napi) static int rocker_port_poll_rx(struct napi_struct *napi, int budget) { struct rocker_port *rocker_port = rocker_port_napi_rx_get(napi); - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; struct rocker_desc_info *desc_info; u32 credits = 0; int err; @@ -4565,9 +4782,9 @@ static int rocker_port_poll_rx(struct napi_struct *napi, int budget) * PCI driver ops *****************/ -static void rocker_carrier_init(struct rocker_port *rocker_port) +static void rocker_carrier_init(const struct rocker_port *rocker_port) { - struct rocker *rocker = rocker_port->rocker; + const struct rocker *rocker = rocker_port->rocker; u64 link_status = rocker_read64(rocker, PORT_PHYS_LINK_STATUS); bool link_up; @@ -4578,23 +4795,24 @@ static void rocker_carrier_init(struct rocker_port *rocker_port) netif_carrier_off(rocker_port->dev); } -static void rocker_remove_ports(struct rocker *rocker) +static void rocker_remove_ports(const struct rocker *rocker) { struct rocker_port *rocker_port; int i; for (i = 0; i < rocker->port_count; i++) { rocker_port = rocker->ports[i]; - rocker_port_ig_tbl(rocker_port, ROCKER_OP_FLAG_REMOVE); + rocker_port_ig_tbl(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE); unregister_netdev(rocker_port->dev); } kfree(rocker->ports); } -static void rocker_port_dev_addr_init(struct rocker *rocker, - struct rocker_port *rocker_port) +static void rocker_port_dev_addr_init(struct rocker_port *rocker_port) { - struct pci_dev *pdev = rocker->pdev; + const struct rocker *rocker = rocker_port->rocker; + const struct pci_dev *pdev = rocker->pdev; int err; err = rocker_cmd_get_port_settings_macaddr(rocker_port, @@ -4607,9 +4825,10 @@ static void rocker_port_dev_addr_init(struct rocker *rocker, static int rocker_probe_port(struct rocker *rocker, unsigned int port_number) { - struct pci_dev *pdev = rocker->pdev; + const struct pci_dev *pdev = rocker->pdev; struct rocker_port *rocker_port; struct net_device *dev; + u16 untagged_vid = 0; int err; dev = alloc_etherdev(sizeof(struct rocker_port)); @@ -4621,20 +4840,19 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number) rocker_port->port_number = port_number; rocker_port->pport = port_number + 1; rocker_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC; + INIT_LIST_HEAD(&rocker_port->trans_mem); - rocker_port_dev_addr_init(rocker, rocker_port); + rocker_port_dev_addr_init(rocker_port); dev->netdev_ops = &rocker_port_netdev_ops; dev->ethtool_ops = &rocker_port_ethtool_ops; - dev->swdev_ops = &rocker_port_swdev_ops; + dev->switchdev_ops = &rocker_port_switchdev_ops; netif_napi_add(dev, &rocker_port->napi_tx, rocker_port_poll_tx, NAPI_POLL_WEIGHT); netif_napi_add(dev, &rocker_port->napi_rx, rocker_port_poll_rx, NAPI_POLL_WEIGHT); rocker_carrier_init(rocker_port); - dev->features |= NETIF_F_NETNS_LOCAL | - NETIF_F_HW_VLAN_CTAG_FILTER | - NETIF_F_HW_SWITCH_OFFLOAD; + dev->features |= NETIF_F_NETNS_LOCAL; err = register_netdev(dev); if (err) { @@ -4643,18 +4861,29 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number) } rocker->ports[port_number] = rocker_port; - rocker_port_set_learning(rocker_port); + rocker_port_set_learning(rocker_port, SWITCHDEV_TRANS_NONE); - rocker_port->internal_vlan_id = - rocker_port_internal_vlan_id_get(rocker_port, dev->ifindex); - err = rocker_port_ig_tbl(rocker_port, 0); + err = rocker_port_ig_tbl(rocker_port, SWITCHDEV_TRANS_NONE, 0); if (err) { dev_err(&pdev->dev, "install ig port table failed\n"); goto err_port_ig_tbl; } + rocker_port->internal_vlan_id = + rocker_port_internal_vlan_id_get(rocker_port, dev->ifindex); + + err = rocker_port_vlan_add(rocker_port, SWITCHDEV_TRANS_NONE, + untagged_vid, 0); + if (err) { + netdev_err(rocker_port->dev, "install untagged VLAN failed\n"); + goto err_untagged_vlan; + } + return 0; +err_untagged_vlan: + rocker_port_ig_tbl(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE); err_port_ig_tbl: unregister_netdev(dev); err_register_netdev: @@ -4669,7 +4898,7 @@ static int rocker_probe_ports(struct rocker *rocker) int err; alloc_size = sizeof(struct rocker_port *) * rocker->port_count; - rocker->ports = kmalloc(alloc_size, GFP_KERNEL); + rocker->ports = kzalloc(alloc_size, GFP_KERNEL); if (!rocker->ports) return -ENOMEM; for (i = 0; i < rocker->port_count; i++) { @@ -4718,7 +4947,7 @@ err_enable_msix: return err; } -static void rocker_msix_fini(struct rocker *rocker) +static void rocker_msix_fini(const struct rocker *rocker) { pci_disable_msix(rocker->pdev); kfree(rocker->msix_entries); @@ -4884,7 +5113,7 @@ static struct pci_driver rocker_pci_driver = { * Net device notifier event handler ************************************/ -static bool rocker_port_dev_check(struct net_device *dev) +static bool rocker_port_dev_check(const struct net_device *dev) { return dev->netdev_ops == &rocker_port_netdev_ops; } @@ -4892,45 +5121,54 @@ static bool rocker_port_dev_check(struct net_device *dev) static int rocker_port_bridge_join(struct rocker_port *rocker_port, struct net_device *bridge) { + u16 untagged_vid = 0; int err; + /* Port is joining bridge, so the internal VLAN for the + * port is going to change to the bridge internal VLAN. + * Let's remove untagged VLAN (vid=0) from port and + * re-add once internal VLAN has changed. + */ + + err = rocker_port_vlan_del(rocker_port, untagged_vid, 0); + if (err) + return err; + rocker_port_internal_vlan_id_put(rocker_port, rocker_port->dev->ifindex); + rocker_port->internal_vlan_id = + rocker_port_internal_vlan_id_get(rocker_port, bridge->ifindex); rocker_port->bridge_dev = bridge; - /* Use bridge internal VLAN ID for untagged pkts */ - err = rocker_port_vlan(rocker_port, ROCKER_OP_FLAG_REMOVE, 0); - if (err) - return err; - rocker_port->internal_vlan_id = - rocker_port_internal_vlan_id_get(rocker_port, - bridge->ifindex); - return rocker_port_vlan(rocker_port, 0, 0); + return rocker_port_vlan_add(rocker_port, SWITCHDEV_TRANS_NONE, + untagged_vid, 0); } static int rocker_port_bridge_leave(struct rocker_port *rocker_port) { + u16 untagged_vid = 0; int err; - rocker_port_internal_vlan_id_put(rocker_port, - rocker_port->bridge_dev->ifindex); - - rocker_port->bridge_dev = NULL; - - /* Use port internal VLAN ID for untagged pkts */ - err = rocker_port_vlan(rocker_port, ROCKER_OP_FLAG_REMOVE, 0); + err = rocker_port_vlan_del(rocker_port, untagged_vid, 0); if (err) return err; + + rocker_port_internal_vlan_id_put(rocker_port, + rocker_port->bridge_dev->ifindex); rocker_port->internal_vlan_id = rocker_port_internal_vlan_id_get(rocker_port, rocker_port->dev->ifindex); - err = rocker_port_vlan(rocker_port, 0, 0); + + rocker_port->bridge_dev = NULL; + + err = rocker_port_vlan_add(rocker_port, SWITCHDEV_TRANS_NONE, + untagged_vid, 0); if (err) return err; if (rocker_port->dev->flags & IFF_UP) - err = rocker_port_fwd_enable(rocker_port); + err = rocker_port_fwd_enable(rocker_port, SWITCHDEV_TRANS_NONE); return err; } @@ -4992,7 +5230,8 @@ static int rocker_neigh_update(struct net_device *dev, struct neighbour *n) int flags = (n->nud_state & NUD_VALID) ? 0 : ROCKER_OP_FLAG_REMOVE; __be32 ip_addr = *(__be32 *)n->primary_key; - return rocker_port_ipv4_neigh(rocker_port, flags, ip_addr, n->ha); + return rocker_port_ipv4_neigh(rocker_port, SWITCHDEV_TRANS_NONE, + flags, ip_addr, n->ha); } static int rocker_netevent_event(struct notifier_block *unused, diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h index a4e9591d7457..c61fbf968036 100644 --- a/drivers/net/ethernet/rocker/rocker.h +++ b/drivers/net/ethernet/rocker/rocker.h @@ -65,9 +65,9 @@ enum { #define ROCKER_TEST_DMA_CTRL 0x0034 /* Rocker test register ctrl */ -#define ROCKER_TEST_DMA_CTRL_CLEAR (1 << 0) -#define ROCKER_TEST_DMA_CTRL_FILL (1 << 1) -#define ROCKER_TEST_DMA_CTRL_INVERT (1 << 2) +#define ROCKER_TEST_DMA_CTRL_CLEAR BIT(0) +#define ROCKER_TEST_DMA_CTRL_FILL BIT(1) +#define ROCKER_TEST_DMA_CTRL_INVERT BIT(2) /* Rocker DMA ring register offsets */ #define ROCKER_DMA_DESC_ADDR(x) (0x1000 + (x) * 32) /* 8-byte */ @@ -79,7 +79,7 @@ enum { #define ROCKER_DMA_DESC_RES1(x) (0x101c + (x) * 32) /* Rocker dma ctrl register bits */ -#define ROCKER_DMA_DESC_CTRL_RESET (1 << 0) +#define ROCKER_DMA_DESC_CTRL_RESET BIT(0) /* Rocker DMA ring types */ enum rocker_dma_type { @@ -111,7 +111,7 @@ struct rocker_desc { u16 comp_err; }; -#define ROCKER_DMA_DESC_COMP_ERR_GEN (1 << 15) +#define ROCKER_DMA_DESC_COMP_ERR_GEN BIT(15) /* Rocker DMA TLV struct */ struct rocker_tlv { @@ -237,14 +237,14 @@ enum { ROCKER_TLV_RX_MAX = __ROCKER_TLV_RX_MAX - 1, }; -#define ROCKER_RX_FLAGS_IPV4 (1 << 0) -#define ROCKER_RX_FLAGS_IPV6 (1 << 1) -#define ROCKER_RX_FLAGS_CSUM_CALC (1 << 2) -#define ROCKER_RX_FLAGS_IPV4_CSUM_GOOD (1 << 3) -#define ROCKER_RX_FLAGS_IP_FRAG (1 << 4) -#define ROCKER_RX_FLAGS_TCP (1 << 5) -#define ROCKER_RX_FLAGS_UDP (1 << 6) -#define ROCKER_RX_FLAGS_TCP_UDP_CSUM_GOOD (1 << 7) +#define ROCKER_RX_FLAGS_IPV4 BIT(0) +#define ROCKER_RX_FLAGS_IPV6 BIT(1) +#define ROCKER_RX_FLAGS_CSUM_CALC BIT(2) +#define ROCKER_RX_FLAGS_IPV4_CSUM_GOOD BIT(3) +#define ROCKER_RX_FLAGS_IP_FRAG BIT(4) +#define ROCKER_RX_FLAGS_TCP BIT(5) +#define ROCKER_RX_FLAGS_UDP BIT(6) +#define ROCKER_RX_FLAGS_TCP_UDP_CSUM_GOOD BIT(7) enum { ROCKER_TLV_TX_UNSPEC, @@ -460,6 +460,6 @@ enum rocker_of_dpa_overlay_type { #define ROCKER_SWITCH_ID 0x0320 /* 8-byte */ /* Rocker control bits */ -#define ROCKER_CONTROL_RESET (1 << 0) +#define ROCKER_CONTROL_RESET BIT(0) #endif diff --git a/drivers/net/ethernet/sfc/Kconfig b/drivers/net/ethernet/sfc/Kconfig index 088921294448..4dd92b7b80f4 100644 --- a/drivers/net/ethernet/sfc/Kconfig +++ b/drivers/net/ethernet/sfc/Kconfig @@ -36,3 +36,12 @@ config SFC_SRIOV This enables support for the SFC9000 I/O Virtualization features, allowing accelerated network performance in virtualized environments. +config SFC_MCDI_LOGGING + bool "Solarflare SFC9000/SFC9100-family MCDI logging support" + depends on SFC + default y + ---help--- + This enables support for tracing of MCDI (Management-Controller-to- + Driver-Interface) commands and responses, allowing debugging of + driver/firmware interaction. The tracing is actually enabled by + a sysfs file 'mcdi_logging' under the PCI device. diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile index 3a83c0dca8e6..ce8470fe79d5 100644 --- a/drivers/net/ethernet/sfc/Makefile +++ b/drivers/net/ethernet/sfc/Makefile @@ -3,6 +3,6 @@ sfc-y += efx.o nic.o farch.o falcon.o siena.o ef10.o tx.o \ tenxpress.o txc43128_phy.o falcon_boards.o \ mcdi.o mcdi_port.o mcdi_mon.o ptp.o sfc-$(CONFIG_SFC_MTD) += mtd.o -sfc-$(CONFIG_SFC_SRIOV) += siena_sriov.o +sfc-$(CONFIG_SFC_SRIOV) += sriov.o siena_sriov.o ef10_sriov.o obj-$(CONFIG_SFC) += sfc.o diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index fbb6cfa0f5f1..4eb6ab748f06 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -15,6 +15,7 @@ #include "nic.h" #include "workarounds.h" #include "selftest.h" +#include "ef10_sriov.h" #include <linux/in.h> #include <linux/jhash.h> #include <linux/wait.h> @@ -30,6 +31,9 @@ enum { /* The reserved RSS context value */ #define EFX_EF10_RSS_CONTEXT_INVALID 0xffffffff +/* The maximum size of a shared RSS context */ +/* TODO: this should really be from the mcdi protocol export */ +#define EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE 64UL /* The filter table(s) are managed by firmware and we have write-only * access. When removing filters we must identify them to the @@ -77,7 +81,6 @@ struct efx_ef10_filter_table { /* An arbitrary search limit for the software hash table */ #define EFX_EF10_FILTER_SEARCH_LIMIT 200 -static void efx_ef10_rx_push_rss_config(struct efx_nic *efx); static void efx_ef10_rx_free_indir_table(struct efx_nic *efx); static void efx_ef10_filter_table_remove(struct efx_nic *efx); @@ -92,9 +95,50 @@ static int efx_ef10_get_warm_boot_count(struct efx_nic *efx) static unsigned int efx_ef10_mem_map_size(struct efx_nic *efx) { - return resource_size(&efx->pci_dev->resource[EFX_MEM_BAR]); + int bar; + + bar = efx->type->mem_bar; + return resource_size(&efx->pci_dev->resource[bar]); +} + +static int efx_ef10_get_pf_index(struct efx_nic *efx) +{ + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_FUNCTION_INFO_OUT_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + size_t outlen; + int rc; + + rc = efx_mcdi_rpc(efx, MC_CMD_GET_FUNCTION_INFO, NULL, 0, outbuf, + sizeof(outbuf), &outlen); + if (rc) + return rc; + if (outlen < sizeof(outbuf)) + return -EIO; + + nic_data->pf_index = MCDI_DWORD(outbuf, GET_FUNCTION_INFO_OUT_PF); + return 0; } +#ifdef CONFIG_SFC_SRIOV +static int efx_ef10_get_vf_index(struct efx_nic *efx) +{ + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_FUNCTION_INFO_OUT_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + size_t outlen; + int rc; + + rc = efx_mcdi_rpc(efx, MC_CMD_GET_FUNCTION_INFO, NULL, 0, outbuf, + sizeof(outbuf), &outlen); + if (rc) + return rc; + if (outlen < sizeof(outbuf)) + return -EIO; + + nic_data->vf_index = MCDI_DWORD(outbuf, GET_FUNCTION_INFO_OUT_VF); + return 0; +} +#endif + static int efx_ef10_init_datapath_caps(struct efx_nic *efx) { MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_CAPABILITIES_OUT_LEN); @@ -117,6 +161,13 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx) nic_data->datapath_caps = MCDI_DWORD(outbuf, GET_CAPABILITIES_OUT_FLAGS1); + /* record the DPCPU firmware IDs to determine VEB vswitching support. + */ + nic_data->rx_dpcpu_fw_id = + MCDI_WORD(outbuf, GET_CAPABILITIES_OUT_RX_DPCPU_FW_ID); + nic_data->tx_dpcpu_fw_id = + MCDI_WORD(outbuf, GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID); + if (!(nic_data->datapath_caps & (1 << MC_CMD_GET_CAPABILITIES_OUT_TX_TSO_LBN))) { netif_err(efx, drv, efx->net_dev, @@ -147,7 +198,7 @@ static int efx_ef10_get_sysclk_freq(struct efx_nic *efx) return rc > 0 ? rc : -ERANGE; } -static int efx_ef10_get_mac_address(struct efx_nic *efx, u8 *mac_address) +static int efx_ef10_get_mac_address_pf(struct efx_nic *efx, u8 *mac_address) { MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_MAC_ADDRESSES_OUT_LEN); size_t outlen; @@ -167,6 +218,34 @@ static int efx_ef10_get_mac_address(struct efx_nic *efx, u8 *mac_address) return 0; } +static int efx_ef10_get_mac_address_vf(struct efx_nic *efx, u8 *mac_address) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_GET_MAC_ADDRESSES_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_VPORT_GET_MAC_ADDRESSES_OUT_LENMAX); + size_t outlen; + int num_addrs, rc; + + MCDI_SET_DWORD(inbuf, VPORT_GET_MAC_ADDRESSES_IN_VPORT_ID, + EVB_PORT_ID_ASSIGNED); + rc = efx_mcdi_rpc(efx, MC_CMD_VPORT_GET_MAC_ADDRESSES, inbuf, + sizeof(inbuf), outbuf, sizeof(outbuf), &outlen); + + if (rc) + return rc; + if (outlen < MC_CMD_VPORT_GET_MAC_ADDRESSES_OUT_LENMIN) + return -EIO; + + num_addrs = MCDI_DWORD(outbuf, + VPORT_GET_MAC_ADDRESSES_OUT_MACADDR_COUNT); + + WARN_ON(num_addrs != 1); + + ether_addr_copy(mac_address, + MCDI_PTR(outbuf, VPORT_GET_MAC_ADDRESSES_OUT_MACADDR)); + + return 0; +} + static int efx_ef10_probe(struct efx_nic *efx) { struct efx_ef10_nic_data *nic_data; @@ -178,7 +257,7 @@ static int efx_ef10_probe(struct efx_nic *efx) efx->max_channels = min_t(unsigned int, EFX_MAX_CHANNELS, - resource_size(&efx->pci_dev->resource[EFX_MEM_BAR]) / + efx_ef10_mem_map_size(efx) / (EFX_VI_PAGE_SIZE * EFX_TXQ_TYPES)); if (WARN_ON(efx->max_channels == 0)) return -EIO; @@ -188,6 +267,9 @@ static int efx_ef10_probe(struct efx_nic *efx) return -ENOMEM; efx->nic_data = nic_data; + /* we assume later that we can copy from this buffer in dwords */ + BUILD_BUG_ON(MCDI_CTL_SDU_LEN_MAX_V2 % 4); + rc = efx_nic_alloc_buffer(efx, &nic_data->mcdi_buf, 8 + MCDI_CTL_SDU_LEN_MAX_V2, GFP_KERNEL); if (rc) @@ -209,6 +291,8 @@ static int efx_ef10_probe(struct efx_nic *efx) nic_data->rx_rss_context = EFX_EF10_RSS_CONTEXT_INVALID; + nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + /* In case we're recovering from a crash (kexec), we want to * cancel any outstanding request by the previous user of this * function. We send a special message using the least @@ -230,6 +314,10 @@ static int efx_ef10_probe(struct efx_nic *efx) if (rc) goto fail3; + rc = efx_ef10_get_pf_index(efx); + if (rc) + goto fail3; + rc = efx_ef10_init_datapath_caps(efx); if (rc < 0) goto fail3; @@ -242,7 +330,7 @@ static int efx_ef10_probe(struct efx_nic *efx) goto fail3; efx->port_num = rc; - rc = efx_ef10_get_mac_address(efx, efx->net_dev->perm_addr); + rc = efx->type->get_mac_address(efx, efx->net_dev->perm_addr); if (rc) goto fail3; @@ -251,10 +339,22 @@ static int efx_ef10_probe(struct efx_nic *efx) goto fail3; efx->timer_quantum_ns = 1536000 / rc; /* 1536 cycles */ - /* Check whether firmware supports bug 35388 workaround */ + /* Check whether firmware supports bug 35388 workaround. + * First try to enable it, then if we get EPERM, just + * ask if it's already enabled + */ rc = efx_mcdi_set_workaround(efx, MC_CMD_WORKAROUND_BUG35388, true); if (rc == 0) nic_data->workaround_35388 = true; + else if (rc == -EPERM) { + unsigned int enabled; + + rc = efx_mcdi_get_workarounds(efx, NULL, &enabled); + if (rc) + goto fail3; + nic_data->workaround_35388 = enabled & + MC_CMD_GET_WORKAROUNDS_OUT_BUG35388; + } else if (rc != -ENOSYS && rc != -ENOENT) goto fail3; netif_dbg(efx, probe, efx->net_dev, @@ -262,7 +362,7 @@ static int efx_ef10_probe(struct efx_nic *efx) nic_data->workaround_35388 ? "en" : "dis"); rc = efx_mcdi_mon_probe(efx); - if (rc) + if (rc && rc != -EPERM) goto fail3; efx_ptp_probe(efx, NULL); @@ -281,7 +381,7 @@ fail1: static int efx_ef10_free_vis(struct efx_nic *efx) { - MCDI_DECLARE_BUF_OUT_OR_ERR(outbuf, 0); + MCDI_DECLARE_BUF_ERR(outbuf); size_t outlen; int rc = efx_mcdi_rpc_quiet(efx, MC_CMD_FREE_VIS, NULL, 0, outbuf, sizeof(outbuf), &outlen); @@ -352,9 +452,9 @@ static int efx_ef10_alloc_piobufs(struct efx_nic *efx, unsigned int n) static int efx_ef10_link_piobufs(struct efx_nic *efx) { struct efx_ef10_nic_data *nic_data = efx->nic_data; - MCDI_DECLARE_BUF(inbuf, - max(MC_CMD_LINK_PIOBUF_IN_LEN, - MC_CMD_UNLINK_PIOBUF_IN_LEN)); + _MCDI_DECLARE_BUF(inbuf, + max(MC_CMD_LINK_PIOBUF_IN_LEN, + MC_CMD_UNLINK_PIOBUF_IN_LEN)); struct efx_channel *channel; struct efx_tx_queue *tx_queue; unsigned int offset, index; @@ -363,6 +463,8 @@ static int efx_ef10_link_piobufs(struct efx_nic *efx) BUILD_BUG_ON(MC_CMD_LINK_PIOBUF_OUT_LEN != 0); BUILD_BUG_ON(MC_CMD_UNLINK_PIOBUF_OUT_LEN != 0); + memset(inbuf, 0, sizeof(inbuf)); + /* Link a buffer to each VI in the write-combining mapping */ for (index = 0; index < nic_data->n_piobufs; ++index) { MCDI_SET_DWORD(inbuf, LINK_PIOBUF_IN_PIOBUF_HANDLE, @@ -475,6 +577,25 @@ static void efx_ef10_remove(struct efx_nic *efx) struct efx_ef10_nic_data *nic_data = efx->nic_data; int rc; +#ifdef CONFIG_SFC_SRIOV + struct efx_ef10_nic_data *nic_data_pf; + struct pci_dev *pci_dev_pf; + struct efx_nic *efx_pf; + struct ef10_vf *vf; + + if (efx->pci_dev->is_virtfn) { + pci_dev_pf = efx->pci_dev->physfn; + if (pci_dev_pf) { + efx_pf = pci_get_drvdata(pci_dev_pf); + nic_data_pf = efx_pf->nic_data; + vf = nic_data_pf->vf + nic_data->vf_index; + vf->efx = NULL; + } else + netif_info(efx, drv, efx->net_dev, + "Could not get the PF id from VF\n"); + } +#endif + efx_ptp_remove(efx); efx_mcdi_mon_remove(efx); @@ -495,6 +616,50 @@ static void efx_ef10_remove(struct efx_nic *efx) kfree(nic_data); } +static int efx_ef10_probe_pf(struct efx_nic *efx) +{ + return efx_ef10_probe(efx); +} + +#ifdef CONFIG_SFC_SRIOV +static int efx_ef10_probe_vf(struct efx_nic *efx) +{ + int rc; + + rc = efx_ef10_probe(efx); + if (rc) + return rc; + + rc = efx_ef10_get_vf_index(efx); + if (rc) + goto fail; + + if (efx->pci_dev->is_virtfn) { + if (efx->pci_dev->physfn) { + struct efx_nic *efx_pf = + pci_get_drvdata(efx->pci_dev->physfn); + struct efx_ef10_nic_data *nic_data_p = efx_pf->nic_data; + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + nic_data_p->vf[nic_data->vf_index].efx = efx; + } else + netif_info(efx, drv, efx->net_dev, + "Could not get the PF id from VF\n"); + } + + return 0; + +fail: + efx_ef10_remove(efx); + return rc; +} +#else +static int efx_ef10_probe_vf(struct efx_nic *efx __attribute__ ((unused))) +{ + return 0; +} +#endif + static int efx_ef10_alloc_vis(struct efx_nic *efx, unsigned int min_vis, unsigned int max_vis) { @@ -687,7 +852,9 @@ static int efx_ef10_init_nic(struct efx_nic *efx) nic_data->must_restore_piobufs = false; } - efx_ef10_rx_push_rss_config(efx); + /* don't fail init if RSS setup doesn't work */ + efx->type->rx_push_rss_config(efx, false, efx->rx_indir_table); + return 0; } @@ -702,6 +869,14 @@ static void efx_ef10_reset_mc_allocations(struct efx_nic *efx) nic_data->rx_rss_context = EFX_EF10_RSS_CONTEXT_INVALID; } +static enum reset_type efx_ef10_map_reset_reason(enum reset_type reason) +{ + if (reason == RESET_TYPE_MC_FAILURE) + return RESET_TYPE_DATAPATH; + + return efx_mcdi_map_reset_reason(reason); +} + static int efx_ef10_map_reset_flags(u32 *flags) { enum { @@ -1044,6 +1219,14 @@ static void efx_ef10_push_irq_moderation(struct efx_channel *channel) } } +static void efx_ef10_get_wol_vf(struct efx_nic *efx, + struct ethtool_wolinfo *wol) {} + +static int efx_ef10_set_wol_vf(struct efx_nic *efx, u32 type) +{ + return -EOPNOTSUPP; +} + static void efx_ef10_get_wol(struct efx_nic *efx, struct ethtool_wolinfo *wol) { wol->supported = 0; @@ -1123,6 +1306,10 @@ static int efx_ef10_mcdi_poll_reboot(struct efx_nic *efx) /* All our allocations have been reset */ efx_ef10_reset_mc_allocations(efx); + /* Driver-created vswitches and vports must be re-created */ + nic_data->must_probe_vswitching = true; + nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + /* The datapath firmware might have been changed */ nic_data->must_check_datapath_caps = true; @@ -1232,16 +1419,17 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) { MCDI_DECLARE_BUF(inbuf, MC_CMD_INIT_TXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 / EFX_BUF_SIZE)); - MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_TXQ_OUT_LEN); bool csum_offload = tx_queue->queue & EFX_TXQ_TYPE_OFFLOAD; size_t entries = tx_queue->txd.buf.len / EFX_BUF_SIZE; struct efx_channel *channel = tx_queue->channel; struct efx_nic *efx = tx_queue->efx; - size_t inlen, outlen; + struct efx_ef10_nic_data *nic_data = efx->nic_data; + size_t inlen; dma_addr_t dma_addr; efx_qword_t *txd; int rc; int i; + BUILD_BUG_ON(MC_CMD_INIT_TXQ_OUT_LEN != 0); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_SIZE, tx_queue->ptr_mask + 1); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_TARGET_EVQ, channel->channel); @@ -1251,7 +1439,7 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) INIT_TXQ_IN_FLAG_IP_CSUM_DIS, !csum_offload, INIT_TXQ_IN_FLAG_TCP_CSUM_DIS, !csum_offload); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_OWNER_ID, 0); - MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, EVB_PORT_ID_ASSIGNED); + MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, nic_data->vport_id); dma_addr = tx_queue->txd.buf.dma_addr; @@ -1266,7 +1454,7 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) inlen = MC_CMD_INIT_TXQ_IN_LEN(entries); rc = efx_mcdi_rpc(efx, MC_CMD_INIT_TXQ, inbuf, inlen, - outbuf, sizeof(outbuf), &outlen); + NULL, 0, NULL); if (rc) goto fail; @@ -1299,7 +1487,7 @@ fail: static void efx_ef10_tx_fini(struct efx_tx_queue *tx_queue) { MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_TXQ_IN_LEN); - MCDI_DECLARE_BUF(outbuf, MC_CMD_FINI_TXQ_OUT_LEN); + MCDI_DECLARE_BUF_ERR(outbuf); struct efx_nic *efx = tx_queue->efx; size_t outlen; int rc; @@ -1378,19 +1566,33 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue) } } -static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context) +static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context, + bool exclusive, unsigned *context_size) { MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN); MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; size_t outlen; int rc; + u32 alloc_type = exclusive ? + MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_EXCLUSIVE : + MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_SHARED; + unsigned rss_spread = exclusive ? + efx->rss_spread : + min(rounddown_pow_of_two(efx->rss_spread), + EFX_EF10_MAX_SHARED_RSS_CONTEXT_SIZE); + + if (!exclusive && rss_spread == 1) { + *context = EFX_EF10_RSS_CONTEXT_INVALID; + if (context_size) + *context_size = 1; + return 0; + } MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_UPSTREAM_PORT_ID, - EVB_PORT_ID_ASSIGNED); - MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_TYPE, - MC_CMD_RSS_CONTEXT_ALLOC_IN_TYPE_EXCLUSIVE); - MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_NUM_QUEUES, - EFX_MAX_CHANNELS); + nic_data->vport_id); + MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_TYPE, alloc_type); + MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_NUM_QUEUES, rss_spread); rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_ALLOC, inbuf, sizeof(inbuf), outbuf, sizeof(outbuf), &outlen); @@ -1402,6 +1604,9 @@ static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context) *context = MCDI_DWORD(outbuf, RSS_CONTEXT_ALLOC_OUT_RSS_CONTEXT_ID); + if (context_size) + *context_size = rss_spread; + return 0; } @@ -1418,7 +1623,8 @@ static void efx_ef10_free_rss_context(struct efx_nic *efx, u32 context) WARN_ON(rc != 0); } -static int efx_ef10_populate_rss_table(struct efx_nic *efx, u32 context) +static int efx_ef10_populate_rss_table(struct efx_nic *efx, u32 context, + const u32 *rx_indir_table) { MCDI_DECLARE_BUF(tablebuf, MC_CMD_RSS_CONTEXT_SET_TABLE_IN_LEN); MCDI_DECLARE_BUF(keybuf, MC_CMD_RSS_CONTEXT_SET_KEY_IN_LEN); @@ -1432,7 +1638,7 @@ static int efx_ef10_populate_rss_table(struct efx_nic *efx, u32 context) for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table); ++i) MCDI_PTR(tablebuf, RSS_CONTEXT_SET_TABLE_IN_INDIRECTION_TABLE)[i] = - (u8) efx->rx_indir_table[i]; + (u8) rx_indir_table[i]; rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_TABLE, tablebuf, sizeof(tablebuf), NULL, 0, NULL); @@ -1460,27 +1666,119 @@ static void efx_ef10_rx_free_indir_table(struct efx_nic *efx) nic_data->rx_rss_context = EFX_EF10_RSS_CONTEXT_INVALID; } -static void efx_ef10_rx_push_rss_config(struct efx_nic *efx) +static int efx_ef10_rx_push_shared_rss_config(struct efx_nic *efx, + unsigned *context_size) { + u32 new_rx_rss_context; struct efx_ef10_nic_data *nic_data = efx->nic_data; - int rc; + int rc = efx_ef10_alloc_rss_context(efx, &new_rx_rss_context, + false, context_size); - netif_dbg(efx, drv, efx->net_dev, "pushing RSS config\n"); + if (rc != 0) + return rc; - if (nic_data->rx_rss_context == EFX_EF10_RSS_CONTEXT_INVALID) { - rc = efx_ef10_alloc_rss_context(efx, &nic_data->rx_rss_context); - if (rc != 0) - goto fail; + nic_data->rx_rss_context = new_rx_rss_context; + nic_data->rx_rss_context_exclusive = false; + efx_set_default_rx_indir_table(efx); + return 0; +} + +static int efx_ef10_rx_push_exclusive_rss_config(struct efx_nic *efx, + const u32 *rx_indir_table) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int rc; + u32 new_rx_rss_context; + + if (nic_data->rx_rss_context == EFX_EF10_RSS_CONTEXT_INVALID || + !nic_data->rx_rss_context_exclusive) { + rc = efx_ef10_alloc_rss_context(efx, &new_rx_rss_context, + true, NULL); + if (rc == -EOPNOTSUPP) + return rc; + else if (rc != 0) + goto fail1; + } else { + new_rx_rss_context = nic_data->rx_rss_context; } - rc = efx_ef10_populate_rss_table(efx, nic_data->rx_rss_context); + rc = efx_ef10_populate_rss_table(efx, new_rx_rss_context, + rx_indir_table); if (rc != 0) - goto fail; + goto fail2; - return; + if (nic_data->rx_rss_context != new_rx_rss_context) + efx_ef10_rx_free_indir_table(efx); + nic_data->rx_rss_context = new_rx_rss_context; + nic_data->rx_rss_context_exclusive = true; + if (rx_indir_table != efx->rx_indir_table) + memcpy(efx->rx_indir_table, rx_indir_table, + sizeof(efx->rx_indir_table)); + return 0; -fail: +fail2: + if (new_rx_rss_context != nic_data->rx_rss_context) + efx_ef10_free_rss_context(efx, new_rx_rss_context); +fail1: netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc); + return rc; +} + +static int efx_ef10_pf_rx_push_rss_config(struct efx_nic *efx, bool user, + const u32 *rx_indir_table) +{ + int rc; + + if (efx->rss_spread == 1) + return 0; + + rc = efx_ef10_rx_push_exclusive_rss_config(efx, rx_indir_table); + + if (rc == -ENOBUFS && !user) { + unsigned context_size; + bool mismatch = false; + size_t i; + + for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table) && !mismatch; + i++) + mismatch = rx_indir_table[i] != + ethtool_rxfh_indir_default(i, efx->rss_spread); + + rc = efx_ef10_rx_push_shared_rss_config(efx, &context_size); + if (rc == 0) { + if (context_size != efx->rss_spread) + netif_warn(efx, probe, efx->net_dev, + "Could not allocate an exclusive RSS" + " context; allocated a shared one of" + " different size." + " Wanted %u, got %u.\n", + efx->rss_spread, context_size); + else if (mismatch) + netif_warn(efx, probe, efx->net_dev, + "Could not allocate an exclusive RSS" + " context; allocated a shared one but" + " could not apply custom" + " indirection.\n"); + else + netif_info(efx, probe, efx->net_dev, + "Could not allocate an exclusive RSS" + " context; allocated a shared one.\n"); + } + } + return rc; +} + +static int efx_ef10_vf_rx_push_rss_config(struct efx_nic *efx, bool user, + const u32 *rx_indir_table + __attribute__ ((unused))) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + if (user) + return -EOPNOTSUPP; + if (nic_data->rx_rss_context != EFX_EF10_RSS_CONTEXT_INVALID) + return 0; + return efx_ef10_rx_push_shared_rss_config(efx, NULL); } static int efx_ef10_rx_probe(struct efx_rx_queue *rx_queue) @@ -1496,14 +1794,15 @@ static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue) MCDI_DECLARE_BUF(inbuf, MC_CMD_INIT_RXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 / EFX_BUF_SIZE)); - MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_RXQ_OUT_LEN); struct efx_channel *channel = efx_rx_queue_channel(rx_queue); size_t entries = rx_queue->rxd.buf.len / EFX_BUF_SIZE; struct efx_nic *efx = rx_queue->efx; - size_t inlen, outlen; + struct efx_ef10_nic_data *nic_data = efx->nic_data; + size_t inlen; dma_addr_t dma_addr; int rc; int i; + BUILD_BUG_ON(MC_CMD_INIT_RXQ_OUT_LEN != 0); rx_queue->scatter_n = 0; rx_queue->scatter_len = 0; @@ -1517,7 +1816,7 @@ static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue) INIT_RXQ_IN_FLAG_PREFIX, 1, INIT_RXQ_IN_FLAG_TIMESTAMP, 1); MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_OWNER_ID, 0); - MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, EVB_PORT_ID_ASSIGNED); + MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, nic_data->vport_id); dma_addr = rx_queue->rxd.buf.dma_addr; @@ -1532,7 +1831,7 @@ static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue) inlen = MC_CMD_INIT_RXQ_IN_LEN(entries); rc = efx_mcdi_rpc(efx, MC_CMD_INIT_RXQ, inbuf, inlen, - outbuf, sizeof(outbuf), &outlen); + NULL, 0, NULL); if (rc) netdev_WARN(efx->net_dev, "failed to initialise RXQ %d\n", efx_rx_queue_index(rx_queue)); @@ -1541,7 +1840,7 @@ static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue) static void efx_ef10_rx_fini(struct efx_rx_queue *rx_queue) { MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_RXQ_IN_LEN); - MCDI_DECLARE_BUF(outbuf, MC_CMD_FINI_RXQ_OUT_LEN); + MCDI_DECLARE_BUF_ERR(outbuf); struct efx_nic *efx = rx_queue->efx; size_t outlen; int rc; @@ -1703,7 +2002,7 @@ static int efx_ef10_ev_init(struct efx_channel *channel) static void efx_ef10_ev_fini(struct efx_channel *channel) { MCDI_DECLARE_BUF(inbuf, MC_CMD_FINI_EVQ_IN_LEN); - MCDI_DECLARE_BUF(outbuf, MC_CMD_FINI_EVQ_OUT_LEN); + MCDI_DECLARE_BUF_ERR(outbuf); struct efx_nic *efx = channel->efx; size_t outlen; int rc; @@ -2286,11 +2585,12 @@ static void efx_ef10_filter_push_prep(struct efx_nic *efx, match_fields); } - MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, EVB_PORT_ID_ASSIGNED); + MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, nic_data->vport_id); MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_DEST, spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ? MC_CMD_FILTER_OP_IN_RX_DEST_DROP : MC_CMD_FILTER_OP_IN_RX_DEST_HOST); + MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DOMAIN, 0); MCDI_SET_DWORD(inbuf, FILTER_OP_IN_TX_DEST, MC_CMD_FILTER_OP_IN_TX_DEST_DEFAULT); MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_QUEUE, @@ -3055,6 +3355,9 @@ fail: return rc; } +/* Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ static void efx_ef10_filter_table_restore(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; @@ -3064,9 +3367,14 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx) bool failed = false; int rc; + WARN_ON(!rwsem_is_locked(&efx->filter_sem)); + if (!nic_data->must_restore_filters) return; + if (!table) + return; + spin_lock_bh(&efx->filter_lock); for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) { @@ -3102,6 +3410,7 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx) nic_data->must_restore_filters = false; } +/* Caller must hold efx->filter_sem for write */ static void efx_ef10_filter_table_remove(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; @@ -3110,6 +3419,10 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) unsigned int filter_idx; int rc; + efx->filter_state = NULL; + if (!table) + return; + for (filter_idx = 0; filter_idx < HUNT_FILTER_TBL_ROWS; filter_idx++) { spec = efx_ef10_filter_entry_spec(table, filter_idx); if (!spec) @@ -3135,6 +3448,9 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) kfree(table); } +/* Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; @@ -3149,6 +3465,9 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) if (!efx_dev_registered(efx)) return; + if (!table) + return; + /* Mark old filters that may need to be removed */ spin_lock_bh(&efx->filter_lock); n = table->dev_uc_count < 0 ? 1 : table->dev_uc_count; @@ -3280,6 +3599,78 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) WARN_ON(remove_failed); } +static int efx_ef10_set_mac_address(struct efx_nic *efx) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_SET_MAC_IN_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + bool was_enabled = efx->port_enabled; + int rc; + + efx_device_detach_sync(efx); + efx_net_stop(efx->net_dev); + down_write(&efx->filter_sem); + efx_ef10_filter_table_remove(efx); + + ether_addr_copy(MCDI_PTR(inbuf, VADAPTOR_SET_MAC_IN_MACADDR), + efx->net_dev->dev_addr); + MCDI_SET_DWORD(inbuf, VADAPTOR_SET_MAC_IN_UPSTREAM_PORT_ID, + nic_data->vport_id); + rc = efx_mcdi_rpc(efx, MC_CMD_VADAPTOR_SET_MAC, inbuf, + sizeof(inbuf), NULL, 0, NULL); + + efx_ef10_filter_table_probe(efx); + up_write(&efx->filter_sem); + if (was_enabled) + efx_net_open(efx->net_dev); + netif_device_attach(efx->net_dev); + +#if !defined(CONFIG_SFC_SRIOV) + if (rc == -EPERM) + netif_err(efx, drv, efx->net_dev, + "Cannot change MAC address; use sfboot to enable mac-spoofing" + " on this interface\n"); +#else + if (rc == -EPERM) { + struct pci_dev *pci_dev_pf = efx->pci_dev->physfn; + + /* Switch to PF and change MAC address on vport */ + if (efx->pci_dev->is_virtfn && pci_dev_pf) { + struct efx_nic *efx_pf = pci_get_drvdata(pci_dev_pf); + + if (!efx_ef10_sriov_set_vf_mac(efx_pf, + nic_data->vf_index, + efx->net_dev->dev_addr)) + return 0; + } + netif_err(efx, drv, efx->net_dev, + "Cannot change MAC address; use sfboot to enable mac-spoofing" + " on this interface\n"); + } else if (efx->pci_dev->is_virtfn) { + /* Successfully changed by VF (with MAC spoofing), so update the + * parent PF if possible. + */ + struct pci_dev *pci_dev_pf = efx->pci_dev->physfn; + + if (pci_dev_pf) { + struct efx_nic *efx_pf = pci_get_drvdata(pci_dev_pf); + struct efx_ef10_nic_data *nic_data = efx_pf->nic_data; + unsigned int i; + + for (i = 0; i < efx_pf->vf_count; ++i) { + struct ef10_vf *vf = nic_data->vf + i; + + if (vf->efx == efx) { + ether_addr_copy(vf->mac, + efx->net_dev->dev_addr); + return 0; + } + } + } + } +#endif + return rc; +} + static int efx_ef10_mac_reconfigure(struct efx_nic *efx) { efx_ef10_filter_sync_rx_mode(efx); @@ -3287,6 +3678,13 @@ static int efx_ef10_mac_reconfigure(struct efx_nic *efx) return efx_mcdi_set_mac(efx); } +static int efx_ef10_mac_reconfigure_vf(struct efx_nic *efx) +{ + efx_ef10_filter_sync_rx_mode(efx); + + return 0; +} + static int efx_ef10_start_bist(struct efx_nic *efx, u32 bist_type) { MCDI_DECLARE_BUF(inbuf, MC_CMD_START_BIST_IN_LEN); @@ -3494,6 +3892,9 @@ static void efx_ef10_ptp_write_host_time(struct efx_nic *efx, u32 host_time) _efx_writed(efx, cpu_to_le32(host_time), ER_DZ_MC_DB_LWRD); } +static void efx_ef10_ptp_write_host_time_vf(struct efx_nic *efx, + u32 host_time) {} + static int efx_ef10_rx_enable_timestamping(struct efx_channel *channel, bool temp) { @@ -3571,6 +3972,12 @@ static int efx_ef10_ptp_set_ts_sync_events(struct efx_nic *efx, bool en, return 0; } +static int efx_ef10_ptp_set_ts_config_vf(struct efx_nic *efx, + struct hwtstamp_config *init) +{ + return -EOPNOTSUPP; +} + static int efx_ef10_ptp_set_ts_config(struct efx_nic *efx, struct hwtstamp_config *init) { @@ -3607,14 +4014,117 @@ static int efx_ef10_ptp_set_ts_config(struct efx_nic *efx, } } +const struct efx_nic_type efx_hunt_a0_vf_nic_type = { + .is_vf = true, + .mem_bar = EFX_MEM_VF_BAR, + .mem_map_size = efx_ef10_mem_map_size, + .probe = efx_ef10_probe_vf, + .remove = efx_ef10_remove, + .dimension_resources = efx_ef10_dimension_resources, + .init = efx_ef10_init_nic, + .fini = efx_port_dummy_op_void, + .map_reset_reason = efx_ef10_map_reset_reason, + .map_reset_flags = efx_ef10_map_reset_flags, + .reset = efx_ef10_reset, + .probe_port = efx_mcdi_port_probe, + .remove_port = efx_mcdi_port_remove, + .fini_dmaq = efx_ef10_fini_dmaq, + .prepare_flr = efx_ef10_prepare_flr, + .finish_flr = efx_port_dummy_op_void, + .describe_stats = efx_ef10_describe_stats, + .update_stats = efx_ef10_update_stats, + .start_stats = efx_port_dummy_op_void, + .pull_stats = efx_port_dummy_op_void, + .stop_stats = efx_port_dummy_op_void, + .set_id_led = efx_mcdi_set_id_led, + .push_irq_moderation = efx_ef10_push_irq_moderation, + .reconfigure_mac = efx_ef10_mac_reconfigure_vf, + .check_mac_fault = efx_mcdi_mac_check_fault, + .reconfigure_port = efx_mcdi_port_reconfigure, + .get_wol = efx_ef10_get_wol_vf, + .set_wol = efx_ef10_set_wol_vf, + .resume_wol = efx_port_dummy_op_void, + .mcdi_request = efx_ef10_mcdi_request, + .mcdi_poll_response = efx_ef10_mcdi_poll_response, + .mcdi_read_response = efx_ef10_mcdi_read_response, + .mcdi_poll_reboot = efx_ef10_mcdi_poll_reboot, + .irq_enable_master = efx_port_dummy_op_void, + .irq_test_generate = efx_ef10_irq_test_generate, + .irq_disable_non_ev = efx_port_dummy_op_void, + .irq_handle_msi = efx_ef10_msi_interrupt, + .irq_handle_legacy = efx_ef10_legacy_interrupt, + .tx_probe = efx_ef10_tx_probe, + .tx_init = efx_ef10_tx_init, + .tx_remove = efx_ef10_tx_remove, + .tx_write = efx_ef10_tx_write, + .rx_push_rss_config = efx_ef10_vf_rx_push_rss_config, + .rx_probe = efx_ef10_rx_probe, + .rx_init = efx_ef10_rx_init, + .rx_remove = efx_ef10_rx_remove, + .rx_write = efx_ef10_rx_write, + .rx_defer_refill = efx_ef10_rx_defer_refill, + .ev_probe = efx_ef10_ev_probe, + .ev_init = efx_ef10_ev_init, + .ev_fini = efx_ef10_ev_fini, + .ev_remove = efx_ef10_ev_remove, + .ev_process = efx_ef10_ev_process, + .ev_read_ack = efx_ef10_ev_read_ack, + .ev_test_generate = efx_ef10_ev_test_generate, + .filter_table_probe = efx_ef10_filter_table_probe, + .filter_table_restore = efx_ef10_filter_table_restore, + .filter_table_remove = efx_ef10_filter_table_remove, + .filter_update_rx_scatter = efx_ef10_filter_update_rx_scatter, + .filter_insert = efx_ef10_filter_insert, + .filter_remove_safe = efx_ef10_filter_remove_safe, + .filter_get_safe = efx_ef10_filter_get_safe, + .filter_clear_rx = efx_ef10_filter_clear_rx, + .filter_count_rx_used = efx_ef10_filter_count_rx_used, + .filter_get_rx_id_limit = efx_ef10_filter_get_rx_id_limit, + .filter_get_rx_ids = efx_ef10_filter_get_rx_ids, +#ifdef CONFIG_RFS_ACCEL + .filter_rfs_insert = efx_ef10_filter_rfs_insert, + .filter_rfs_expire_one = efx_ef10_filter_rfs_expire_one, +#endif +#ifdef CONFIG_SFC_MTD + .mtd_probe = efx_port_dummy_op_int, +#endif + .ptp_write_host_time = efx_ef10_ptp_write_host_time_vf, + .ptp_set_ts_config = efx_ef10_ptp_set_ts_config_vf, +#ifdef CONFIG_SFC_SRIOV + .vswitching_probe = efx_ef10_vswitching_probe_vf, + .vswitching_restore = efx_ef10_vswitching_restore_vf, + .vswitching_remove = efx_ef10_vswitching_remove_vf, +#endif + .get_mac_address = efx_ef10_get_mac_address_vf, + .set_mac_address = efx_ef10_set_mac_address, + + .revision = EFX_REV_HUNT_A0, + .max_dma_mask = DMA_BIT_MASK(ESF_DZ_TX_KER_BUF_ADDR_WIDTH), + .rx_prefix_size = ES_DZ_RX_PREFIX_SIZE, + .rx_hash_offset = ES_DZ_RX_PREFIX_HASH_OFST, + .rx_ts_offset = ES_DZ_RX_PREFIX_TSTAMP_OFST, + .can_rx_scatter = true, + .always_rx_scatter = true, + .max_interrupt_mode = EFX_INT_MODE_MSIX, + .timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH, + .offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | + NETIF_F_RXHASH | NETIF_F_NTUPLE), + .mcdi_max_ver = 2, + .max_rx_ip_filters = HUNT_FILTER_TBL_ROWS, + .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | + 1 << HWTSTAMP_FILTER_ALL, +}; + const struct efx_nic_type efx_hunt_a0_nic_type = { + .is_vf = false, + .mem_bar = EFX_MEM_BAR, .mem_map_size = efx_ef10_mem_map_size, - .probe = efx_ef10_probe, + .probe = efx_ef10_probe_pf, .remove = efx_ef10_remove, .dimension_resources = efx_ef10_dimension_resources, .init = efx_ef10_init_nic, .fini = efx_port_dummy_op_void, - .map_reset_reason = efx_mcdi_map_reset_reason, + .map_reset_reason = efx_ef10_map_reset_reason, .map_reset_flags = efx_ef10_map_reset_flags, .reset = efx_ef10_reset, .probe_port = efx_mcdi_port_probe, @@ -3650,7 +4160,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .tx_init = efx_ef10_tx_init, .tx_remove = efx_ef10_tx_remove, .tx_write = efx_ef10_tx_write, - .rx_push_rss_config = efx_ef10_rx_push_rss_config, + .rx_push_rss_config = efx_ef10_pf_rx_push_rss_config, .rx_probe = efx_ef10_rx_probe, .rx_init = efx_ef10_rx_init, .rx_remove = efx_ef10_rx_remove, @@ -3689,11 +4199,24 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .ptp_write_host_time = efx_ef10_ptp_write_host_time, .ptp_set_ts_sync_events = efx_ef10_ptp_set_ts_sync_events, .ptp_set_ts_config = efx_ef10_ptp_set_ts_config, +#ifdef CONFIG_SFC_SRIOV + .sriov_configure = efx_ef10_sriov_configure, .sriov_init = efx_ef10_sriov_init, .sriov_fini = efx_ef10_sriov_fini, - .sriov_mac_address_changed = efx_ef10_sriov_mac_address_changed, .sriov_wanted = efx_ef10_sriov_wanted, .sriov_reset = efx_ef10_sriov_reset, + .sriov_flr = efx_ef10_sriov_flr, + .sriov_set_vf_mac = efx_ef10_sriov_set_vf_mac, + .sriov_set_vf_vlan = efx_ef10_sriov_set_vf_vlan, + .sriov_set_vf_spoofchk = efx_ef10_sriov_set_vf_spoofchk, + .sriov_get_vf_config = efx_ef10_sriov_get_vf_config, + .sriov_set_vf_link_state = efx_ef10_sriov_set_vf_link_state, + .vswitching_probe = efx_ef10_vswitching_probe_pf, + .vswitching_restore = efx_ef10_vswitching_restore_pf, + .vswitching_remove = efx_ef10_vswitching_remove_pf, +#endif + .get_mac_address = efx_ef10_get_mac_address_pf, + .set_mac_address = efx_ef10_set_mac_address, .revision = EFX_REV_HUNT_A0, .max_dma_mask = DMA_BIT_MASK(ESF_DZ_TX_KER_BUF_ADDR_WIDTH), diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c new file mode 100644 index 000000000000..3969b1bf7ef3 --- /dev/null +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -0,0 +1,738 @@ +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2015 Solarflare Communications Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ +#include <linux/pci.h> +#include <linux/module.h> +#include "net_driver.h" +#include "ef10_sriov.h" +#include "efx.h" +#include "nic.h" +#include "mcdi_pcol.h" + +static int efx_ef10_evb_port_assign(struct efx_nic *efx, unsigned int port_id, + unsigned int vf_fn) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_EVB_PORT_ASSIGN_IN_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + MCDI_SET_DWORD(inbuf, EVB_PORT_ASSIGN_IN_PORT_ID, port_id); + MCDI_POPULATE_DWORD_2(inbuf, EVB_PORT_ASSIGN_IN_FUNCTION, + EVB_PORT_ASSIGN_IN_PF, nic_data->pf_index, + EVB_PORT_ASSIGN_IN_VF, vf_fn); + + return efx_mcdi_rpc(efx, MC_CMD_EVB_PORT_ASSIGN, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +static int efx_ef10_vport_add_mac(struct efx_nic *efx, + unsigned int port_id, u8 *mac) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_ADD_MAC_ADDRESS_IN_LEN); + + MCDI_SET_DWORD(inbuf, VPORT_ADD_MAC_ADDRESS_IN_VPORT_ID, port_id); + ether_addr_copy(MCDI_PTR(inbuf, VPORT_ADD_MAC_ADDRESS_IN_MACADDR), mac); + + return efx_mcdi_rpc(efx, MC_CMD_VPORT_ADD_MAC_ADDRESS, inbuf, + sizeof(inbuf), NULL, 0, NULL); +} + +static int efx_ef10_vport_del_mac(struct efx_nic *efx, + unsigned int port_id, u8 *mac) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_DEL_MAC_ADDRESS_IN_LEN); + + MCDI_SET_DWORD(inbuf, VPORT_DEL_MAC_ADDRESS_IN_VPORT_ID, port_id); + ether_addr_copy(MCDI_PTR(inbuf, VPORT_DEL_MAC_ADDRESS_IN_MACADDR), mac); + + return efx_mcdi_rpc(efx, MC_CMD_VPORT_DEL_MAC_ADDRESS, inbuf, + sizeof(inbuf), NULL, 0, NULL); +} + +static int efx_ef10_vswitch_alloc(struct efx_nic *efx, unsigned int port_id, + unsigned int vswitch_type) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VSWITCH_ALLOC_IN_LEN); + int rc; + + MCDI_SET_DWORD(inbuf, VSWITCH_ALLOC_IN_UPSTREAM_PORT_ID, port_id); + MCDI_SET_DWORD(inbuf, VSWITCH_ALLOC_IN_TYPE, vswitch_type); + MCDI_SET_DWORD(inbuf, VSWITCH_ALLOC_IN_NUM_VLAN_TAGS, 2); + MCDI_POPULATE_DWORD_1(inbuf, VSWITCH_ALLOC_IN_FLAGS, + VSWITCH_ALLOC_IN_FLAG_AUTO_PORT, 0); + + /* Quietly try to allocate 2 VLAN tags */ + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_VSWITCH_ALLOC, inbuf, sizeof(inbuf), + NULL, 0, NULL); + + /* If 2 VLAN tags is too many, revert to trying with 1 VLAN tags */ + if (rc == -EPROTO) { + MCDI_SET_DWORD(inbuf, VSWITCH_ALLOC_IN_NUM_VLAN_TAGS, 1); + rc = efx_mcdi_rpc(efx, MC_CMD_VSWITCH_ALLOC, inbuf, + sizeof(inbuf), NULL, 0, NULL); + } else if (rc) { + efx_mcdi_display_error(efx, MC_CMD_VSWITCH_ALLOC, + MC_CMD_VSWITCH_ALLOC_IN_LEN, + NULL, 0, rc); + } + return rc; +} + +static int efx_ef10_vswitch_free(struct efx_nic *efx, unsigned int port_id) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VSWITCH_FREE_IN_LEN); + + MCDI_SET_DWORD(inbuf, VSWITCH_FREE_IN_UPSTREAM_PORT_ID, port_id); + + return efx_mcdi_rpc(efx, MC_CMD_VSWITCH_FREE, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +static int efx_ef10_vport_alloc(struct efx_nic *efx, + unsigned int port_id_in, + unsigned int vport_type, + u16 vlan, + unsigned int *port_id_out) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_ALLOC_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_VPORT_ALLOC_OUT_LEN); + size_t outlen; + int rc; + + EFX_WARN_ON_PARANOID(!port_id_out); + + MCDI_SET_DWORD(inbuf, VPORT_ALLOC_IN_UPSTREAM_PORT_ID, port_id_in); + MCDI_SET_DWORD(inbuf, VPORT_ALLOC_IN_TYPE, vport_type); + MCDI_SET_DWORD(inbuf, VPORT_ALLOC_IN_NUM_VLAN_TAGS, + (vlan != EFX_EF10_NO_VLAN)); + MCDI_POPULATE_DWORD_1(inbuf, VPORT_ALLOC_IN_FLAGS, + VPORT_ALLOC_IN_FLAG_AUTO_PORT, 0); + if (vlan != EFX_EF10_NO_VLAN) + MCDI_POPULATE_DWORD_1(inbuf, VPORT_ALLOC_IN_VLAN_TAGS, + VPORT_ALLOC_IN_VLAN_TAG_0, vlan); + + rc = efx_mcdi_rpc(efx, MC_CMD_VPORT_ALLOC, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &outlen); + if (rc) + return rc; + if (outlen < MC_CMD_VPORT_ALLOC_OUT_LEN) + return -EIO; + + *port_id_out = MCDI_DWORD(outbuf, VPORT_ALLOC_OUT_VPORT_ID); + return 0; +} + +static int efx_ef10_vport_free(struct efx_nic *efx, unsigned int port_id) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_FREE_IN_LEN); + + MCDI_SET_DWORD(inbuf, VPORT_FREE_IN_VPORT_ID, port_id); + + return efx_mcdi_rpc(efx, MC_CMD_VPORT_FREE, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +static int efx_ef10_vadaptor_alloc(struct efx_nic *efx, unsigned int port_id) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_ALLOC_IN_LEN); + + MCDI_SET_DWORD(inbuf, VADAPTOR_ALLOC_IN_UPSTREAM_PORT_ID, port_id); + return efx_mcdi_rpc(efx, MC_CMD_VADAPTOR_ALLOC, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +static int efx_ef10_vadaptor_free(struct efx_nic *efx, unsigned int port_id) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_FREE_IN_LEN); + + MCDI_SET_DWORD(inbuf, VADAPTOR_FREE_IN_UPSTREAM_PORT_ID, port_id); + return efx_mcdi_rpc(efx, MC_CMD_VADAPTOR_FREE, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +static void efx_ef10_sriov_free_vf_vports(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int i; + + if (!nic_data->vf) + return; + + for (i = 0; i < efx->vf_count; i++) { + struct ef10_vf *vf = nic_data->vf + i; + + if (vf->vport_assigned) { + efx_ef10_evb_port_assign(efx, EVB_PORT_ID_NULL, i); + vf->vport_assigned = 0; + } + + if (!is_zero_ether_addr(vf->mac)) { + efx_ef10_vport_del_mac(efx, vf->vport_id, vf->mac); + eth_zero_addr(vf->mac); + } + + if (vf->vport_id) { + efx_ef10_vport_free(efx, vf->vport_id); + vf->vport_id = 0; + } + + vf->efx = NULL; + } +} + +static void efx_ef10_sriov_free_vf_vswitching(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + efx_ef10_sriov_free_vf_vports(efx); + kfree(nic_data->vf); + nic_data->vf = NULL; +} + +static int efx_ef10_sriov_assign_vf_vport(struct efx_nic *efx, + unsigned int vf_i) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct ef10_vf *vf = nic_data->vf + vf_i; + int rc; + + if (WARN_ON_ONCE(!nic_data->vf)) + return -EOPNOTSUPP; + + rc = efx_ef10_vport_alloc(efx, EVB_PORT_ID_ASSIGNED, + MC_CMD_VPORT_ALLOC_IN_VPORT_TYPE_NORMAL, + vf->vlan, &vf->vport_id); + if (rc) + return rc; + + rc = efx_ef10_vport_add_mac(efx, vf->vport_id, vf->mac); + if (rc) { + eth_zero_addr(vf->mac); + return rc; + } + + rc = efx_ef10_evb_port_assign(efx, vf->vport_id, vf_i); + if (rc) + return rc; + + vf->vport_assigned = 1; + return 0; +} + +static int efx_ef10_sriov_alloc_vf_vswitching(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + unsigned int i; + int rc; + + nic_data->vf = kcalloc(efx->vf_count, sizeof(struct ef10_vf), + GFP_KERNEL); + if (!nic_data->vf) + return -ENOMEM; + + for (i = 0; i < efx->vf_count; i++) { + random_ether_addr(nic_data->vf[i].mac); + nic_data->vf[i].efx = NULL; + nic_data->vf[i].vlan = EFX_EF10_NO_VLAN; + + rc = efx_ef10_sriov_assign_vf_vport(efx, i); + if (rc) + goto fail; + } + + return 0; +fail: + efx_ef10_sriov_free_vf_vports(efx); + kfree(nic_data->vf); + nic_data->vf = NULL; + return rc; +} + +static int efx_ef10_sriov_restore_vf_vswitching(struct efx_nic *efx) +{ + unsigned int i; + int rc; + + for (i = 0; i < efx->vf_count; i++) { + rc = efx_ef10_sriov_assign_vf_vport(efx, i); + if (rc) + goto fail; + } + + return 0; +fail: + efx_ef10_sriov_free_vf_vswitching(efx); + return rc; +} + +/* On top of the default firmware vswitch setup, create a VEB vswitch and + * expansion vport for use by this function. + */ +int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct net_device *net_dev = efx->net_dev; + int rc; + + if (pci_sriov_get_totalvfs(efx->pci_dev) <= 0) { + /* vswitch not needed as we have no VFs */ + efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + return 0; + } + + rc = efx_ef10_vswitch_alloc(efx, EVB_PORT_ID_ASSIGNED, + MC_CMD_VSWITCH_ALLOC_IN_VSWITCH_TYPE_VEB); + if (rc) + goto fail1; + + rc = efx_ef10_vport_alloc(efx, EVB_PORT_ID_ASSIGNED, + MC_CMD_VPORT_ALLOC_IN_VPORT_TYPE_NORMAL, + EFX_EF10_NO_VLAN, &nic_data->vport_id); + if (rc) + goto fail2; + + rc = efx_ef10_vport_add_mac(efx, nic_data->vport_id, net_dev->dev_addr); + if (rc) + goto fail3; + ether_addr_copy(nic_data->vport_mac, net_dev->dev_addr); + + rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + if (rc) + goto fail4; + + return 0; +fail4: + efx_ef10_vport_del_mac(efx, nic_data->vport_id, nic_data->vport_mac); + eth_zero_addr(nic_data->vport_mac); +fail3: + efx_ef10_vport_free(efx, nic_data->vport_id); + nic_data->vport_id = EVB_PORT_ID_ASSIGNED; +fail2: + efx_ef10_vswitch_free(efx, EVB_PORT_ID_ASSIGNED); +fail1: + return rc; +} + +int efx_ef10_vswitching_probe_vf(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + return efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); +} + +int efx_ef10_vswitching_restore_pf(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int rc; + + if (!nic_data->must_probe_vswitching) + return 0; + + rc = efx_ef10_vswitching_probe_pf(efx); + if (rc) + goto fail; + + rc = efx_ef10_sriov_restore_vf_vswitching(efx); + if (rc) + goto fail; + + nic_data->must_probe_vswitching = false; +fail: + return rc; +} + +int efx_ef10_vswitching_restore_vf(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int rc; + + if (!nic_data->must_probe_vswitching) + return 0; + + rc = efx_ef10_vadaptor_free(efx, EVB_PORT_ID_ASSIGNED); + if (rc) + return rc; + + nic_data->must_probe_vswitching = false; + return 0; +} + +void efx_ef10_vswitching_remove_pf(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + efx_ef10_sriov_free_vf_vswitching(efx); + + efx_ef10_vadaptor_free(efx, nic_data->vport_id); + + if (nic_data->vport_id == EVB_PORT_ID_ASSIGNED) + return; /* No vswitch was ever created */ + + if (!is_zero_ether_addr(nic_data->vport_mac)) { + efx_ef10_vport_del_mac(efx, nic_data->vport_id, + efx->net_dev->dev_addr); + eth_zero_addr(nic_data->vport_mac); + } + efx_ef10_vport_free(efx, nic_data->vport_id); + nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + + efx_ef10_vswitch_free(efx, nic_data->vport_id); +} + +void efx_ef10_vswitching_remove_vf(struct efx_nic *efx) +{ + efx_ef10_vadaptor_free(efx, EVB_PORT_ID_ASSIGNED); +} + +static int efx_ef10_pci_sriov_enable(struct efx_nic *efx, int num_vfs) +{ + int rc = 0; + struct pci_dev *dev = efx->pci_dev; + + efx->vf_count = num_vfs; + + rc = efx_ef10_sriov_alloc_vf_vswitching(efx); + if (rc) + goto fail1; + + rc = pci_enable_sriov(dev, num_vfs); + if (rc) + goto fail2; + + return 0; +fail2: + efx_ef10_sriov_free_vf_vswitching(efx); +fail1: + efx->vf_count = 0; + netif_err(efx, probe, efx->net_dev, + "Failed to enable SRIOV VFs\n"); + return rc; +} + +static int efx_ef10_pci_sriov_disable(struct efx_nic *efx) +{ + struct pci_dev *dev = efx->pci_dev; + + pci_disable_sriov(dev); + efx_ef10_sriov_free_vf_vswitching(efx); + efx->vf_count = 0; + return 0; +} + +int efx_ef10_sriov_configure(struct efx_nic *efx, int num_vfs) +{ + if (num_vfs == 0) + return efx_ef10_pci_sriov_disable(efx); + else + return efx_ef10_pci_sriov_enable(efx, num_vfs); +} + +int efx_ef10_sriov_init(struct efx_nic *efx) +{ + return 0; +} + +void efx_ef10_sriov_fini(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int rc; + + if (!nic_data->vf) + return; + + rc = efx_ef10_pci_sriov_disable(efx); + if (rc) + netif_dbg(efx, drv, efx->net_dev, + "Disabling SRIOV was not successful rc=%d\n", rc); + else + netif_dbg(efx, drv, efx->net_dev, "SRIOV disabled\n"); +} + +static int efx_ef10_vport_del_vf_mac(struct efx_nic *efx, unsigned int port_id, + u8 *mac) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_VPORT_DEL_MAC_ADDRESS_IN_LEN); + MCDI_DECLARE_BUF_ERR(outbuf); + size_t outlen; + int rc; + + MCDI_SET_DWORD(inbuf, VPORT_DEL_MAC_ADDRESS_IN_VPORT_ID, port_id); + ether_addr_copy(MCDI_PTR(inbuf, VPORT_DEL_MAC_ADDRESS_IN_MACADDR), mac); + + rc = efx_mcdi_rpc(efx, MC_CMD_VPORT_DEL_MAC_ADDRESS, inbuf, + sizeof(inbuf), outbuf, sizeof(outbuf), &outlen); + + return rc; +} + +int efx_ef10_sriov_set_vf_mac(struct efx_nic *efx, int vf_i, u8 *mac) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct ef10_vf *vf; + int rc; + + if (!nic_data->vf) + return -EOPNOTSUPP; + + if (vf_i >= efx->vf_count) + return -EINVAL; + vf = nic_data->vf + vf_i; + + if (vf->efx) { + efx_device_detach_sync(vf->efx); + efx_net_stop(vf->efx->net_dev); + + down_write(&vf->efx->filter_sem); + vf->efx->type->filter_table_remove(vf->efx); + + rc = efx_ef10_vadaptor_free(vf->efx, EVB_PORT_ID_ASSIGNED); + if (rc) { + up_write(&vf->efx->filter_sem); + return rc; + } + } + + rc = efx_ef10_evb_port_assign(efx, EVB_PORT_ID_NULL, vf_i); + if (rc) + return rc; + + if (!is_zero_ether_addr(vf->mac)) { + rc = efx_ef10_vport_del_vf_mac(efx, vf->vport_id, vf->mac); + if (rc) + return rc; + } + + if (!is_zero_ether_addr(mac)) { + rc = efx_ef10_vport_add_mac(efx, vf->vport_id, mac); + if (rc) { + eth_zero_addr(vf->mac); + goto fail; + } + if (vf->efx) + ether_addr_copy(vf->efx->net_dev->dev_addr, mac); + } + + ether_addr_copy(vf->mac, mac); + + rc = efx_ef10_evb_port_assign(efx, vf->vport_id, vf_i); + if (rc) + goto fail; + + if (vf->efx) { + /* VF cannot use the vport_id that the PF created */ + rc = efx_ef10_vadaptor_alloc(vf->efx, EVB_PORT_ID_ASSIGNED); + if (rc) { + up_write(&vf->efx->filter_sem); + return rc; + } + vf->efx->type->filter_table_probe(vf->efx); + up_write(&vf->efx->filter_sem); + efx_net_open(vf->efx->net_dev); + netif_device_attach(vf->efx->net_dev); + } + + return 0; + +fail: + memset(vf->mac, 0, ETH_ALEN); + return rc; +} + +int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, u16 vlan, + u8 qos) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct ef10_vf *vf; + u16 old_vlan, new_vlan; + int rc = 0, rc2 = 0; + + if (vf_i >= efx->vf_count) + return -EINVAL; + if (qos != 0) + return -EINVAL; + + vf = nic_data->vf + vf_i; + + new_vlan = (vlan == 0) ? EFX_EF10_NO_VLAN : vlan; + if (new_vlan == vf->vlan) + return 0; + + if (vf->efx) { + efx_device_detach_sync(vf->efx); + efx_net_stop(vf->efx->net_dev); + + down_write(&vf->efx->filter_sem); + vf->efx->type->filter_table_remove(vf->efx); + + rc = efx_ef10_vadaptor_free(vf->efx, EVB_PORT_ID_ASSIGNED); + if (rc) + goto restore_filters; + } + + if (vf->vport_assigned) { + rc = efx_ef10_evb_port_assign(efx, EVB_PORT_ID_NULL, vf_i); + if (rc) { + netif_warn(efx, drv, efx->net_dev, + "Failed to change vlan on VF %d.\n", vf_i); + netif_warn(efx, drv, efx->net_dev, + "This is likely because the VF is bound to a driver in a VM.\n"); + netif_warn(efx, drv, efx->net_dev, + "Please unload the driver in the VM.\n"); + goto restore_vadaptor; + } + vf->vport_assigned = 0; + } + + if (!is_zero_ether_addr(vf->mac)) { + rc = efx_ef10_vport_del_mac(efx, vf->vport_id, vf->mac); + if (rc) + goto restore_evb_port; + } + + if (vf->vport_id) { + rc = efx_ef10_vport_free(efx, vf->vport_id); + if (rc) + goto restore_mac; + vf->vport_id = 0; + } + + /* Do the actual vlan change */ + old_vlan = vf->vlan; + vf->vlan = new_vlan; + + /* Restore everything in reverse order */ + rc = efx_ef10_vport_alloc(efx, EVB_PORT_ID_ASSIGNED, + MC_CMD_VPORT_ALLOC_IN_VPORT_TYPE_NORMAL, + vf->vlan, &vf->vport_id); + if (rc) + goto reset_nic; + +restore_mac: + if (!is_zero_ether_addr(vf->mac)) { + rc2 = efx_ef10_vport_add_mac(efx, vf->vport_id, vf->mac); + if (rc2) { + eth_zero_addr(vf->mac); + goto reset_nic; + } + } + +restore_evb_port: + rc2 = efx_ef10_evb_port_assign(efx, vf->vport_id, vf_i); + if (rc2) + goto reset_nic; + else + vf->vport_assigned = 1; + +restore_vadaptor: + if (vf->efx) { + rc2 = efx_ef10_vadaptor_alloc(vf->efx, EVB_PORT_ID_ASSIGNED); + if (rc2) + goto reset_nic; + } + +restore_filters: + if (vf->efx) { + rc2 = vf->efx->type->filter_table_probe(vf->efx); + if (rc2) + goto reset_nic; + + up_write(&vf->efx->filter_sem); + + rc2 = efx_net_open(vf->efx->net_dev); + if (rc2) + goto reset_nic; + + netif_device_attach(vf->efx->net_dev); + } + return rc; + +reset_nic: + if (vf->efx) { + up_write(&vf->efx->filter_sem); + netif_err(efx, drv, efx->net_dev, + "Failed to restore VF - scheduling reset.\n"); + efx_schedule_reset(vf->efx, RESET_TYPE_DATAPATH); + } else { + netif_err(efx, drv, efx->net_dev, + "Failed to restore the VF and cannot reset the VF " + "- VF is not functional.\n"); + netif_err(efx, drv, efx->net_dev, + "Please reload the driver attached to the VF.\n"); + } + + return rc ? rc : rc2; +} + +int efx_ef10_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf_i, + bool spoofchk) +{ + return spoofchk ? -EOPNOTSUPP : 0; +} + +int efx_ef10_sriov_set_vf_link_state(struct efx_nic *efx, int vf_i, + int link_state) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_LINK_STATE_MODE_IN_LEN); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + BUILD_BUG_ON(IFLA_VF_LINK_STATE_AUTO != + MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_AUTO); + BUILD_BUG_ON(IFLA_VF_LINK_STATE_ENABLE != + MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_UP); + BUILD_BUG_ON(IFLA_VF_LINK_STATE_DISABLE != + MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_DOWN); + MCDI_POPULATE_DWORD_2(inbuf, LINK_STATE_MODE_IN_FUNCTION, + LINK_STATE_MODE_IN_FUNCTION_PF, + nic_data->pf_index, + LINK_STATE_MODE_IN_FUNCTION_VF, vf_i); + MCDI_SET_DWORD(inbuf, LINK_STATE_MODE_IN_NEW_MODE, link_state); + return efx_mcdi_rpc(efx, MC_CMD_LINK_STATE_MODE, inbuf, sizeof(inbuf), + NULL, 0, NULL); /* don't care what old mode was */ +} + +int efx_ef10_sriov_get_vf_config(struct efx_nic *efx, int vf_i, + struct ifla_vf_info *ivf) +{ + MCDI_DECLARE_BUF(inbuf, MC_CMD_LINK_STATE_MODE_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_LINK_STATE_MODE_OUT_LEN); + + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct ef10_vf *vf; + size_t outlen; + int rc; + + if (vf_i >= efx->vf_count) + return -EINVAL; + + if (!nic_data->vf) + return -EOPNOTSUPP; + + vf = nic_data->vf + vf_i; + + ivf->vf = vf_i; + ivf->min_tx_rate = 0; + ivf->max_tx_rate = 0; + ether_addr_copy(ivf->mac, vf->mac); + ivf->vlan = (vf->vlan == EFX_EF10_NO_VLAN) ? 0 : vf->vlan; + ivf->qos = 0; + + MCDI_POPULATE_DWORD_2(inbuf, LINK_STATE_MODE_IN_FUNCTION, + LINK_STATE_MODE_IN_FUNCTION_PF, + nic_data->pf_index, + LINK_STATE_MODE_IN_FUNCTION_VF, vf_i); + MCDI_SET_DWORD(inbuf, LINK_STATE_MODE_IN_NEW_MODE, + MC_CMD_LINK_STATE_MODE_IN_DO_NOT_CHANGE); + rc = efx_mcdi_rpc(efx, MC_CMD_LINK_STATE_MODE, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &outlen); + if (rc) + return rc; + if (outlen < MC_CMD_LINK_STATE_MODE_OUT_LEN) + return -EIO; + ivf->linkstate = MCDI_DWORD(outbuf, LINK_STATE_MODE_OUT_OLD_MODE); + + return 0; +} diff --git a/drivers/net/ethernet/sfc/ef10_sriov.h b/drivers/net/ethernet/sfc/ef10_sriov.h new file mode 100644 index 000000000000..b98557670f73 --- /dev/null +++ b/drivers/net/ethernet/sfc/ef10_sriov.h @@ -0,0 +1,64 @@ +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2015 Solarflare Communications Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ + +#ifndef EF10_SRIOV_H +#define EF10_SRIOV_H + +#include "net_driver.h" + +/** + * struct ef10_vf - PF's store of VF data + * @efx: efx_nic struct for the current VF + * @vport_id: vport ID for the VF + * @vport_assigned: record whether the vport is currently assigned to the VF + * @mac: MAC address for the VF, zero when address is removed from the vport + * @vlan: Default VLAN for the VF or #EFX_EF10_NO_VLAN + */ +struct ef10_vf { + struct efx_nic *efx; + unsigned int vport_id; + unsigned int vport_assigned; + u8 mac[ETH_ALEN]; + u16 vlan; +#define EFX_EF10_NO_VLAN 0 +}; + +static inline bool efx_ef10_sriov_wanted(struct efx_nic *efx) +{ + return false; +} + +int efx_ef10_sriov_configure(struct efx_nic *efx, int num_vfs); +int efx_ef10_sriov_init(struct efx_nic *efx); +static inline void efx_ef10_sriov_reset(struct efx_nic *efx) {} +void efx_ef10_sriov_fini(struct efx_nic *efx); +static inline void efx_ef10_sriov_flr(struct efx_nic *efx, unsigned vf_i) {} + +int efx_ef10_sriov_set_vf_mac(struct efx_nic *efx, int vf, u8 *mac); + +int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, + u16 vlan, u8 qos); + +int efx_ef10_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf, + bool spoofchk); + +int efx_ef10_sriov_get_vf_config(struct efx_nic *efx, int vf_i, + struct ifla_vf_info *ivf); + +int efx_ef10_sriov_set_vf_link_state(struct efx_nic *efx, int vf_i, + int link_state); + +int efx_ef10_vswitching_probe_pf(struct efx_nic *efx); +int efx_ef10_vswitching_probe_vf(struct efx_nic *efx); +int efx_ef10_vswitching_restore_pf(struct efx_nic *efx); +int efx_ef10_vswitching_restore_vf(struct efx_nic *efx); +void efx_ef10_vswitching_remove_pf(struct efx_nic *efx); +void efx_ef10_vswitching_remove_vf(struct efx_nic *efx); + +#endif /* EF10_SRIOV_H */ diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 4b00545a3ace..2d4853c032c3 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -26,6 +26,7 @@ #include "efx.h" #include "nic.h" #include "selftest.h" +#include "sriov.h" #include "mcdi.h" #include "workarounds.h" @@ -76,6 +77,7 @@ const char *const efx_reset_type_names[] = { [RESET_TYPE_RECOVER_OR_ALL] = "RECOVER_OR_ALL", [RESET_TYPE_WORLD] = "WORLD", [RESET_TYPE_RECOVER_OR_DISABLE] = "RECOVER_OR_DISABLE", + [RESET_TYPE_DATAPATH] = "DATAPATH", [RESET_TYPE_MC_BIST] = "MC_BIST", [RESET_TYPE_DISABLE] = "DISABLE", [RESET_TYPE_TX_WATCHDOG] = "TX_WATCHDOG", @@ -948,6 +950,16 @@ void efx_link_set_wanted_fc(struct efx_nic *efx, u8 wanted_fc) static void efx_fini_port(struct efx_nic *efx); +/* We assume that efx->type->reconfigure_mac will always try to sync RX + * filters and therefore needs to read-lock the filter table against freeing + */ +void efx_mac_reconfigure(struct efx_nic *efx) +{ + down_read(&efx->filter_sem); + efx->type->reconfigure_mac(efx); + up_read(&efx->filter_sem); +} + /* Push loopback/power/transmit disable settings to the PHY, and reconfigure * the MAC appropriately. All other PHY configuration changes are pushed * through phy_op->set_settings(), and pushed asynchronously to the MAC @@ -1001,7 +1013,7 @@ static void efx_mac_work(struct work_struct *data) mutex_lock(&efx->mac_lock); if (efx->port_enabled) - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); mutex_unlock(&efx->mac_lock); } @@ -1041,11 +1053,11 @@ static int efx_init_port(struct efx_nic *efx) /* Reconfigure the MAC before creating dma queues (required for * Falcon/A1 where RX_INGR_EN/TX_DRAIN_EN isn't supported) */ - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); /* Ensure the PHY advertises the correct flow control settings */ rc = efx->phy_op->reconfigure(efx); - if (rc) + if (rc && rc != -EPERM) goto fail2; mutex_unlock(&efx->mac_lock); @@ -1067,7 +1079,7 @@ static void efx_start_port(struct efx_nic *efx) efx->port_enabled = true; /* Ensure MAC ingress/egress is enabled */ - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); mutex_unlock(&efx->mac_lock); } @@ -1200,10 +1212,12 @@ static int efx_init_io(struct efx_nic *efx) struct pci_dev *pci_dev = efx->pci_dev; dma_addr_t dma_mask = efx->type->max_dma_mask; unsigned int mem_map_size = efx->type->mem_map_size(efx); - int rc; + int rc, bar; netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n"); + bar = efx->type->mem_bar; + rc = pci_enable_device(pci_dev); if (rc) { netif_err(efx, probe, efx->net_dev, @@ -1234,8 +1248,8 @@ static int efx_init_io(struct efx_nic *efx) netif_dbg(efx, probe, efx->net_dev, "using DMA mask %llx\n", (unsigned long long) dma_mask); - efx->membase_phys = pci_resource_start(efx->pci_dev, EFX_MEM_BAR); - rc = pci_request_region(pci_dev, EFX_MEM_BAR, "sfc"); + efx->membase_phys = pci_resource_start(efx->pci_dev, bar); + rc = pci_request_region(pci_dev, bar, "sfc"); if (rc) { netif_err(efx, probe, efx->net_dev, "request for memory BAR failed\n"); @@ -1258,7 +1272,7 @@ static int efx_init_io(struct efx_nic *efx) return 0; fail4: - pci_release_region(efx->pci_dev, EFX_MEM_BAR); + pci_release_region(efx->pci_dev, bar); fail3: efx->membase_phys = 0; fail2: @@ -1269,6 +1283,8 @@ static int efx_init_io(struct efx_nic *efx) static void efx_fini_io(struct efx_nic *efx) { + int bar; + netif_dbg(efx, drv, efx->net_dev, "shutting down I/O\n"); if (efx->membase) { @@ -1277,13 +1293,23 @@ static void efx_fini_io(struct efx_nic *efx) } if (efx->membase_phys) { - pci_release_region(efx->pci_dev, EFX_MEM_BAR); + bar = efx->type->mem_bar; + pci_release_region(efx->pci_dev, bar); efx->membase_phys = 0; } pci_disable_device(efx->pci_dev); } +void efx_set_default_rx_indir_table(struct efx_nic *efx) +{ + size_t i; + + for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table); i++) + efx->rx_indir_table[i] = + ethtool_rxfh_indir_default(i, efx->rss_spread); +} + static unsigned int efx_wanted_parallelism(struct efx_nic *efx) { cpumask_var_t thread_mask; @@ -1314,15 +1340,19 @@ static unsigned int efx_wanted_parallelism(struct efx_nic *efx) /* If RSS is requested for the PF *and* VFs then we can't write RSS * table entries that are inaccessible to VFs */ - if (efx->type->sriov_wanted(efx) && efx_vf_size(efx) > 1 && - count > efx_vf_size(efx)) { - netif_warn(efx, probe, efx->net_dev, - "Reducing number of RSS channels from %u to %u for " - "VF support. Increase vf-msix-limit to use more " - "channels on the PF.\n", - count, efx_vf_size(efx)); - count = efx_vf_size(efx); +#ifdef CONFIG_SFC_SRIOV + if (efx->type->sriov_wanted) { + if (efx->type->sriov_wanted(efx) && efx_vf_size(efx) > 1 && + count > efx_vf_size(efx)) { + netif_warn(efx, probe, efx->net_dev, + "Reducing number of RSS channels from %u to %u for " + "VF support. Increase vf-msix-limit to use more " + "channels on the PF.\n", + count, efx_vf_size(efx)); + count = efx_vf_size(efx); + } } +#endif return count; } @@ -1426,10 +1456,15 @@ static int efx_probe_interrupts(struct efx_nic *efx) } /* RSS might be usable on VFs even if it is disabled on the PF */ - - efx->rss_spread = ((efx->n_rx_channels > 1 || - !efx->type->sriov_wanted(efx)) ? - efx->n_rx_channels : efx_vf_size(efx)); +#ifdef CONFIG_SFC_SRIOV + if (efx->type->sriov_wanted) { + efx->rss_spread = ((efx->n_rx_channels > 1 || + !efx->type->sriov_wanted(efx)) ? + efx->n_rx_channels : efx_vf_size(efx)); + return 0; + } +#endif + efx->rss_spread = efx->n_rx_channels; return 0; } @@ -1593,7 +1628,6 @@ static void efx_set_channels(struct efx_nic *efx) static int efx_probe_nic(struct efx_nic *efx) { - size_t i; int rc; netif_dbg(efx, probe, efx->net_dev, "creating NIC\n"); @@ -1616,10 +1650,9 @@ static int efx_probe_nic(struct efx_nic *efx) goto fail2; if (efx->n_channels > 1) - netdev_rss_key_fill(&efx->rx_hash_key, sizeof(efx->rx_hash_key)); - for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table); i++) - efx->rx_indir_table[i] = - ethtool_rxfh_indir_default(i, efx->rss_spread); + netdev_rss_key_fill(&efx->rx_hash_key, + sizeof(efx->rx_hash_key)); + efx_set_default_rx_indir_table(efx); netif_set_real_num_tx_queues(efx->net_dev, efx->n_tx_channels); netif_set_real_num_rx_queues(efx->net_dev, efx->n_rx_channels); @@ -1650,10 +1683,11 @@ static int efx_probe_filters(struct efx_nic *efx) int rc; spin_lock_init(&efx->filter_lock); - + init_rwsem(&efx->filter_sem); + down_write(&efx->filter_sem); rc = efx->type->filter_table_probe(efx); if (rc) - return rc; + goto out_unlock; #ifdef CONFIG_RFS_ACCEL if (efx->type->offload_features & NETIF_F_NTUPLE) { @@ -1662,12 +1696,14 @@ static int efx_probe_filters(struct efx_nic *efx) GFP_KERNEL); if (!efx->rps_flow_id) { efx->type->filter_table_remove(efx); - return -ENOMEM; + rc = -ENOMEM; + goto out_unlock; } } #endif - - return 0; +out_unlock: + up_write(&efx->filter_sem); + return rc; } static void efx_remove_filters(struct efx_nic *efx) @@ -1675,12 +1711,16 @@ static void efx_remove_filters(struct efx_nic *efx) #ifdef CONFIG_RFS_ACCEL kfree(efx->rps_flow_id); #endif + down_write(&efx->filter_sem); efx->type->filter_table_remove(efx); + up_write(&efx->filter_sem); } static void efx_restore_filters(struct efx_nic *efx) { + down_read(&efx->filter_sem); efx->type->filter_table_restore(efx); + up_read(&efx->filter_sem); } /************************************************************************** @@ -1712,21 +1752,33 @@ static int efx_probe_all(struct efx_nic *efx) } efx->rxq_entries = efx->txq_entries = EFX_DEFAULT_DMAQ_SIZE; +#ifdef CONFIG_SFC_SRIOV + rc = efx->type->vswitching_probe(efx); + if (rc) /* not fatal; the PF will still work fine */ + netif_warn(efx, probe, efx->net_dev, + "failed to setup vswitching rc=%d;" + " VFs may not function\n", rc); +#endif + rc = efx_probe_filters(efx); if (rc) { netif_err(efx, probe, efx->net_dev, "failed to create filter tables\n"); - goto fail3; + goto fail4; } rc = efx_probe_channels(efx); if (rc) - goto fail4; + goto fail5; return 0; - fail4: + fail5: efx_remove_filters(efx); + fail4: +#ifdef CONFIG_SFC_SRIOV + efx->type->vswitching_remove(efx); +#endif fail3: efx_remove_port(efx); fail2: @@ -1816,6 +1868,9 @@ static void efx_remove_all(struct efx_nic *efx) { efx_remove_channels(efx); efx_remove_filters(efx); +#ifdef CONFIG_SFC_SRIOV + efx->type->vswitching_remove(efx); +#endif efx_remove_port(efx); efx_remove_nic(efx); } @@ -2059,7 +2114,7 @@ static int efx_busy_poll(struct napi_struct *napi) *************************************************************************/ /* Context: process, rtnl_lock() held. */ -static int efx_net_open(struct net_device *net_dev) +int efx_net_open(struct net_device *net_dev) { struct efx_nic *efx = netdev_priv(net_dev); int rc; @@ -2088,7 +2143,7 @@ static int efx_net_open(struct net_device *net_dev) * Note that the kernel will ignore our return code; this method * should really be a void. */ -static int efx_net_stop(struct net_device *net_dev) +int efx_net_stop(struct net_device *net_dev) { struct efx_nic *efx = netdev_priv(net_dev); @@ -2146,7 +2201,7 @@ static int efx_change_mtu(struct net_device *net_dev, int new_mtu) mutex_lock(&efx->mac_lock); net_dev->mtu = new_mtu; - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); mutex_unlock(&efx->mac_lock); efx_start_all(efx); @@ -2159,6 +2214,8 @@ static int efx_set_mac_address(struct net_device *net_dev, void *data) struct efx_nic *efx = netdev_priv(net_dev); struct sockaddr *addr = data; u8 *new_addr = addr->sa_data; + u8 old_addr[6]; + int rc; if (!is_valid_ether_addr(new_addr)) { netif_err(efx, drv, efx->net_dev, @@ -2167,12 +2224,20 @@ static int efx_set_mac_address(struct net_device *net_dev, void *data) return -EADDRNOTAVAIL; } + /* save old address */ + ether_addr_copy(old_addr, net_dev->dev_addr); ether_addr_copy(net_dev->dev_addr, new_addr); - efx->type->sriov_mac_address_changed(efx); + if (efx->type->set_mac_address) { + rc = efx->type->set_mac_address(efx); + if (rc) { + ether_addr_copy(net_dev->dev_addr, old_addr); + return rc; + } + } /* Reconfigure the MAC */ mutex_lock(&efx->mac_lock); - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); mutex_unlock(&efx->mac_lock); return 0; @@ -2199,7 +2264,7 @@ static int efx_set_features(struct net_device *net_dev, netdev_features_t data) return 0; } -static const struct net_device_ops efx_farch_netdev_ops = { +static const struct net_device_ops efx_netdev_ops = { .ndo_open = efx_net_open, .ndo_stop = efx_net_stop, .ndo_get_stats64 = efx_net_stats, @@ -2212,10 +2277,11 @@ static const struct net_device_ops efx_farch_netdev_ops = { .ndo_set_rx_mode = efx_set_rx_mode, .ndo_set_features = efx_set_features, #ifdef CONFIG_SFC_SRIOV - .ndo_set_vf_mac = efx_siena_sriov_set_vf_mac, - .ndo_set_vf_vlan = efx_siena_sriov_set_vf_vlan, - .ndo_set_vf_spoofchk = efx_siena_sriov_set_vf_spoofchk, - .ndo_get_vf_config = efx_siena_sriov_get_vf_config, + .ndo_set_vf_mac = efx_sriov_set_vf_mac, + .ndo_set_vf_vlan = efx_sriov_set_vf_vlan, + .ndo_set_vf_spoofchk = efx_sriov_set_vf_spoofchk, + .ndo_get_vf_config = efx_sriov_get_vf_config, + .ndo_set_vf_link_state = efx_sriov_set_vf_link_state, #endif #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = efx_netpoll, @@ -2229,29 +2295,6 @@ static const struct net_device_ops efx_farch_netdev_ops = { #endif }; -static const struct net_device_ops efx_ef10_netdev_ops = { - .ndo_open = efx_net_open, - .ndo_stop = efx_net_stop, - .ndo_get_stats64 = efx_net_stats, - .ndo_tx_timeout = efx_watchdog, - .ndo_start_xmit = efx_hard_start_xmit, - .ndo_validate_addr = eth_validate_addr, - .ndo_do_ioctl = efx_ioctl, - .ndo_change_mtu = efx_change_mtu, - .ndo_set_mac_address = efx_set_mac_address, - .ndo_set_rx_mode = efx_set_rx_mode, - .ndo_set_features = efx_set_features, -#ifdef CONFIG_NET_POLL_CONTROLLER - .ndo_poll_controller = efx_netpoll, -#endif -#ifdef CONFIG_NET_RX_BUSY_POLL - .ndo_busy_poll = efx_busy_poll, -#endif -#ifdef CONFIG_RFS_ACCEL - .ndo_rx_flow_steer = efx_filter_rfs, -#endif -}; - static void efx_update_name(struct efx_nic *efx) { strcpy(efx->name, efx->net_dev->name); @@ -2264,8 +2307,7 @@ static int efx_netdev_event(struct notifier_block *this, { struct net_device *net_dev = netdev_notifier_info_to_dev(ptr); - if ((net_dev->netdev_ops == &efx_farch_netdev_ops || - net_dev->netdev_ops == &efx_ef10_netdev_ops) && + if ((net_dev->netdev_ops == &efx_netdev_ops) && event == NETDEV_CHANGENAME) efx_update_name(netdev_priv(net_dev)); @@ -2284,6 +2326,28 @@ show_phy_type(struct device *dev, struct device_attribute *attr, char *buf) } static DEVICE_ATTR(phy_type, 0444, show_phy_type, NULL); +#ifdef CONFIG_SFC_MCDI_LOGGING +static ssize_t show_mcdi_log(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev)); + struct efx_mcdi_iface *mcdi = efx_mcdi(efx); + + return scnprintf(buf, PAGE_SIZE, "%d\n", mcdi->logging_enabled); +} +static ssize_t set_mcdi_log(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev)); + struct efx_mcdi_iface *mcdi = efx_mcdi(efx); + bool enable = count > 0 && *buf != '0'; + + mcdi->logging_enabled = enable; + return count; +} +static DEVICE_ATTR(mcdi_logging, 0644, show_mcdi_log, set_mcdi_log); +#endif + static int efx_register_netdev(struct efx_nic *efx) { struct net_device *net_dev = efx->net_dev; @@ -2292,12 +2356,9 @@ static int efx_register_netdev(struct efx_nic *efx) net_dev->watchdog_timeo = 5 * HZ; net_dev->irq = efx->pci_dev->irq; - if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) { - net_dev->netdev_ops = &efx_ef10_netdev_ops; + net_dev->netdev_ops = &efx_netdev_ops; + if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) net_dev->priv_flags |= IFF_UNICAST_FLT; - } else { - net_dev->netdev_ops = &efx_farch_netdev_ops; - } net_dev->ethtool_ops = &efx_ethtool_ops; net_dev->gso_max_segs = EFX_TSO_MAX_SEGS; @@ -2344,9 +2405,21 @@ static int efx_register_netdev(struct efx_nic *efx) "failed to init net dev attributes\n"); goto fail_registered; } +#ifdef CONFIG_SFC_MCDI_LOGGING + rc = device_create_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging); + if (rc) { + netif_err(efx, drv, efx->net_dev, + "failed to init net dev attributes\n"); + goto fail_attr_mcdi_logging; + } +#endif return 0; +#ifdef CONFIG_SFC_MCDI_LOGGING +fail_attr_mcdi_logging: + device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type); +#endif fail_registered: rtnl_lock(); efx_dissociate(efx); @@ -2365,13 +2438,14 @@ static void efx_unregister_netdev(struct efx_nic *efx) BUG_ON(netdev_priv(efx->net_dev) != efx); - strlcpy(efx->name, pci_name(efx->pci_dev), sizeof(efx->name)); - device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type); - - rtnl_lock(); - unregister_netdevice(efx->net_dev); - efx->state = STATE_UNINIT; - rtnl_unlock(); + if (efx_dev_registered(efx)) { + strlcpy(efx->name, pci_name(efx->pci_dev), sizeof(efx->name)); +#ifdef CONFIG_SFC_MCDI_LOGGING + device_remove_file(&efx->pci_dev->dev, &dev_attr_mcdi_logging); +#endif + device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type); + unregister_netdev(efx->net_dev); + } } /************************************************************************** @@ -2393,7 +2467,8 @@ void efx_reset_down(struct efx_nic *efx, enum reset_type method) efx_disable_interrupts(efx); mutex_lock(&efx->mac_lock); - if (efx->port_initialized && method != RESET_TYPE_INVISIBLE) + if (efx->port_initialized && method != RESET_TYPE_INVISIBLE && + method != RESET_TYPE_DATAPATH) efx->phy_op->fini(efx); efx->type->fini(efx); } @@ -2422,11 +2497,13 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok) if (!ok) goto fail; - if (efx->port_initialized && method != RESET_TYPE_INVISIBLE) { + if (efx->port_initialized && method != RESET_TYPE_INVISIBLE && + method != RESET_TYPE_DATAPATH) { rc = efx->phy_op->init(efx); if (rc) goto fail; - if (efx->phy_op->reconfigure(efx)) + rc = efx->phy_op->reconfigure(efx); + if (rc && rc != -EPERM) netif_err(efx, drv, efx->net_dev, "could not restore PHY settings\n"); } @@ -2434,8 +2511,20 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok) rc = efx_enable_interrupts(efx); if (rc) goto fail; + +#ifdef CONFIG_SFC_SRIOV + rc = efx->type->vswitching_restore(efx); + if (rc) /* not fatal; the PF will still work fine */ + netif_warn(efx, probe, efx->net_dev, + "failed to restore vswitching rc=%d;" + " VFs may not function\n", rc); +#endif + + down_read(&efx->filter_sem); efx_restore_filters(efx); - efx->type->sriov_reset(efx); + up_read(&efx->filter_sem); + if (efx->type->sriov_reset) + efx->type->sriov_reset(efx); mutex_unlock(&efx->mac_lock); @@ -2605,6 +2694,7 @@ void efx_schedule_reset(struct efx_nic *efx, enum reset_type type) case RESET_TYPE_WORLD: case RESET_TYPE_DISABLE: case RESET_TYPE_RECOVER_OR_DISABLE: + case RESET_TYPE_DATAPATH: case RESET_TYPE_MC_BIST: case RESET_TYPE_MCDI_TIMEOUT: method = type; @@ -2655,6 +2745,8 @@ static const struct pci_device_id efx_pci_table[] = { .driver_data = (unsigned long) &siena_a0_nic_type}, {PCI_DEVICE(PCI_VENDOR_ID_SOLARFLARE, 0x0903), /* SFC9120 PF */ .driver_data = (unsigned long) &efx_hunt_a0_nic_type}, + {PCI_DEVICE(PCI_VENDOR_ID_SOLARFLARE, 0x1903), /* SFC9120 VF */ + .driver_data = (unsigned long) &efx_hunt_a0_vf_nic_type}, {PCI_DEVICE(PCI_VENDOR_ID_SOLARFLARE, 0x0923), /* SFC9140 PF */ .driver_data = (unsigned long) &efx_hunt_a0_nic_type}, {0} /* end of list */ @@ -2826,7 +2918,9 @@ static void efx_pci_remove(struct pci_dev *pci_dev) efx_disable_interrupts(efx); rtnl_unlock(); - efx->type->sriov_fini(efx); + if (efx->type->sriov_fini) + efx->type->sriov_fini(efx); + efx_unregister_netdev(efx); efx_mtd_remove(efx); @@ -3008,7 +3102,8 @@ static int efx_pci_probe(struct pci_dev *pci_dev, netif_info(efx, probe, efx->net_dev, "Solarflare NIC detected\n"); - efx_probe_vpd_strings(efx); + if (!efx->type->is_vf) + efx_probe_vpd_strings(efx); /* Set up basic I/O (BAR mappings etc) */ rc = efx_init_io(efx); @@ -3023,10 +3118,12 @@ static int efx_pci_probe(struct pci_dev *pci_dev, if (rc) goto fail4; - rc = efx->type->sriov_init(efx); - if (rc) - netif_err(efx, probe, efx->net_dev, - "SR-IOV can't be enabled rc %d\n", rc); + if (efx->type->sriov_init) { + rc = efx->type->sriov_init(efx); + if (rc) + netif_err(efx, probe, efx->net_dev, + "SR-IOV can't be enabled rc %d\n", rc); + } netif_dbg(efx, probe, efx->net_dev, "initialisation successful\n"); @@ -3058,6 +3155,26 @@ static int efx_pci_probe(struct pci_dev *pci_dev, return rc; } +/* efx_pci_sriov_configure returns the actual number of Virtual Functions + * enabled on success + */ +#ifdef CONFIG_SFC_SRIOV +static int efx_pci_sriov_configure(struct pci_dev *dev, int num_vfs) +{ + int rc; + struct efx_nic *efx = pci_get_drvdata(dev); + + if (efx->type->sriov_configure) { + rc = efx->type->sriov_configure(efx, num_vfs); + if (rc) + return rc; + else + return num_vfs; + } else + return -EOPNOTSUPP; +} +#endif + static int efx_pm_freeze(struct device *dev) { struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev)); @@ -3280,6 +3397,9 @@ static struct pci_driver efx_pci_driver = { .remove = efx_pci_remove, .driver.pm = &efx_pm_ops, .err_handler = &efx_err_handlers, +#ifdef CONFIG_SFC_SRIOV + .sriov_configure = efx_pci_sriov_configure, +#endif }; /************************************************************************** @@ -3302,9 +3422,11 @@ static int __init efx_init_module(void) if (rc) goto err_notifier; +#ifdef CONFIG_SFC_SRIOV rc = efx_init_sriov(); if (rc) goto err_sriov; +#endif reset_workqueue = create_singlethread_workqueue("sfc_reset"); if (!reset_workqueue) { @@ -3321,8 +3443,10 @@ static int __init efx_init_module(void) err_pci: destroy_workqueue(reset_workqueue); err_reset: +#ifdef CONFIG_SFC_SRIOV efx_fini_sriov(); err_sriov: +#endif unregister_netdevice_notifier(&efx_netdev_notifier); err_notifier: return rc; @@ -3334,7 +3458,9 @@ static void __exit efx_exit_module(void) pci_unregister_driver(&efx_pci_driver); destroy_workqueue(reset_workqueue); +#ifdef CONFIG_SFC_SRIOV efx_fini_sriov(); +#endif unregister_netdevice_notifier(&efx_netdev_notifier); } diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index 2587c582a821..acb1e0718485 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -15,7 +15,12 @@ #include "filter.h" /* All controllers use BAR 0 for I/O space and BAR 2(&3) for memory */ +/* All VFs use BAR 0/1 for memory */ #define EFX_MEM_BAR 2 +#define EFX_MEM_VF_BAR 0 + +int efx_net_open(struct net_device *net_dev); +int efx_net_stop(struct net_device *net_dev); /* TX */ int efx_probe_tx_queue(struct efx_tx_queue *tx_queue); @@ -32,6 +37,7 @@ unsigned int efx_tx_max_skb_descs(struct efx_nic *efx); extern unsigned int efx_piobuf_size; /* RX */ +void efx_set_default_rx_indir_table(struct efx_nic *efx); void efx_rx_config_page_split(struct efx_nic *efx); int efx_probe_rx_queue(struct efx_rx_queue *rx_queue); void efx_remove_rx_queue(struct efx_rx_queue *rx_queue); @@ -71,6 +77,8 @@ void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue); /* Filters */ +void efx_mac_reconfigure(struct efx_nic *efx); + /** * efx_filter_insert_filter - add or replace a filter * @efx: NIC in which to insert the filter @@ -220,6 +228,13 @@ static inline void efx_mtd_rename(struct efx_nic *efx) {} static inline void efx_mtd_remove(struct efx_nic *efx) {} #endif +#ifdef CONFIG_SFC_SRIOV +static inline unsigned int efx_vf_size(struct efx_nic *efx) +{ + return 1 << efx->vi_scale; +} +#endif + static inline void efx_schedule_channel(struct efx_channel *channel) { netif_vdbg(channel->efx, intr, channel->efx->net_dev, diff --git a/drivers/net/ethernet/sfc/enum.h b/drivers/net/ethernet/sfc/enum.h index d1dbb5fb31bb..c94f56271dd4 100644 --- a/drivers/net/ethernet/sfc/enum.h +++ b/drivers/net/ethernet/sfc/enum.h @@ -143,6 +143,7 @@ enum efx_loopback_mode { * @RESET_TYPE_WORLD: Reset as much as possible * @RESET_TYPE_RECOVER_OR_DISABLE: Try to recover. Apply RESET_TYPE_DISABLE if * unsuccessful. + * @RESET_TYPE_DATAPATH: Reset datapath only. * @RESET_TYPE_MC_BIST: MC entering BIST mode. * @RESET_TYPE_DISABLE: Reset datapath, MAC and PHY; leave NIC disabled * @RESET_TYPE_TX_WATCHDOG: reset due to TX watchdog @@ -159,6 +160,7 @@ enum reset_type { RESET_TYPE_ALL, RESET_TYPE_WORLD, RESET_TYPE_RECOVER_OR_DISABLE, + RESET_TYPE_DATAPATH, RESET_TYPE_MC_BIST, RESET_TYPE_DISABLE, RESET_TYPE_MAX_METHOD, diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c index 4835bc0d0de8..034797661f96 100644 --- a/drivers/net/ethernet/sfc/ethtool.c +++ b/drivers/net/ethernet/sfc/ethtool.c @@ -734,7 +734,7 @@ static int efx_ethtool_set_pauseparam(struct net_device *net_dev, /* Reconfigure the MAC. The PHY *may* generate a link state change event * if the user just changed the advertised capabilities, but there's no * harm doing this twice */ - efx->type->reconfigure_mac(efx); + efx_mac_reconfigure(efx); out: mutex_unlock(&efx->mac_lock); @@ -1109,9 +1109,8 @@ static int efx_ethtool_set_rxfh(struct net_device *net_dev, const u32 *indir, return -EOPNOTSUPP; if (!indir) return 0; - memcpy(efx->rx_indir_table, indir, sizeof(efx->rx_indir_table)); - efx->type->rx_push_rss_config(efx); - return 0; + + return efx->type->rx_push_rss_config(efx, true, indir); } static int efx_ethtool_get_ts_info(struct net_device *net_dev, diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c index f166c8ef38a3..80e69af21642 100644 --- a/drivers/net/ethernet/sfc/falcon.c +++ b/drivers/net/ethernet/sfc/falcon.c @@ -477,16 +477,29 @@ static irqreturn_t falcon_legacy_interrupt_a1(int irq, void *dev_id) * ************************************************************************** */ +static int dummy_rx_push_rss_config(struct efx_nic *efx, bool user, + const u32 *rx_indir_table) +{ + (void) efx; + (void) user; + (void) rx_indir_table; + return -ENOSYS; +} -static void falcon_b0_rx_push_rss_config(struct efx_nic *efx) +static int falcon_b0_rx_push_rss_config(struct efx_nic *efx, bool user, + const u32 *rx_indir_table) { efx_oword_t temp; + (void) user; /* Set hash key for IPv4 */ memcpy(&temp, efx->rx_hash_key, sizeof(temp)); efx_writeo(efx, &temp, FR_BZ_RX_RSS_TKEY); + memcpy(efx->rx_indir_table, rx_indir_table, + sizeof(efx->rx_indir_table)); efx_farch_rx_push_indir_table(efx); + return 0; } /************************************************************************** @@ -2507,7 +2520,7 @@ static int falcon_init_nic(struct efx_nic *efx) falcon_init_rx_cfg(efx); if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) { - falcon_b0_rx_push_rss_config(efx); + falcon_b0_rx_push_rss_config(efx, false, efx->rx_indir_table); /* Set destination of both TX and RX Flush events */ EFX_POPULATE_OWORD_1(temp, FRF_BZ_FLS_EVQ_ID, 0); @@ -2687,6 +2700,8 @@ static int falcon_set_wol(struct efx_nic *efx, u32 type) */ const struct efx_nic_type falcon_a1_nic_type = { + .is_vf = false, + .mem_bar = EFX_MEM_BAR, .mem_map_size = falcon_a1_mem_map_size, .probe = falcon_probe_nic, .remove = falcon_remove_nic, @@ -2729,7 +2744,7 @@ const struct efx_nic_type falcon_a1_nic_type = { .tx_init = efx_farch_tx_init, .tx_remove = efx_farch_tx_remove, .tx_write = efx_farch_tx_write, - .rx_push_rss_config = efx_port_dummy_op_void, + .rx_push_rss_config = dummy_rx_push_rss_config, .rx_probe = efx_farch_rx_probe, .rx_init = efx_farch_rx_init, .rx_remove = efx_farch_rx_remove, @@ -2766,11 +2781,6 @@ const struct efx_nic_type falcon_a1_nic_type = { .mtd_write = falcon_mtd_write, .mtd_sync = falcon_mtd_sync, #endif - .sriov_init = efx_falcon_sriov_init, - .sriov_fini = efx_falcon_sriov_fini, - .sriov_mac_address_changed = efx_falcon_sriov_mac_address_changed, - .sriov_wanted = efx_falcon_sriov_wanted, - .sriov_reset = efx_falcon_sriov_reset, .revision = EFX_REV_FALCON_A1, .txd_ptr_tbl_base = FR_AA_TX_DESC_PTR_TBL_KER, @@ -2788,6 +2798,8 @@ const struct efx_nic_type falcon_a1_nic_type = { }; const struct efx_nic_type falcon_b0_nic_type = { + .is_vf = false, + .mem_bar = EFX_MEM_BAR, .mem_map_size = falcon_b0_mem_map_size, .probe = falcon_probe_nic, .remove = falcon_remove_nic, @@ -2867,11 +2879,6 @@ const struct efx_nic_type falcon_b0_nic_type = { .mtd_write = falcon_mtd_write, .mtd_sync = falcon_mtd_sync, #endif - .sriov_init = efx_falcon_sriov_init, - .sriov_fini = efx_falcon_sriov_fini, - .sriov_mac_address_changed = efx_falcon_sriov_mac_address_changed, - .sriov_wanted = efx_falcon_sriov_wanted, - .sriov_reset = efx_falcon_sriov_reset, .revision = EFX_REV_FALCON_B0, .txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL, diff --git a/drivers/net/ethernet/sfc/farch.c b/drivers/net/ethernet/sfc/farch.c index bb89e96a125e..f08266f0eca2 100644 --- a/drivers/net/ethernet/sfc/farch.c +++ b/drivers/net/ethernet/sfc/farch.c @@ -20,6 +20,8 @@ #include "efx.h" #include "nic.h" #include "farch_regs.h" +#include "sriov.h" +#include "siena_sriov.h" #include "io.h" #include "workarounds.h" @@ -1198,13 +1200,17 @@ efx_farch_handle_driver_event(struct efx_channel *channel, efx_qword_t *event) netif_vdbg(efx, hw, efx->net_dev, "channel %d TXQ %d flushed\n", channel->channel, ev_sub_data); efx_farch_handle_tx_flush_done(efx, event); +#ifdef CONFIG_SFC_SRIOV efx_siena_sriov_tx_flush_done(efx, event); +#endif break; case FSE_AZ_RX_DESCQ_FLS_DONE_EV: netif_vdbg(efx, hw, efx->net_dev, "channel %d RXQ %d flushed\n", channel->channel, ev_sub_data); efx_farch_handle_rx_flush_done(efx, event); +#ifdef CONFIG_SFC_SRIOV efx_siena_sriov_rx_flush_done(efx, event); +#endif break; case FSE_AZ_EVQ_INIT_DONE_EV: netif_dbg(efx, hw, efx->net_dev, @@ -1242,8 +1248,11 @@ efx_farch_handle_driver_event(struct efx_channel *channel, efx_qword_t *event) " RX Q %d is disabled.\n", ev_sub_data, ev_sub_data); efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR); - } else + } +#ifdef CONFIG_SFC_SRIOV + else efx_siena_sriov_desc_fetch_err(efx, ev_sub_data); +#endif break; case FSE_BZ_TX_DSC_ERROR_EV: if (ev_sub_data < EFX_VI_BASE) { @@ -1252,8 +1261,11 @@ efx_farch_handle_driver_event(struct efx_channel *channel, efx_qword_t *event) " TX Q %d is disabled.\n", ev_sub_data, ev_sub_data); efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR); - } else + } +#ifdef CONFIG_SFC_SRIOV + else efx_siena_sriov_desc_fetch_err(efx, ev_sub_data); +#endif break; default: netif_vdbg(efx, hw, efx->net_dev, @@ -1317,9 +1329,11 @@ int efx_farch_ev_process(struct efx_channel *channel, int budget) case FSE_AZ_EV_CODE_DRIVER_EV: efx_farch_handle_driver_event(channel, &event); break; +#ifdef CONFIG_SFC_SRIOV case FSE_CZ_EV_CODE_USER_EV: efx_siena_sriov_event(channel, &event); break; +#endif case FSE_CZ_EV_CODE_MCDI_EV: efx_mcdi_process_event(channel, &event); break; @@ -1685,28 +1699,32 @@ void efx_farch_dimension_resources(struct efx_nic *efx, unsigned sram_lim_qw) vi_count = max(efx->n_channels, efx->n_tx_channels * EFX_TXQ_TYPES); #ifdef CONFIG_SFC_SRIOV - if (efx->type->sriov_wanted(efx)) { - unsigned vi_dc_entries, buftbl_free, entries_per_vf, vf_limit; - - nic_data->vf_buftbl_base = buftbl_min; - - vi_dc_entries = RX_DC_ENTRIES + TX_DC_ENTRIES; - vi_count = max(vi_count, EFX_VI_BASE); - buftbl_free = (sram_lim_qw - buftbl_min - - vi_count * vi_dc_entries); - - entries_per_vf = ((vi_dc_entries + EFX_VF_BUFTBL_PER_VI) * - efx_vf_size(efx)); - vf_limit = min(buftbl_free / entries_per_vf, - (1024U - EFX_VI_BASE) >> efx->vi_scale); - - if (efx->vf_count > vf_limit) { - netif_err(efx, probe, efx->net_dev, - "Reducing VF count from from %d to %d\n", - efx->vf_count, vf_limit); - efx->vf_count = vf_limit; + if (efx->type->sriov_wanted) { + if (efx->type->sriov_wanted(efx)) { + unsigned vi_dc_entries, buftbl_free; + unsigned entries_per_vf, vf_limit; + + nic_data->vf_buftbl_base = buftbl_min; + + vi_dc_entries = RX_DC_ENTRIES + TX_DC_ENTRIES; + vi_count = max(vi_count, EFX_VI_BASE); + buftbl_free = (sram_lim_qw - buftbl_min - + vi_count * vi_dc_entries); + + entries_per_vf = ((vi_dc_entries + + EFX_VF_BUFTBL_PER_VI) * + efx_vf_size(efx)); + vf_limit = min(buftbl_free / entries_per_vf, + (1024U - EFX_VI_BASE) >> efx->vi_scale); + + if (efx->vf_count > vf_limit) { + netif_err(efx, probe, efx->net_dev, + "Reducing VF count from from %d to %d\n", + efx->vf_count, vf_limit); + efx->vf_count = vf_limit; + } + vi_count += efx->vf_count * efx_vf_size(efx); } - vi_count += efx->vf_count * efx_vf_size(efx); } #endif diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c index d37928f01949..81640f8bb811 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -8,6 +8,7 @@ */ #include <linux/delay.h> +#include <linux/moduleparam.h> #include <asm/cmpxchg.h> #include "net_driver.h" #include "nic.h" @@ -54,18 +55,32 @@ static int efx_mcdi_drv_attach(struct efx_nic *efx, bool driver_operating, static bool efx_mcdi_poll_once(struct efx_nic *efx); static void efx_mcdi_abandon(struct efx_nic *efx); +#ifdef CONFIG_SFC_MCDI_LOGGING +static bool mcdi_logging_default; +module_param(mcdi_logging_default, bool, 0644); +MODULE_PARM_DESC(mcdi_logging_default, + "Enable MCDI logging on newly-probed functions"); +#endif + int efx_mcdi_init(struct efx_nic *efx) { struct efx_mcdi_iface *mcdi; bool already_attached; - int rc; + int rc = -ENOMEM; efx->mcdi = kzalloc(sizeof(*efx->mcdi), GFP_KERNEL); if (!efx->mcdi) - return -ENOMEM; + goto fail; mcdi = efx_mcdi(efx); mcdi->efx = efx; +#ifdef CONFIG_SFC_MCDI_LOGGING + /* consuming code assumes buffer is page-sized */ + mcdi->logging_buffer = (char *)__get_free_page(GFP_KERNEL); + if (!mcdi->logging_buffer) + goto fail1; + mcdi->logging_enabled = mcdi_logging_default; +#endif init_waitqueue_head(&mcdi->wq); spin_lock_init(&mcdi->iface_lock); mcdi->state = MCDI_STATE_QUIESCENT; @@ -81,7 +96,7 @@ int efx_mcdi_init(struct efx_nic *efx) /* Recover from a failed assertion before probing */ rc = efx_mcdi_handle_assertion(efx); if (rc) - return rc; + goto fail2; /* Let the MC (and BMC, if this is a LOM) know that the driver * is loaded. We should do this before we reset the NIC. @@ -90,7 +105,7 @@ int efx_mcdi_init(struct efx_nic *efx) if (rc) { netif_err(efx, probe, efx->net_dev, "Unable to register driver with MCPU\n"); - return rc; + goto fail2; } if (already_attached) /* Not a fatal error */ @@ -102,6 +117,15 @@ int efx_mcdi_init(struct efx_nic *efx) efx->primary = efx; return 0; +fail2: +#ifdef CONFIG_SFC_MCDI_LOGGING + free_page((unsigned long)mcdi->logging_buffer); +fail1: +#endif + kfree(efx->mcdi); + efx->mcdi = NULL; +fail: + return rc; } void efx_mcdi_fini(struct efx_nic *efx) @@ -114,6 +138,10 @@ void efx_mcdi_fini(struct efx_nic *efx) /* Relinquish the device (back to the BMC, if this is a LOM) */ efx_mcdi_drv_attach(efx, false, NULL); +#ifdef CONFIG_SFC_MCDI_LOGGING + free_page((unsigned long)efx->mcdi->iface.logging_buffer); +#endif + kfree(efx->mcdi); } @@ -121,6 +149,9 @@ static void efx_mcdi_send_request(struct efx_nic *efx, unsigned cmd, const efx_dword_t *inbuf, size_t inlen) { struct efx_mcdi_iface *mcdi = efx_mcdi(efx); +#ifdef CONFIG_SFC_MCDI_LOGGING + char *buf = mcdi->logging_buffer; /* page-sized */ +#endif efx_dword_t hdr[2]; size_t hdr_len; u32 xflags, seqno; @@ -165,6 +196,31 @@ static void efx_mcdi_send_request(struct efx_nic *efx, unsigned cmd, hdr_len = 8; } +#ifdef CONFIG_SFC_MCDI_LOGGING + if (mcdi->logging_enabled && !WARN_ON_ONCE(!buf)) { + int bytes = 0; + int i; + /* Lengths should always be a whole number of dwords, so scream + * if they're not. + */ + WARN_ON_ONCE(hdr_len % 4); + WARN_ON_ONCE(inlen % 4); + + /* We own the logging buffer, as only one MCDI can be in + * progress on a NIC at any one time. So no need for locking. + */ + for (i = 0; i < hdr_len / 4 && bytes < PAGE_SIZE; i++) + bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(hdr[i].u32[0])); + + for (i = 0; i < inlen / 4 && bytes < PAGE_SIZE; i++) + bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(inbuf[i].u32[0])); + + netif_info(efx, hw, efx->net_dev, "MCDI RPC REQ:%s\n", buf); + } +#endif + efx->type->mcdi_request(efx, hdr, hdr_len, inbuf, inlen); mcdi->new_epoch = false; @@ -206,6 +262,9 @@ static void efx_mcdi_read_response_header(struct efx_nic *efx) { struct efx_mcdi_iface *mcdi = efx_mcdi(efx); unsigned int respseq, respcmd, error; +#ifdef CONFIG_SFC_MCDI_LOGGING + char *buf = mcdi->logging_buffer; /* page-sized */ +#endif efx_dword_t hdr; efx->type->mcdi_read_response(efx, &hdr, 0, 4); @@ -223,6 +282,39 @@ static void efx_mcdi_read_response_header(struct efx_nic *efx) EFX_DWORD_FIELD(hdr, MC_CMD_V2_EXTN_IN_ACTUAL_LEN); } +#ifdef CONFIG_SFC_MCDI_LOGGING + if (mcdi->logging_enabled && !WARN_ON_ONCE(!buf)) { + size_t hdr_len, data_len; + int bytes = 0; + int i; + + WARN_ON_ONCE(mcdi->resp_hdr_len % 4); + hdr_len = mcdi->resp_hdr_len / 4; + /* MCDI_DECLARE_BUF ensures that underlying buffer is padded + * to dword size, and the MCDI buffer is always dword size + */ + data_len = DIV_ROUND_UP(mcdi->resp_data_len, 4); + + /* We own the logging buffer, as only one MCDI can be in + * progress on a NIC at any one time. So no need for locking. + */ + for (i = 0; i < hdr_len && bytes < PAGE_SIZE; i++) { + efx->type->mcdi_read_response(efx, &hdr, (i * 4), 4); + bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(hdr.u32[0])); + } + + for (i = 0; i < data_len && bytes < PAGE_SIZE; i++) { + efx->type->mcdi_read_response(efx, &hdr, + mcdi->resp_hdr_len + (i * 4), 4); + bytes += snprintf(buf + bytes, PAGE_SIZE - bytes, + " %08x", le32_to_cpu(hdr.u32[0])); + } + + netif_info(efx, hw, efx->net_dev, "MCDI RPC RESP:%s\n", buf); + } +#endif + if (error && mcdi->resp_data_len == 0) { netif_err(efx, hw, efx->net_dev, "MC rebooted\n"); mcdi->resprc = -EIO; @@ -406,7 +498,7 @@ static bool efx_mcdi_complete_async(struct efx_mcdi_iface *mcdi, bool timeout) struct efx_mcdi_async_param *async; size_t hdr_len, data_len, err_len; efx_dword_t *outbuf; - MCDI_DECLARE_BUF_OUT_OR_ERR(errbuf, 0); + MCDI_DECLARE_BUF_ERR(errbuf); int rc; if (cmpxchg(&mcdi->state, @@ -534,7 +626,7 @@ static int _efx_mcdi_rpc_finish(struct efx_nic *efx, unsigned cmd, size_t inlen, size_t *outlen_actual, bool quiet) { struct efx_mcdi_iface *mcdi = efx_mcdi(efx); - MCDI_DECLARE_BUF_OUT_OR_ERR(errbuf, 0); + MCDI_DECLARE_BUF_ERR(errbuf); int rc; if (mcdi->mode == MCDI_MODE_POLL) @@ -1035,7 +1127,9 @@ void efx_mcdi_process_event(struct efx_channel *channel, /* MAC stats are gather lazily. We can ignore this. */ break; case MCDI_EVENT_CODE_FLR: - efx_siena_sriov_flr(efx, MCDI_EVENT_FIELD(*event, FLR_VF)); + if (efx->type->sriov_flr) + efx->type->sriov_flr(efx, + MCDI_EVENT_FIELD(*event, FLR_VF)); break; case MCDI_EVENT_CODE_PTP_RX: case MCDI_EVENT_CODE_PTP_FAULT: @@ -1081,9 +1175,7 @@ void efx_mcdi_process_event(struct efx_channel *channel, void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len) { - MCDI_DECLARE_BUF(outbuf, - max(MC_CMD_GET_VERSION_OUT_LEN, - MC_CMD_GET_CAPABILITIES_OUT_LEN)); + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_VERSION_OUT_LEN); size_t outlength; const __le16 *ver_words; size_t offset; @@ -1108,19 +1200,11 @@ void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len) * single version. Report which variants are running. */ if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) { - BUILD_BUG_ON(MC_CMD_GET_CAPABILITIES_IN_LEN != 0); - rc = efx_mcdi_rpc(efx, MC_CMD_GET_CAPABILITIES, NULL, 0, - outbuf, sizeof(outbuf), &outlength); - if (rc || outlength < MC_CMD_GET_CAPABILITIES_OUT_LEN) - offset += snprintf( - buf + offset, len - offset, " rx? tx?"); - else - offset += snprintf( - buf + offset, len - offset, " rx%x tx%x", - MCDI_WORD(outbuf, - GET_CAPABILITIES_OUT_RX_DPCPU_FW_ID), - MCDI_WORD(outbuf, - GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID)); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + offset += snprintf(buf + offset, len - offset, " rx%x tx%x", + nic_data->rx_dpcpu_fw_id, + nic_data->tx_dpcpu_fw_id); /* It's theoretically possible for the string to exceed 31 * characters, though in practice the first three version @@ -1150,10 +1234,26 @@ static int efx_mcdi_drv_attach(struct efx_nic *efx, bool driver_operating, MCDI_SET_DWORD(inbuf, DRV_ATTACH_IN_UPDATE, 1); MCDI_SET_DWORD(inbuf, DRV_ATTACH_IN_FIRMWARE_ID, MC_CMD_FW_LOW_LATENCY); - rc = efx_mcdi_rpc(efx, MC_CMD_DRV_ATTACH, inbuf, sizeof(inbuf), - outbuf, sizeof(outbuf), &outlen); - if (rc) + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_DRV_ATTACH, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &outlen); + /* If we're not the primary PF, trying to ATTACH with a FIRMWARE_ID + * specified will fail with EPERM, and we have to tell the MC we don't + * care what firmware we get. + */ + if (rc == -EPERM) { + netif_dbg(efx, probe, efx->net_dev, + "efx_mcdi_drv_attach with fw-variant setting failed EPERM, trying without it\n"); + MCDI_SET_DWORD(inbuf, DRV_ATTACH_IN_FIRMWARE_ID, + MC_CMD_FW_DONT_CARE); + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_DRV_ATTACH, inbuf, + sizeof(inbuf), outbuf, sizeof(outbuf), + &outlen); + } + if (rc) { + efx_mcdi_display_error(efx, MC_CMD_DRV_ATTACH, sizeof(inbuf), + outbuf, outlen, rc); goto fail; + } if (outlen < MC_CMD_DRV_ATTACH_OUT_LEN) { rc = -EIO; goto fail; @@ -1178,16 +1278,6 @@ static int efx_mcdi_drv_attach(struct efx_nic *efx, bool driver_operating, * and are completely trusted by firmware. Abort probing * if that's not true for this function. */ - if (driver_operating && - (efx->mcdi->fn_flags & - (1 << MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_LINKCTRL | - 1 << MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_TRUSTED)) != - (1 << MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_LINKCTRL | - 1 << MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_TRUSTED)) { - netif_err(efx, probe, efx->net_dev, - "This driver version only supports one function per port\n"); - return -ENODEV; - } if (was_attached != NULL) *was_attached = MCDI_DWORD(outbuf, DRV_ATTACH_OUT_OLD_STATE); @@ -1385,10 +1475,13 @@ fail1: return rc; } +/* Returns 1 if an assertion was read, 0 if no assertion had fired, + * negative on error. + */ static int efx_mcdi_read_assertion(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_ASSERTS_IN_LEN); - MCDI_DECLARE_BUF_OUT_OR_ERR(outbuf, MC_CMD_GET_ASSERTS_OUT_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_ASSERTS_OUT_LEN); unsigned int flags, index; const char *reason; size_t outlen; @@ -1406,6 +1499,8 @@ static int efx_mcdi_read_assertion(struct efx_nic *efx) rc = efx_mcdi_rpc_quiet(efx, MC_CMD_GET_ASSERTS, inbuf, MC_CMD_GET_ASSERTS_IN_LEN, outbuf, sizeof(outbuf), &outlen); + if (rc == -EPERM) + return 0; } while ((rc == -EINTR || rc == -EIO) && retry-- > 0); if (rc) { @@ -1443,24 +1538,31 @@ static int efx_mcdi_read_assertion(struct efx_nic *efx) MCDI_ARRAY_DWORD(outbuf, GET_ASSERTS_OUT_GP_REGS_OFFS, index)); - return 0; + return 1; } -static void efx_mcdi_exit_assertion(struct efx_nic *efx) +static int efx_mcdi_exit_assertion(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_REBOOT_IN_LEN); + int rc; /* If the MC is running debug firmware, it might now be * waiting for a debugger to attach, but we just want it to * reboot. We set a flag that makes the command a no-op if it - * has already done so. We don't know what return code to - * expect (0 or -EIO), so ignore it. + * has already done so. + * The MCDI will thus return either 0 or -EIO. */ BUILD_BUG_ON(MC_CMD_REBOOT_OUT_LEN != 0); MCDI_SET_DWORD(inbuf, REBOOT_IN_FLAGS, MC_CMD_REBOOT_FLAGS_AFTER_ASSERTION); - (void) efx_mcdi_rpc(efx, MC_CMD_REBOOT, inbuf, MC_CMD_REBOOT_IN_LEN, - NULL, 0, NULL); + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_REBOOT, inbuf, MC_CMD_REBOOT_IN_LEN, + NULL, 0, NULL); + if (rc == -EIO) + rc = 0; + if (rc) + efx_mcdi_display_error(efx, MC_CMD_REBOOT, MC_CMD_REBOOT_IN_LEN, + NULL, 0, rc); + return rc; } int efx_mcdi_handle_assertion(struct efx_nic *efx) @@ -1468,12 +1570,10 @@ int efx_mcdi_handle_assertion(struct efx_nic *efx) int rc; rc = efx_mcdi_read_assertion(efx); - if (rc) + if (rc <= 0) return rc; - efx_mcdi_exit_assertion(efx); - - return 0; + return efx_mcdi_exit_assertion(efx); } void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode) @@ -1550,7 +1650,9 @@ int efx_mcdi_reset(struct efx_nic *efx, enum reset_type method) if (rc) return rc; - if (method == RESET_TYPE_WORLD) + if (method == RESET_TYPE_DATAPATH) + return 0; + else if (method == RESET_TYPE_WORLD) return efx_mcdi_reset_mc(efx); else return efx_mcdi_reset_func(efx); @@ -1688,6 +1790,36 @@ int efx_mcdi_set_workaround(struct efx_nic *efx, u32 type, bool enabled) NULL, 0, NULL); } +int efx_mcdi_get_workarounds(struct efx_nic *efx, unsigned int *impl_out, + unsigned int *enabled_out) +{ + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_WORKAROUNDS_OUT_LEN); + size_t outlen; + int rc; + + rc = efx_mcdi_rpc(efx, MC_CMD_GET_WORKAROUNDS, NULL, 0, + outbuf, sizeof(outbuf), &outlen); + if (rc) + goto fail; + + if (outlen < MC_CMD_GET_WORKAROUNDS_OUT_LEN) { + rc = -EIO; + goto fail; + } + + if (impl_out) + *impl_out = MCDI_DWORD(outbuf, GET_WORKAROUNDS_OUT_IMPLEMENTED); + + if (enabled_out) + *enabled_out = MCDI_DWORD(outbuf, GET_WORKAROUNDS_OUT_ENABLED); + + return 0; + +fail: + netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc); + return rc; +} + #ifdef CONFIG_SFC_MTD #define EFX_MCDI_NVRAM_LEN_MAX 128 diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h index 56465f7465a2..1838afe2da92 100644 --- a/drivers/net/ethernet/sfc/mcdi.h +++ b/drivers/net/ethernet/sfc/mcdi.h @@ -58,6 +58,8 @@ enum efx_mcdi_mode { * enabled * @async_list: Queue of asynchronous requests * @async_timer: Timer for asynchronous request timeout + * @logging_buffer: buffer that may be used to build MCDI tracing messages + * @logging_enabled: whether to trace MCDI */ struct efx_mcdi_iface { struct efx_nic *efx; @@ -74,6 +76,10 @@ struct efx_mcdi_iface { spinlock_t async_lock; struct list_head async_list; struct timer_list async_timer; +#ifdef CONFIG_SFC_MCDI_LOGGING + char *logging_buffer; + bool logging_enabled; +#endif }; struct efx_mcdi_mon { @@ -176,10 +182,12 @@ void efx_mcdi_sensor_event(struct efx_nic *efx, efx_qword_t *ev); * 32-bit-aligned. Also, on Siena we must copy to the MC shared * memory strictly 32 bits at a time, so add any necessary padding. */ -#define MCDI_DECLARE_BUF(_name, _len) \ +#define _MCDI_DECLARE_BUF(_name, _len) \ efx_dword_t _name[DIV_ROUND_UP(_len, 4)] -#define MCDI_DECLARE_BUF_OUT_OR_ERR(_name, _len) \ - MCDI_DECLARE_BUF(_name, max_t(size_t, _len, 8)) +#define MCDI_DECLARE_BUF(_name, _len) \ + _MCDI_DECLARE_BUF(_name, _len) = {{{0}}} +#define MCDI_DECLARE_BUF_ERR(_name) \ + MCDI_DECLARE_BUF(_name, 8) #define _MCDI_PTR(_buf, _offset) \ ((u8 *)(_buf) + (_offset)) #define MCDI_PTR(_buf, _field) \ @@ -339,6 +347,8 @@ bool efx_mcdi_mac_check_fault(struct efx_nic *efx); enum reset_type efx_mcdi_map_reset_reason(enum reset_type reason); int efx_mcdi_reset(struct efx_nic *efx, enum reset_type method); int efx_mcdi_set_workaround(struct efx_nic *efx, u32 type, bool enabled); +int efx_mcdi_get_workarounds(struct efx_nic *efx, unsigned int *impl_out, + unsigned int *enabled_out); #ifdef CONFIG_SFC_MCDI_MON int efx_mcdi_mon_probe(struct efx_nic *efx); diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h index e028de10e1b7..9efdf0a5df64 100644 --- a/drivers/net/ethernet/sfc/mcdi_pcol.h +++ b/drivers/net/ethernet/sfc/mcdi_pcol.h @@ -638,6 +638,8 @@ */ #define MC_CMD_READ32 0x1 +#define MC_CMD_0x1_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_READ32_IN msgrequest */ #define MC_CMD_READ32_IN_LEN 8 #define MC_CMD_READ32_IN_ADDR_OFST 0 @@ -659,6 +661,8 @@ */ #define MC_CMD_WRITE32 0x2 +#define MC_CMD_0x2_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_WRITE32_IN msgrequest */ #define MC_CMD_WRITE32_IN_LENMIN 8 #define MC_CMD_WRITE32_IN_LENMAX 252 @@ -679,6 +683,8 @@ */ #define MC_CMD_COPYCODE 0x3 +#define MC_CMD_0x3_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_COPYCODE_IN msgrequest */ #define MC_CMD_COPYCODE_IN_LEN 16 /* Source address */ @@ -717,6 +723,8 @@ */ #define MC_CMD_SET_FUNC 0x4 +#define MC_CMD_0x4_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_FUNC_IN msgrequest */ #define MC_CMD_SET_FUNC_IN_LEN 4 /* Set function */ @@ -732,6 +740,8 @@ */ #define MC_CMD_GET_BOOT_STATUS 0x5 +#define MC_CMD_0x5_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_BOOT_STATUS_IN msgrequest */ #define MC_CMD_GET_BOOT_STATUS_IN_LEN 0 @@ -758,6 +768,8 @@ */ #define MC_CMD_GET_ASSERTS 0x6 +#define MC_CMD_0x6_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_ASSERTS_IN msgrequest */ #define MC_CMD_GET_ASSERTS_IN_LEN 4 /* Set to clear assertion */ @@ -794,6 +806,8 @@ */ #define MC_CMD_LOG_CTRL 0x7 +#define MC_CMD_0x7_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_LOG_CTRL_IN msgrequest */ #define MC_CMD_LOG_CTRL_IN_LEN 8 /* Log destination */ @@ -814,6 +828,8 @@ */ #define MC_CMD_GET_VERSION 0x8 +#define MC_CMD_0x8_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_VERSION_IN msgrequest */ #define MC_CMD_GET_VERSION_IN_LEN 0 @@ -870,6 +886,8 @@ */ #define MC_CMD_PTP 0xb +#define MC_CMD_0xb_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_PTP_IN msgrequest */ #define MC_CMD_PTP_IN_LEN 1 /* PTP operation code */ @@ -1404,6 +1422,8 @@ */ #define MC_CMD_CSR_READ32 0xc +#define MC_CMD_0xc_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_CSR_READ32_IN msgrequest */ #define MC_CMD_CSR_READ32_IN_LEN 12 /* Address */ @@ -1428,6 +1448,8 @@ */ #define MC_CMD_CSR_WRITE32 0xd +#define MC_CMD_0xd_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_CSR_WRITE32_IN msgrequest */ #define MC_CMD_CSR_WRITE32_IN_LENMIN 12 #define MC_CMD_CSR_WRITE32_IN_LENMAX 252 @@ -1452,6 +1474,8 @@ */ #define MC_CMD_HP 0x54 +#define MC_CMD_0x54_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_HP_IN msgrequest */ #define MC_CMD_HP_IN_LEN 16 /* HP OCSD sub-command. When address is not NULL, request activation of OCSD at @@ -1493,6 +1517,8 @@ */ #define MC_CMD_STACKINFO 0xf +#define MC_CMD_0xf_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_STACKINFO_IN msgrequest */ #define MC_CMD_STACKINFO_IN_LEN 0 @@ -1513,6 +1539,8 @@ */ #define MC_CMD_MDIO_READ 0x10 +#define MC_CMD_0x10_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_MDIO_READ_IN msgrequest */ #define MC_CMD_MDIO_READ_IN_LEN 16 /* Bus number; there are two MDIO buses: one for the internal PHY, and one for @@ -1552,6 +1580,8 @@ */ #define MC_CMD_MDIO_WRITE 0x11 +#define MC_CMD_0x11_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_MDIO_WRITE_IN msgrequest */ #define MC_CMD_MDIO_WRITE_IN_LEN 20 /* Bus number; there are two MDIO buses: one for the internal PHY, and one for @@ -1591,6 +1621,8 @@ */ #define MC_CMD_DBI_WRITE 0x12 +#define MC_CMD_0x12_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DBI_WRITE_IN msgrequest */ #define MC_CMD_DBI_WRITE_IN_LENMIN 12 #define MC_CMD_DBI_WRITE_IN_LENMAX 252 @@ -1739,6 +1771,8 @@ */ #define MC_CMD_GET_BOARD_CFG 0x18 +#define MC_CMD_0x18_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_BOARD_CFG_IN msgrequest */ #define MC_CMD_GET_BOARD_CFG_IN_LEN 0 @@ -1778,6 +1812,8 @@ */ #define MC_CMD_DBI_READX 0x19 +#define MC_CMD_0x19_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DBI_READX_IN msgrequest */ #define MC_CMD_DBI_READX_IN_LENMIN 8 #define MC_CMD_DBI_READX_IN_LENMAX 248 @@ -1822,6 +1858,8 @@ */ #define MC_CMD_SET_RAND_SEED 0x1a +#define MC_CMD_0x1a_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_RAND_SEED_IN msgrequest */ #define MC_CMD_SET_RAND_SEED_IN_LEN 16 /* Seed value. */ @@ -1863,6 +1901,8 @@ */ #define MC_CMD_DRV_ATTACH 0x1c +#define MC_CMD_0x1c_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_DRV_ATTACH_IN msgrequest */ #define MC_CMD_DRV_ATTACH_IN_LEN 12 /* new state (0=detached, 1=attached) to set if UPDATE=1 */ @@ -1875,6 +1915,8 @@ #define MC_CMD_FW_FULL_FEATURED 0x0 /* enum: Prefer to use firmware with fewer features but lower latency */ #define MC_CMD_FW_LOW_LATENCY 0x1 +/* enum: Only this option is allowed for non-admin functions */ +#define MC_CMD_FW_DONT_CARE 0xffffffff /* MC_CMD_DRV_ATTACH_OUT msgresponse */ #define MC_CMD_DRV_ATTACH_OUT_LEN 4 @@ -1920,6 +1962,8 @@ */ #define MC_CMD_PORT_RESET 0x20 +#define MC_CMD_0x20_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_PORT_RESET_IN msgrequest */ #define MC_CMD_PORT_RESET_IN_LEN 0 @@ -1934,6 +1978,7 @@ * extended version of the deprecated MC_CMD_PORT_RESET with added fields. */ #define MC_CMD_ENTITY_RESET 0x20 +/* MC_CMD_0x20_PRIVILEGE_CTG SRIOV_CTG_GENERAL */ /* MC_CMD_ENTITY_RESET_IN msgrequest */ #define MC_CMD_ENTITY_RESET_IN_LEN 4 @@ -2023,6 +2068,8 @@ */ #define MC_CMD_PUTS 0x23 +#define MC_CMD_0x23_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_PUTS_IN msgrequest */ #define MC_CMD_PUTS_IN_LENMIN 13 #define MC_CMD_PUTS_IN_LENMAX 252 @@ -2050,6 +2097,8 @@ */ #define MC_CMD_GET_PHY_CFG 0x24 +#define MC_CMD_0x24_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_PHY_CFG_IN msgrequest */ #define MC_CMD_GET_PHY_CFG_IN_LEN 0 @@ -2149,6 +2198,8 @@ */ #define MC_CMD_START_BIST 0x25 +#define MC_CMD_0x25_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_START_BIST_IN msgrequest */ #define MC_CMD_START_BIST_IN_LEN 4 /* Type of test. */ @@ -2185,6 +2236,8 @@ */ #define MC_CMD_POLL_BIST 0x26 +#define MC_CMD_0x26_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_POLL_BIST_IN msgrequest */ #define MC_CMD_POLL_BIST_IN_LEN 0 @@ -2344,6 +2397,8 @@ */ #define MC_CMD_GET_LOOPBACK_MODES 0x28 +#define MC_CMD_0x28_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_LOOPBACK_MODES_IN msgrequest */ #define MC_CMD_GET_LOOPBACK_MODES_IN_LEN 0 @@ -2463,6 +2518,8 @@ */ #define MC_CMD_GET_LINK 0x29 +#define MC_CMD_0x29_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_LINK_IN msgrequest */ #define MC_CMD_GET_LINK_IN_LEN 0 @@ -2519,6 +2576,8 @@ */ #define MC_CMD_SET_LINK 0x2a +#define MC_CMD_0x2a_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_SET_LINK_IN msgrequest */ #define MC_CMD_SET_LINK_IN_LEN 16 /* ??? */ @@ -2550,6 +2609,8 @@ */ #define MC_CMD_SET_ID_LED 0x2b +#define MC_CMD_0x2b_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_SET_ID_LED_IN msgrequest */ #define MC_CMD_SET_ID_LED_IN_LEN 4 /* Set LED state. */ @@ -2568,6 +2629,8 @@ */ #define MC_CMD_SET_MAC 0x2c +#define MC_CMD_0x2c_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_SET_MAC_IN msgrequest */ #define MC_CMD_SET_MAC_IN_LEN 24 /* The MTU is the MTU programmed directly into the XMAC/GMAC (inclusive of @@ -2609,6 +2672,8 @@ */ #define MC_CMD_PHY_STATS 0x2d +#define MC_CMD_0x2d_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_PHY_STATS_IN msgrequest */ #define MC_CMD_PHY_STATS_IN_LEN 8 /* ??? */ @@ -2687,6 +2752,8 @@ */ #define MC_CMD_MAC_STATS 0x2e +#define MC_CMD_0x2e_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_MAC_STATS_IN msgrequest */ #define MC_CMD_MAC_STATS_IN_LEN 16 /* ??? */ @@ -2926,6 +2993,8 @@ */ #define MC_CMD_WOL_FILTER_SET 0x32 +#define MC_CMD_0x32_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_WOL_FILTER_SET_IN msgrequest */ #define MC_CMD_WOL_FILTER_SET_IN_LEN 192 #define MC_CMD_WOL_FILTER_SET_IN_FILTER_MODE_OFST 0 @@ -3020,6 +3089,8 @@ */ #define MC_CMD_WOL_FILTER_REMOVE 0x33 +#define MC_CMD_0x33_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_WOL_FILTER_REMOVE_IN msgrequest */ #define MC_CMD_WOL_FILTER_REMOVE_IN_LEN 4 #define MC_CMD_WOL_FILTER_REMOVE_IN_FILTER_ID_OFST 0 @@ -3035,6 +3106,8 @@ */ #define MC_CMD_WOL_FILTER_RESET 0x34 +#define MC_CMD_0x34_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_WOL_FILTER_RESET_IN msgrequest */ #define MC_CMD_WOL_FILTER_RESET_IN_LEN 4 #define MC_CMD_WOL_FILTER_RESET_IN_MASK_OFST 0 @@ -3069,6 +3142,8 @@ */ #define MC_CMD_NVRAM_TYPES 0x36 +#define MC_CMD_0x36_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_TYPES_IN msgrequest */ #define MC_CMD_NVRAM_TYPES_IN_LEN 0 @@ -3125,6 +3200,8 @@ */ #define MC_CMD_NVRAM_INFO 0x37 +#define MC_CMD_0x37_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_INFO_IN msgrequest */ #define MC_CMD_NVRAM_INFO_IN_LEN 4 #define MC_CMD_NVRAM_INFO_IN_TYPE_OFST 0 @@ -3157,6 +3234,8 @@ */ #define MC_CMD_NVRAM_UPDATE_START 0x38 +#define MC_CMD_0x38_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_UPDATE_START_IN msgrequest */ #define MC_CMD_NVRAM_UPDATE_START_IN_LEN 4 #define MC_CMD_NVRAM_UPDATE_START_IN_TYPE_OFST 0 @@ -3175,6 +3254,8 @@ */ #define MC_CMD_NVRAM_READ 0x39 +#define MC_CMD_0x39_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_READ_IN msgrequest */ #define MC_CMD_NVRAM_READ_IN_LEN 12 #define MC_CMD_NVRAM_READ_IN_TYPE_OFST 0 @@ -3202,6 +3283,8 @@ */ #define MC_CMD_NVRAM_WRITE 0x3a +#define MC_CMD_0x3a_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_WRITE_IN msgrequest */ #define MC_CMD_NVRAM_WRITE_IN_LENMIN 13 #define MC_CMD_NVRAM_WRITE_IN_LENMAX 252 @@ -3228,6 +3311,8 @@ */ #define MC_CMD_NVRAM_ERASE 0x3b +#define MC_CMD_0x3b_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_ERASE_IN msgrequest */ #define MC_CMD_NVRAM_ERASE_IN_LEN 12 #define MC_CMD_NVRAM_ERASE_IN_TYPE_OFST 0 @@ -3248,6 +3333,8 @@ */ #define MC_CMD_NVRAM_UPDATE_FINISH 0x3c +#define MC_CMD_0x3c_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_UPDATE_FINISH_IN msgrequest */ #define MC_CMD_NVRAM_UPDATE_FINISH_IN_LEN 8 #define MC_CMD_NVRAM_UPDATE_FINISH_IN_TYPE_OFST 0 @@ -3279,6 +3366,8 @@ */ #define MC_CMD_REBOOT 0x3d +#define MC_CMD_0x3d_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_REBOOT_IN msgrequest */ #define MC_CMD_REBOOT_IN_LEN 4 #define MC_CMD_REBOOT_IN_FLAGS_OFST 0 @@ -3316,6 +3405,8 @@ */ #define MC_CMD_REBOOT_MODE 0x3f +#define MC_CMD_0x3f_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_REBOOT_MODE_IN msgrequest */ #define MC_CMD_REBOOT_MODE_IN_LEN 4 #define MC_CMD_REBOOT_MODE_IN_VALUE_OFST 0 @@ -3368,6 +3459,8 @@ */ #define MC_CMD_SENSOR_INFO 0x41 +#define MC_CMD_0x41_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SENSOR_INFO_IN msgrequest */ #define MC_CMD_SENSOR_INFO_IN_LEN 0 @@ -3542,6 +3635,8 @@ */ #define MC_CMD_READ_SENSORS 0x42 +#define MC_CMD_0x42_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_READ_SENSORS_IN msgrequest */ #define MC_CMD_READ_SENSORS_IN_LEN 8 /* DMA address of host buffer for sensor readings (must be 4Kbyte aligned). */ @@ -3602,6 +3697,8 @@ */ #define MC_CMD_GET_PHY_STATE 0x43 +#define MC_CMD_0x43_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_PHY_STATE_IN msgrequest */ #define MC_CMD_GET_PHY_STATE_IN_LEN 0 @@ -3636,6 +3733,8 @@ */ #define MC_CMD_WOL_FILTER_GET 0x45 +#define MC_CMD_0x45_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_WOL_FILTER_GET_IN msgrequest */ #define MC_CMD_WOL_FILTER_GET_IN_LEN 0 @@ -3651,6 +3750,8 @@ */ #define MC_CMD_ADD_LIGHTSOUT_OFFLOAD 0x46 +#define MC_CMD_0x46_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN msgrequest */ #define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_LENMIN 8 #define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_LENMAX 252 @@ -3692,6 +3793,8 @@ */ #define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD 0x47 +#define MC_CMD_0x47_PRIVILEGE_CTG SRIOV_CTG_LINK + /* MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN msgrequest */ #define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN_LEN 8 #define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN_PROTOCOL_OFST 0 @@ -3722,6 +3825,8 @@ */ #define MC_CMD_TESTASSERT 0x49 +#define MC_CMD_0x49_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_TESTASSERT_IN msgrequest */ #define MC_CMD_TESTASSERT_IN_LEN 0 @@ -3739,6 +3844,8 @@ */ #define MC_CMD_WORKAROUND 0x4a +#define MC_CMD_0x4a_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_WORKAROUND_IN msgrequest */ #define MC_CMD_WORKAROUND_IN_LEN 8 #define MC_CMD_WORKAROUND_IN_TYPE_OFST 0 @@ -3765,6 +3872,8 @@ */ #define MC_CMD_GET_PHY_MEDIA_INFO 0x4b +#define MC_CMD_0x4b_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_PHY_MEDIA_INFO_IN msgrequest */ #define MC_CMD_GET_PHY_MEDIA_INFO_IN_LEN 4 #define MC_CMD_GET_PHY_MEDIA_INFO_IN_PAGE_OFST 0 @@ -3788,6 +3897,8 @@ */ #define MC_CMD_NVRAM_TEST 0x4c +#define MC_CMD_0x4c_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_TEST_IN msgrequest */ #define MC_CMD_NVRAM_TEST_IN_LEN 4 #define MC_CMD_NVRAM_TEST_IN_TYPE_OFST 0 @@ -3849,6 +3960,8 @@ */ #define MC_CMD_SENSOR_SET_LIMS 0x4e +#define MC_CMD_0x4e_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SENSOR_SET_LIMS_IN msgrequest */ #define MC_CMD_SENSOR_SET_LIMS_IN_LEN 20 #define MC_CMD_SENSOR_SET_LIMS_IN_SENSOR_OFST 0 @@ -3890,6 +4003,8 @@ */ #define MC_CMD_NVRAM_PARTITIONS 0x51 +#define MC_CMD_0x51_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_PARTITIONS_IN msgrequest */ #define MC_CMD_NVRAM_PARTITIONS_IN_LEN 0 @@ -3913,6 +4028,8 @@ */ #define MC_CMD_NVRAM_METADATA 0x52 +#define MC_CMD_0x52_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_NVRAM_METADATA_IN msgrequest */ #define MC_CMD_NVRAM_METADATA_IN_LEN 4 /* Partition type ID code */ @@ -3958,6 +4075,8 @@ */ #define MC_CMD_GET_MAC_ADDRESSES 0x55 +#define MC_CMD_0x55_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_MAC_ADDRESSES_IN msgrequest */ #define MC_CMD_GET_MAC_ADDRESSES_IN_LEN 0 @@ -4087,11 +4206,66 @@ /***********************************/ +/* MC_CMD_GET_WORKAROUNDS + * Read the list of all implemented and all currently enabled workarounds. The + * enums here must correspond with those in MC_CMD_WORKAROUND. + */ +#define MC_CMD_GET_WORKAROUNDS 0x59 + +/* MC_CMD_GET_WORKAROUNDS_OUT msgresponse */ +#define MC_CMD_GET_WORKAROUNDS_OUT_LEN 8 +/* Each workaround is represented by a single bit according to the enums below. + */ +#define MC_CMD_GET_WORKAROUNDS_OUT_IMPLEMENTED_OFST 0 +#define MC_CMD_GET_WORKAROUNDS_OUT_ENABLED_OFST 4 +/* enum: Bug 17230 work around. */ +#define MC_CMD_GET_WORKAROUNDS_OUT_BUG17230 0x2 +/* enum: Bug 35388 work around (unsafe EVQ writes). */ +#define MC_CMD_GET_WORKAROUNDS_OUT_BUG35388 0x4 +/* enum: Bug35017 workaround (A64 tables must be identity map) */ +#define MC_CMD_GET_WORKAROUNDS_OUT_BUG35017 0x8 + + +/***********************************/ +/* MC_CMD_LINK_STATE_MODE + * Read/set link state mode of a VF + */ +#define MC_CMD_LINK_STATE_MODE 0x5c + +#define MC_CMD_0x5c_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LINK_STATE_MODE_IN msgrequest */ +#define MC_CMD_LINK_STATE_MODE_IN_LEN 8 +/* The target function to have its link state mode read or set, must be a VF + * e.g. VF 1,3 = 0x00030001 + */ +#define MC_CMD_LINK_STATE_MODE_IN_FUNCTION_OFST 0 +#define MC_CMD_LINK_STATE_MODE_IN_FUNCTION_PF_LBN 0 +#define MC_CMD_LINK_STATE_MODE_IN_FUNCTION_PF_WIDTH 16 +#define MC_CMD_LINK_STATE_MODE_IN_FUNCTION_VF_LBN 16 +#define MC_CMD_LINK_STATE_MODE_IN_FUNCTION_VF_WIDTH 16 +/* New link state mode to be set */ +#define MC_CMD_LINK_STATE_MODE_IN_NEW_MODE_OFST 4 +#define MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_AUTO 0x0 /* enum */ +#define MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_UP 0x1 /* enum */ +#define MC_CMD_LINK_STATE_MODE_IN_LINK_STATE_DOWN 0x2 /* enum */ +/* enum: Use this value to just read the existing setting without modifying it. + */ +#define MC_CMD_LINK_STATE_MODE_IN_DO_NOT_CHANGE 0xffffffff + +/* MC_CMD_LINK_STATE_MODE_OUT msgresponse */ +#define MC_CMD_LINK_STATE_MODE_OUT_LEN 4 +#define MC_CMD_LINK_STATE_MODE_OUT_OLD_MODE_OFST 0 + + +/***********************************/ /* MC_CMD_READ_REGS * Get a dump of the MCPU registers */ #define MC_CMD_READ_REGS 0x50 +#define MC_CMD_0x50_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_READ_REGS_IN msgrequest */ #define MC_CMD_READ_REGS_IN_LEN 0 @@ -4115,6 +4289,8 @@ */ #define MC_CMD_INIT_EVQ 0x80 +#define MC_CMD_0x80_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_INIT_EVQ_IN msgrequest */ #define MC_CMD_INIT_EVQ_IN_LENMIN 44 #define MC_CMD_INIT_EVQ_IN_LENMAX 548 @@ -4213,6 +4389,8 @@ */ #define MC_CMD_INIT_RXQ 0x81 +#define MC_CMD_0x81_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_INIT_RXQ_IN msgrequest */ #define MC_CMD_INIT_RXQ_IN_LENMIN 36 #define MC_CMD_INIT_RXQ_IN_LENMAX 252 @@ -4265,6 +4443,8 @@ */ #define MC_CMD_INIT_TXQ 0x82 +#define MC_CMD_0x82_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_INIT_TXQ_IN msgrequest */ #define MC_CMD_INIT_TXQ_IN_LENMIN 36 #define MC_CMD_INIT_TXQ_IN_LENMAX 252 @@ -4322,6 +4502,8 @@ */ #define MC_CMD_FINI_EVQ 0x83 +#define MC_CMD_0x83_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_FINI_EVQ_IN msgrequest */ #define MC_CMD_FINI_EVQ_IN_LEN 4 /* Instance of EVQ to destroy. Should be the same instance as that previously @@ -4339,6 +4521,8 @@ */ #define MC_CMD_FINI_RXQ 0x84 +#define MC_CMD_0x84_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_FINI_RXQ_IN msgrequest */ #define MC_CMD_FINI_RXQ_IN_LEN 4 /* Instance of RXQ to destroy */ @@ -4354,6 +4538,8 @@ */ #define MC_CMD_FINI_TXQ 0x85 +#define MC_CMD_0x85_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_FINI_TXQ_IN msgrequest */ #define MC_CMD_FINI_TXQ_IN_LEN 4 /* Instance of TXQ to destroy */ @@ -4369,6 +4555,8 @@ */ #define MC_CMD_DRIVER_EVENT 0x86 +#define MC_CMD_0x86_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_DRIVER_EVENT_IN msgrequest */ #define MC_CMD_DRIVER_EVENT_IN_LEN 12 /* Handle of target EVQ */ @@ -4392,6 +4580,8 @@ */ #define MC_CMD_PROXY_CMD 0x5b +#define MC_CMD_0x5b_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_PROXY_CMD_IN msgrequest */ #define MC_CMD_PROXY_CMD_IN_LEN 4 /* The handle of the target function. */ @@ -4414,6 +4604,8 @@ */ #define MC_CMD_ALLOC_BUFTBL_CHUNK 0x87 +#define MC_CMD_0x87_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_ALLOC_BUFTBL_CHUNK_IN msgrequest */ #define MC_CMD_ALLOC_BUFTBL_CHUNK_IN_LEN 8 /* Owner ID to use */ @@ -4437,6 +4629,8 @@ */ #define MC_CMD_PROGRAM_BUFTBL_ENTRIES 0x88 +#define MC_CMD_0x88_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_PROGRAM_BUFTBL_ENTRIES_IN msgrequest */ #define MC_CMD_PROGRAM_BUFTBL_ENTRIES_IN_LENMIN 20 #define MC_CMD_PROGRAM_BUFTBL_ENTRIES_IN_LENMAX 268 @@ -4463,6 +4657,8 @@ */ #define MC_CMD_FREE_BUFTBL_CHUNK 0x89 +#define MC_CMD_0x89_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_FREE_BUFTBL_CHUNK_IN msgrequest */ #define MC_CMD_FREE_BUFTBL_CHUNK_IN_LEN 4 #define MC_CMD_FREE_BUFTBL_CHUNK_IN_HANDLE_OFST 0 @@ -4477,6 +4673,8 @@ */ #define MC_CMD_FILTER_OP 0x8a +#define MC_CMD_0x8a_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_FILTER_OP_IN msgrequest */ #define MC_CMD_FILTER_OP_IN_LEN 108 /* identifies the type of operation requested */ @@ -4637,6 +4835,8 @@ */ #define MC_CMD_GET_PARSER_DISP_INFO 0xe4 +#define MC_CMD_0xe4_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_PARSER_DISP_INFO_IN msgrequest */ #define MC_CMD_GET_PARSER_DISP_INFO_IN_LEN 4 /* identifies the type of operation requested */ @@ -4669,6 +4869,8 @@ */ #define MC_CMD_PARSER_DISP_RW 0xe5 +#define MC_CMD_0xe5_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_PARSER_DISP_RW_IN msgrequest */ #define MC_CMD_PARSER_DISP_RW_IN_LEN 32 /* identifies the target of the operation */ @@ -4719,6 +4921,8 @@ */ #define MC_CMD_GET_PF_COUNT 0xb6 +#define MC_CMD_0xb6_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_PF_COUNT_IN msgrequest */ #define MC_CMD_GET_PF_COUNT_IN_LEN 0 @@ -4750,6 +4954,8 @@ */ #define MC_CMD_GET_PORT_ASSIGNMENT 0xb8 +#define MC_CMD_0xb8_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_PORT_ASSIGNMENT_IN msgrequest */ #define MC_CMD_GET_PORT_ASSIGNMENT_IN_LEN 0 @@ -4765,6 +4971,8 @@ */ #define MC_CMD_SET_PORT_ASSIGNMENT 0xb9 +#define MC_CMD_0xb9_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_PORT_ASSIGNMENT_IN msgrequest */ #define MC_CMD_SET_PORT_ASSIGNMENT_IN_LEN 4 /* Identifies the port assignment for this function. */ @@ -4780,6 +4988,8 @@ */ #define MC_CMD_ALLOC_VIS 0x8b +#define MC_CMD_0x8b_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_ALLOC_VIS_IN msgrequest */ #define MC_CMD_ALLOC_VIS_IN_LEN 8 /* The minimum number of VIs that is acceptable */ @@ -4804,6 +5014,8 @@ */ #define MC_CMD_FREE_VIS 0x8c +#define MC_CMD_0x8c_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_FREE_VIS_IN msgrequest */ #define MC_CMD_FREE_VIS_IN_LEN 0 @@ -4817,6 +5029,8 @@ */ #define MC_CMD_GET_SRIOV_CFG 0xba +#define MC_CMD_0xba_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_SRIOV_CFG_IN msgrequest */ #define MC_CMD_GET_SRIOV_CFG_IN_LEN 0 @@ -4841,6 +5055,8 @@ */ #define MC_CMD_SET_SRIOV_CFG 0xbb +#define MC_CMD_0xbb_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_SRIOV_CFG_IN msgrequest */ #define MC_CMD_SET_SRIOV_CFG_IN_LEN 20 /* Number of VFs currently enabled. */ @@ -4870,6 +5086,8 @@ */ #define MC_CMD_GET_VI_ALLOC_INFO 0x8d +#define MC_CMD_0x8d_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_VI_ALLOC_INFO_IN msgrequest */ #define MC_CMD_GET_VI_ALLOC_INFO_IN_LEN 0 @@ -4889,6 +5107,8 @@ */ #define MC_CMD_DUMP_VI_STATE 0x8e +#define MC_CMD_0x8e_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_DUMP_VI_STATE_IN msgrequest */ #define MC_CMD_DUMP_VI_STATE_IN_LEN 4 /* The VI number to query. */ @@ -4998,6 +5218,8 @@ */ #define MC_CMD_ALLOC_PIOBUF 0x8f +#define MC_CMD_0x8f_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_ALLOC_PIOBUF_IN msgrequest */ #define MC_CMD_ALLOC_PIOBUF_IN_LEN 0 @@ -5013,6 +5235,8 @@ */ #define MC_CMD_FREE_PIOBUF 0x90 +#define MC_CMD_0x90_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_FREE_PIOBUF_IN msgrequest */ #define MC_CMD_FREE_PIOBUF_IN_LEN 4 /* Handle for allocated push I/O buffer. */ @@ -5028,6 +5252,8 @@ */ #define MC_CMD_GET_VI_TLP_PROCESSING 0xb0 +#define MC_CMD_0xb0_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_VI_TLP_PROCESSING_IN msgrequest */ #define MC_CMD_GET_VI_TLP_PROCESSING_IN_LEN 4 /* VI number to get information for. */ @@ -5062,6 +5288,8 @@ */ #define MC_CMD_SET_VI_TLP_PROCESSING 0xb1 +#define MC_CMD_0xb1_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_SET_VI_TLP_PROCESSING_IN msgrequest */ #define MC_CMD_SET_VI_TLP_PROCESSING_IN_LEN 8 /* VI number to set information for. */ @@ -5096,6 +5324,8 @@ */ #define MC_CMD_GET_TLP_PROCESSING_GLOBALS 0xbc +#define MC_CMD_0xbc_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_TLP_PROCESSING_GLOBALS_IN msgrequest */ #define MC_CMD_GET_TLP_PROCESSING_GLOBALS_IN_LEN 4 #define MC_CMD_GET_TLP_PROCESSING_GLOBALS_IN_TLP_GLOBAL_CATEGORY_OFST 0 @@ -5157,6 +5387,8 @@ */ #define MC_CMD_SET_TLP_PROCESSING_GLOBALS 0xbd +#define MC_CMD_0xbd_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_TLP_PROCESSING_GLOBALS_IN msgrequest */ #define MC_CMD_SET_TLP_PROCESSING_GLOBALS_IN_LEN 8 #define MC_CMD_SET_TLP_PROCESSING_GLOBALS_IN_TLP_GLOBAL_CATEGORY_OFST 0 @@ -5203,6 +5435,8 @@ */ #define MC_CMD_SATELLITE_DOWNLOAD 0x91 +#define MC_CMD_0x91_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SATELLITE_DOWNLOAD_IN msgrequest: The reset requirements for the CPUs * are subtle, and so downloads must proceed in a number of phases. * @@ -5318,6 +5552,7 @@ */ #define MC_CMD_GET_CAPABILITIES 0xbe +#define MC_CMD_0xbe_PRIVILEGE_CTG SRIOV_CTG_GENERAL /* MC_CMD_GET_CAPABILITIES_IN msgrequest */ #define MC_CMD_GET_CAPABILITIES_IN_LEN 0 @@ -5433,6 +5668,8 @@ */ #define MC_CMD_TCM_BUCKET_ALLOC 0xb2 +#define MC_CMD_0xb2_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_TCM_BUCKET_ALLOC_IN msgrequest */ #define MC_CMD_TCM_BUCKET_ALLOC_IN_LEN 0 @@ -5448,6 +5685,8 @@ */ #define MC_CMD_TCM_BUCKET_FREE 0xb3 +#define MC_CMD_0xb3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_TCM_BUCKET_FREE_IN msgrequest */ #define MC_CMD_TCM_BUCKET_FREE_IN_LEN 4 /* the bucket id */ @@ -5463,6 +5702,8 @@ */ #define MC_CMD_TCM_BUCKET_INIT 0xb4 +#define MC_CMD_0xb4_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_TCM_BUCKET_INIT_IN msgrequest */ #define MC_CMD_TCM_BUCKET_INIT_IN_LEN 8 /* the bucket id */ @@ -5480,6 +5721,8 @@ */ #define MC_CMD_TCM_TXQ_INIT 0xb5 +#define MC_CMD_0xb5_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_TCM_TXQ_INIT_IN msgrequest */ #define MC_CMD_TCM_TXQ_INIT_IN_LEN 28 /* the txq id */ @@ -5511,6 +5754,8 @@ */ #define MC_CMD_LINK_PIOBUF 0x92 +#define MC_CMD_0x92_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_LINK_PIOBUF_IN msgrequest */ #define MC_CMD_LINK_PIOBUF_IN_LEN 8 /* Handle for allocated push I/O buffer. */ @@ -5528,6 +5773,8 @@ */ #define MC_CMD_UNLINK_PIOBUF 0x93 +#define MC_CMD_0x93_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_UNLINK_PIOBUF_IN msgrequest */ #define MC_CMD_UNLINK_PIOBUF_IN_LEN 4 /* Function Local Instance (VI) number. */ @@ -5543,6 +5790,8 @@ */ #define MC_CMD_VSWITCH_ALLOC 0x94 +#define MC_CMD_0x94_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VSWITCH_ALLOC_IN msgrequest */ #define MC_CMD_VSWITCH_ALLOC_IN_LEN 16 /* The port to connect to the v-switch's upstream port. */ @@ -5572,6 +5821,8 @@ */ #define MC_CMD_VSWITCH_FREE 0x95 +#define MC_CMD_0x95_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VSWITCH_FREE_IN msgrequest */ #define MC_CMD_VSWITCH_FREE_IN_LEN 4 /* The port to which the v-switch is connected. */ @@ -5587,6 +5838,8 @@ */ #define MC_CMD_VPORT_ALLOC 0x96 +#define MC_CMD_0x96_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VPORT_ALLOC_IN msgrequest */ #define MC_CMD_VPORT_ALLOC_IN_LEN 20 /* The port to which the v-switch is connected. */ @@ -5636,6 +5889,8 @@ */ #define MC_CMD_VPORT_FREE 0x97 +#define MC_CMD_0x97_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VPORT_FREE_IN msgrequest */ #define MC_CMD_VPORT_FREE_IN_LEN 4 /* The handle of the v-port */ @@ -5651,8 +5906,10 @@ */ #define MC_CMD_VADAPTOR_ALLOC 0x98 +#define MC_CMD_0x98_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VADAPTOR_ALLOC_IN msgrequest */ -#define MC_CMD_VADAPTOR_ALLOC_IN_LEN 16 +#define MC_CMD_VADAPTOR_ALLOC_IN_LEN 30 /* The port to connect to the v-adaptor's port. */ #define MC_CMD_VADAPTOR_ALLOC_IN_UPSTREAM_PORT_ID_OFST 0 /* Flags controlling v-adaptor creation */ @@ -5661,6 +5918,19 @@ #define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_AUTO_VADAPTOR_WIDTH 1 /* The number of VLAN tags to strip on receive */ #define MC_CMD_VADAPTOR_ALLOC_IN_NUM_VLANS_OFST 12 +/* The number of VLAN tags to transparently insert/remove. */ +#define MC_CMD_VADAPTOR_ALLOC_IN_NUM_VLAN_TAGS_OFST 16 +/* The actual VLAN tags to insert/remove */ +#define MC_CMD_VADAPTOR_ALLOC_IN_VLAN_TAGS_OFST 20 +#define MC_CMD_VADAPTOR_ALLOC_IN_VLAN_TAG_0_LBN 0 +#define MC_CMD_VADAPTOR_ALLOC_IN_VLAN_TAG_0_WIDTH 16 +#define MC_CMD_VADAPTOR_ALLOC_IN_VLAN_TAG_1_LBN 16 +#define MC_CMD_VADAPTOR_ALLOC_IN_VLAN_TAG_1_WIDTH 16 +/* The MAC address to assign to this v-adaptor */ +#define MC_CMD_VADAPTOR_ALLOC_IN_MACADDR_OFST 24 +#define MC_CMD_VADAPTOR_ALLOC_IN_MACADDR_LEN 6 +/* enum: Derive the MAC address from the upstream port */ +#define MC_CMD_VADAPTOR_ALLOC_IN_AUTO_MAC 0x0 /* MC_CMD_VADAPTOR_ALLOC_OUT msgresponse */ #define MC_CMD_VADAPTOR_ALLOC_OUT_LEN 0 @@ -5672,6 +5942,8 @@ */ #define MC_CMD_VADAPTOR_FREE 0x99 +#define MC_CMD_0x99_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VADAPTOR_FREE_IN msgrequest */ #define MC_CMD_VADAPTOR_FREE_IN_LEN 4 /* The port to which the v-adaptor is connected. */ @@ -5682,11 +5954,53 @@ /***********************************/ +/* MC_CMD_VADAPTOR_SET_MAC + * assign a new MAC address to a v-adaptor. + */ +#define MC_CMD_VADAPTOR_SET_MAC 0x5d + +#define MC_CMD_0x5d_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VADAPTOR_SET_MAC_IN msgrequest */ +#define MC_CMD_VADAPTOR_SET_MAC_IN_LEN 10 +/* The port to which the v-adaptor is connected. */ +#define MC_CMD_VADAPTOR_SET_MAC_IN_UPSTREAM_PORT_ID_OFST 0 +/* The new MAC address to assign to this v-adaptor */ +#define MC_CMD_VADAPTOR_SET_MAC_IN_MACADDR_OFST 4 +#define MC_CMD_VADAPTOR_SET_MAC_IN_MACADDR_LEN 6 + +/* MC_CMD_VADAPTOR_SET_MAC_OUT msgresponse */ +#define MC_CMD_VADAPTOR_SET_MAC_OUT_LEN 0 + + +/***********************************/ +/* MC_CMD_VADAPTOR_GET_MAC + * read the MAC address assigned to a v-adaptor. + */ +#define MC_CMD_VADAPTOR_GET_MAC 0x5e + +#define MC_CMD_0x5e_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VADAPTOR_GET_MAC_IN msgrequest */ +#define MC_CMD_VADAPTOR_GET_MAC_IN_LEN 4 +/* The port to which the v-adaptor is connected. */ +#define MC_CMD_VADAPTOR_GET_MAC_IN_UPSTREAM_PORT_ID_OFST 0 + +/* MC_CMD_VADAPTOR_GET_MAC_OUT msgresponse */ +#define MC_CMD_VADAPTOR_GET_MAC_OUT_LEN 6 +/* The MAC address assigned to this v-adaptor */ +#define MC_CMD_VADAPTOR_GET_MAC_OUT_MACADDR_OFST 0 +#define MC_CMD_VADAPTOR_GET_MAC_OUT_MACADDR_LEN 6 + + +/***********************************/ /* MC_CMD_EVB_PORT_ASSIGN * assign a port to a PCI function. */ #define MC_CMD_EVB_PORT_ASSIGN 0x9a +#define MC_CMD_0x9a_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_EVB_PORT_ASSIGN_IN msgrequest */ #define MC_CMD_EVB_PORT_ASSIGN_IN_LEN 8 /* The port to assign. */ @@ -5708,6 +6022,8 @@ */ #define MC_CMD_RDWR_A64_REGIONS 0x9b +#define MC_CMD_0x9b_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_RDWR_A64_REGIONS_IN msgrequest */ #define MC_CMD_RDWR_A64_REGIONS_IN_LEN 17 #define MC_CMD_RDWR_A64_REGIONS_IN_REGION0_OFST 0 @@ -5736,6 +6052,8 @@ */ #define MC_CMD_ONLOAD_STACK_ALLOC 0x9c +#define MC_CMD_0x9c_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_ONLOAD_STACK_ALLOC_IN msgrequest */ #define MC_CMD_ONLOAD_STACK_ALLOC_IN_LEN 4 /* The handle of the owning upstream port */ @@ -5753,6 +6071,8 @@ */ #define MC_CMD_ONLOAD_STACK_FREE 0x9d +#define MC_CMD_0x9d_PRIVILEGE_CTG SRIOV_CTG_ONLOAD + /* MC_CMD_ONLOAD_STACK_FREE_IN msgrequest */ #define MC_CMD_ONLOAD_STACK_FREE_IN_LEN 4 /* The handle of the Onload stack */ @@ -5768,6 +6088,8 @@ */ #define MC_CMD_RSS_CONTEXT_ALLOC 0x9e +#define MC_CMD_0x9e_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_ALLOC_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN 12 /* The handle of the owning upstream port */ @@ -5800,6 +6122,8 @@ */ #define MC_CMD_RSS_CONTEXT_FREE 0x9f +#define MC_CMD_0x9f_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_FREE_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_FREE_IN_LEN 4 /* The handle of the RSS context */ @@ -5815,6 +6139,8 @@ */ #define MC_CMD_RSS_CONTEXT_SET_KEY 0xa0 +#define MC_CMD_0xa0_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_SET_KEY_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_SET_KEY_IN_LEN 44 /* The handle of the RSS context */ @@ -5833,6 +6159,8 @@ */ #define MC_CMD_RSS_CONTEXT_GET_KEY 0xa1 +#define MC_CMD_0xa1_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_GET_KEY_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_GET_KEY_IN_LEN 4 /* The handle of the RSS context */ @@ -5851,6 +6179,8 @@ */ #define MC_CMD_RSS_CONTEXT_SET_TABLE 0xa2 +#define MC_CMD_0xa2_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_SET_TABLE_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_SET_TABLE_IN_LEN 132 /* The handle of the RSS context */ @@ -5869,6 +6199,8 @@ */ #define MC_CMD_RSS_CONTEXT_GET_TABLE 0xa3 +#define MC_CMD_0xa3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_GET_TABLE_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN 4 /* The handle of the RSS context */ @@ -5887,6 +6219,8 @@ */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS 0xe1 +#define MC_CMD_0xe1_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_SET_FLAGS_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN 8 /* The handle of the RSS context */ @@ -5912,6 +6246,8 @@ */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS 0xe2 +#define MC_CMD_0xe2_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_RSS_CONTEXT_GET_FLAGS_IN msgrequest */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS_IN_LEN 4 /* The handle of the RSS context */ @@ -5937,6 +6273,8 @@ */ #define MC_CMD_DOT1P_MAPPING_ALLOC 0xa4 +#define MC_CMD_0xa4_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DOT1P_MAPPING_ALLOC_IN msgrequest */ #define MC_CMD_DOT1P_MAPPING_ALLOC_IN_LEN 8 /* The handle of the owning upstream port */ @@ -5959,6 +6297,8 @@ */ #define MC_CMD_DOT1P_MAPPING_FREE 0xa5 +#define MC_CMD_0xa5_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DOT1P_MAPPING_FREE_IN msgrequest */ #define MC_CMD_DOT1P_MAPPING_FREE_IN_LEN 4 /* The handle of the .1p mapping */ @@ -5974,6 +6314,8 @@ */ #define MC_CMD_DOT1P_MAPPING_SET_TABLE 0xa6 +#define MC_CMD_0xa6_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DOT1P_MAPPING_SET_TABLE_IN msgrequest */ #define MC_CMD_DOT1P_MAPPING_SET_TABLE_IN_LEN 36 /* The handle of the .1p mapping */ @@ -5994,6 +6336,8 @@ */ #define MC_CMD_DOT1P_MAPPING_GET_TABLE 0xa7 +#define MC_CMD_0xa7_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DOT1P_MAPPING_GET_TABLE_IN msgrequest */ #define MC_CMD_DOT1P_MAPPING_GET_TABLE_IN_LEN 4 /* The handle of the .1p mapping */ @@ -6014,6 +6358,8 @@ */ #define MC_CMD_GET_VECTOR_CFG 0xbf +#define MC_CMD_0xbf_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_VECTOR_CFG_IN msgrequest */ #define MC_CMD_GET_VECTOR_CFG_IN_LEN 0 @@ -6033,6 +6379,8 @@ */ #define MC_CMD_SET_VECTOR_CFG 0xc0 +#define MC_CMD_0xc0_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_SET_VECTOR_CFG_IN msgrequest */ #define MC_CMD_SET_VECTOR_CFG_IN_LEN 12 /* Base absolute interrupt vector number, or MC_CMD_RESOURCE_INSTANCE_ANY to @@ -6423,6 +6771,8 @@ */ #define MC_CMD_VPORT_ADD_MAC_ADDRESS 0xa8 +#define MC_CMD_0xa8_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VPORT_ADD_MAC_ADDRESS_IN msgrequest */ #define MC_CMD_VPORT_ADD_MAC_ADDRESS_IN_LEN 10 /* The handle of the v-port */ @@ -6441,6 +6791,8 @@ */ #define MC_CMD_VPORT_DEL_MAC_ADDRESS 0xa9 +#define MC_CMD_0xa9_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VPORT_DEL_MAC_ADDRESS_IN msgrequest */ #define MC_CMD_VPORT_DEL_MAC_ADDRESS_IN_LEN 10 /* The handle of the v-port */ @@ -6459,6 +6811,8 @@ */ #define MC_CMD_VPORT_GET_MAC_ADDRESSES 0xaa +#define MC_CMD_0xaa_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_VPORT_GET_MAC_ADDRESSES_IN msgrequest */ #define MC_CMD_VPORT_GET_MAC_ADDRESSES_IN_LEN 4 /* The handle of the v-port */ @@ -6486,6 +6840,8 @@ */ #define MC_CMD_DUMP_BUFTBL_ENTRIES 0xab +#define MC_CMD_0xab_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DUMP_BUFTBL_ENTRIES_IN msgrequest */ #define MC_CMD_DUMP_BUFTBL_ENTRIES_IN_LEN 8 /* Index of the first buffer table entry. */ @@ -6510,6 +6866,8 @@ */ #define MC_CMD_SET_RXDP_CONFIG 0xc1 +#define MC_CMD_0xc1_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_RXDP_CONFIG_IN msgrequest */ #define MC_CMD_SET_RXDP_CONFIG_IN_LEN 4 #define MC_CMD_SET_RXDP_CONFIG_IN_DATA_OFST 0 @@ -6526,6 +6884,8 @@ */ #define MC_CMD_GET_RXDP_CONFIG 0xc2 +#define MC_CMD_0xc2_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_RXDP_CONFIG_IN msgrequest */ #define MC_CMD_GET_RXDP_CONFIG_IN_LEN 0 @@ -6890,6 +7250,8 @@ */ #define MC_CMD_GET_CLOCK 0xac +#define MC_CMD_0xac_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_CLOCK_IN msgrequest */ #define MC_CMD_GET_CLOCK_IN_LEN 0 @@ -6907,6 +7269,8 @@ */ #define MC_CMD_SET_CLOCK 0xad +#define MC_CMD_0xad_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_CLOCK_IN msgrequest */ #define MC_CMD_SET_CLOCK_IN_LEN 12 /* Requested system frequency in MHz; 0 leaves unchanged. */ @@ -6932,6 +7296,8 @@ */ #define MC_CMD_DPCPU_RPC 0xae +#define MC_CMD_0xae_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DPCPU_RPC_IN msgrequest */ #define MC_CMD_DPCPU_RPC_IN_LEN 36 #define MC_CMD_DPCPU_RPC_IN_CPU_OFST 0 @@ -7016,6 +7382,8 @@ */ #define MC_CMD_TRIGGER_INTERRUPT 0xe3 +#define MC_CMD_0xe3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_TRIGGER_INTERRUPT_IN msgrequest */ #define MC_CMD_TRIGGER_INTERRUPT_IN_LEN 4 /* Interrupt level relative to base for function. */ @@ -7031,6 +7399,8 @@ */ #define MC_CMD_CAP_BLK_READ 0xe7 +#define MC_CMD_0xe7_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_CAP_BLK_READ_IN msgrequest */ #define MC_CMD_CAP_BLK_READ_IN_LEN 12 #define MC_CMD_CAP_BLK_READ_IN_CAP_REG_OFST 0 @@ -7055,6 +7425,8 @@ */ #define MC_CMD_DUMP_DO 0xe8 +#define MC_CMD_0xe8_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DUMP_DO_IN msgrequest */ #define MC_CMD_DUMP_DO_IN_LEN 52 #define MC_CMD_DUMP_DO_IN_PADDING_OFST 0 @@ -7108,6 +7480,8 @@ */ #define MC_CMD_DUMP_CONFIGURE_UNSOLICITED 0xe9 +#define MC_CMD_0xe9_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_DUMP_CONFIGURE_UNSOLICITED_IN msgrequest */ #define MC_CMD_DUMP_CONFIGURE_UNSOLICITED_IN_LEN 52 #define MC_CMD_DUMP_CONFIGURE_UNSOLICITED_IN_ENABLE_OFST 0 @@ -7151,6 +7525,8 @@ */ #define MC_CMD_SET_PSU 0xea +#define MC_CMD_0xea_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_PSU_IN msgrequest */ #define MC_CMD_SET_PSU_IN_LEN 12 #define MC_CMD_SET_PSU_IN_PARAM_OFST 0 @@ -7171,6 +7547,8 @@ */ #define MC_CMD_GET_FUNCTION_INFO 0xec +#define MC_CMD_0xec_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_FUNCTION_INFO_IN msgrequest */ #define MC_CMD_GET_FUNCTION_INFO_IN_LEN 0 @@ -7188,6 +7566,8 @@ */ #define MC_CMD_ENABLE_OFFLINE_BIST 0xed +#define MC_CMD_0xed_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_ENABLE_OFFLINE_BIST_IN msgrequest */ #define MC_CMD_ENABLE_OFFLINE_BIST_IN_LEN 0 @@ -7203,6 +7583,8 @@ */ #define MC_CMD_UART_SEND_DATA 0xee +#define MC_CMD_0xee_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_UART_SEND_DATA_OUT msgrequest */ #define MC_CMD_UART_SEND_DATA_OUT_LENMIN 16 #define MC_CMD_UART_SEND_DATA_OUT_LENMAX 252 @@ -7231,6 +7613,8 @@ */ #define MC_CMD_UART_RECV_DATA 0xef +#define MC_CMD_0xef_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_UART_RECV_DATA_OUT msgrequest */ #define MC_CMD_UART_RECV_DATA_OUT_LEN 16 /* CRC32 over OFFSET, LENGTH, RESERVED */ @@ -7266,6 +7650,8 @@ */ #define MC_CMD_READ_FUSES 0xf0 +#define MC_CMD_0xf0_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_READ_FUSES_IN msgrequest */ #define MC_CMD_READ_FUSES_IN_LEN 8 /* Offset in OTP to read */ @@ -7292,6 +7678,8 @@ */ #define MC_CMD_KR_TUNE 0xf1 +#define MC_CMD_0xf1_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_KR_TUNE_IN msgrequest */ #define MC_CMD_KR_TUNE_IN_LENMIN 4 #define MC_CMD_KR_TUNE_IN_LENMAX 252 @@ -7550,6 +7938,8 @@ */ #define MC_CMD_PCIE_TUNE 0xf2 +#define MC_CMD_0xf2_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_PCIE_TUNE_IN msgrequest */ #define MC_CMD_PCIE_TUNE_IN_LENMIN 4 #define MC_CMD_PCIE_TUNE_IN_LENMAX 252 @@ -7711,6 +8101,8 @@ */ #define MC_CMD_LICENSING 0xf3 +#define MC_CMD_0xf3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_LICENSING_IN msgrequest */ #define MC_CMD_LICENSING_IN_LEN 4 /* identifies the type of operation requested */ @@ -7756,6 +8148,8 @@ */ #define MC_CMD_MC2MC_PROXY 0xf4 +#define MC_CMD_0xf4_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_MC2MC_PROXY_IN msgrequest */ #define MC_CMD_MC2MC_PROXY_IN_LEN 0 @@ -7771,6 +8165,8 @@ */ #define MC_CMD_GET_LICENSED_APP_STATE 0xf5 +#define MC_CMD_0xf5_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_GET_LICENSED_APP_STATE_IN msgrequest */ #define MC_CMD_GET_LICENSED_APP_STATE_IN_LEN 4 /* application ID to query (LICENSED_APP_ID_xxx) */ @@ -7792,6 +8188,8 @@ */ #define MC_CMD_LICENSED_APP_OP 0xf6 +#define MC_CMD_0xf6_PRIVILEGE_CTG SRIOV_CTG_GENERAL + /* MC_CMD_LICENSED_APP_OP_IN msgrequest */ #define MC_CMD_LICENSED_APP_OP_IN_LENMIN 8 #define MC_CMD_LICENSED_APP_OP_IN_LENMAX 252 @@ -7847,6 +8245,8 @@ */ #define MC_CMD_SET_PORT_SNIFF_CONFIG 0xf7 +#define MC_CMD_0xf7_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_SET_PORT_SNIFF_CONFIG_IN msgrequest */ #define MC_CMD_SET_PORT_SNIFF_CONFIG_IN_LEN 16 /* configuration flags */ @@ -7881,6 +8281,8 @@ */ #define MC_CMD_GET_PORT_SNIFF_CONFIG 0xf8 +#define MC_CMD_0xf8_PRIVILEGE_CTG SRIOV_CTG_ADMIN + /* MC_CMD_GET_PORT_SNIFF_CONFIG_IN msgrequest */ #define MC_CMD_GET_PORT_SNIFF_CONFIG_IN_LEN 0 diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c index fb19b70eac01..9bf04cbce20a 100644 --- a/drivers/net/ethernet/sfc/mcdi_port.c +++ b/drivers/net/ethernet/sfc/mcdi_port.c @@ -865,6 +865,7 @@ int efx_mcdi_set_mac(struct efx_nic *efx) BUILD_BUG_ON(MC_CMD_SET_MAC_OUT_LEN != 0); + /* This has no effect on EF10 */ ether_addr_copy(MCDI_PTR(cmdbytes, SET_MAC_IN_ADDR), efx->net_dev->dev_addr); diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 325dd94bca46..a468a22e7a88 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -25,6 +25,7 @@ #include <linux/highmem.h> #include <linux/workqueue.h> #include <linux/mutex.h> +#include <linux/rwsem.h> #include <linux/vmalloc.h> #include <linux/i2c.h> #include <linux/mtd/mtd.h> @@ -793,7 +794,6 @@ union efx_multicast_hash { efx_oword_t oword[EFX_MCAST_HASH_ENTRIES / sizeof(efx_oword_t) / 8]; }; -struct efx_vf; struct vfdi_status; /** @@ -897,7 +897,8 @@ struct vfdi_status; * @loopback_mode: Loopback status * @loopback_modes: Supported loopback mode bitmask * @loopback_selftest: Offline self-test private state - * @filter_lock: Filter table lock + * @filter_sem: Filter table rw_semaphore, for freeing the table + * @filter_lock: Filter table lock, for mere content changes * @filter_state: Architecture-dependent filter table state * @rps_flow_id: Flow IDs of filters allocated for accelerated RFS, * indexed by filter ID @@ -909,7 +910,6 @@ struct vfdi_status; * completed (either success or failure). Not used when MCDI is used to * flush receive queues. * @flush_wq: wait queue used by efx_nic_flush_queues() to wait for flush completions. - * @vf: Array of &struct efx_vf objects. * @vf_count: Number of VFs intended to be enabled. * @vf_init_count: Number of VFs that have been fully initialised. * @vi_scale: log2 number of vnics per VF. @@ -1040,6 +1040,7 @@ struct efx_nic { void *loopback_selftest; + struct rw_semaphore filter_sem; spinlock_t filter_lock; void *filter_state; #ifdef CONFIG_RFS_ACCEL @@ -1053,7 +1054,6 @@ struct efx_nic { wait_queue_head_t flush_wq; #ifdef CONFIG_SFC_SRIOV - struct efx_vf *vf; unsigned vf_count; unsigned vf_init_count; unsigned vi_scale; @@ -1092,6 +1092,7 @@ struct efx_mtd_partition { /** * struct efx_nic_type - Efx device type definition + * @mem_bar: Get the memory BAR * @mem_map_size: Get memory BAR mapped size * @probe: Probe the controller * @remove: Free resources allocated by probe() @@ -1204,6 +1205,7 @@ struct efx_mtd_partition { * @ptp_set_ts_config: Set hardware timestamp configuration. The flags * and tx_type will already have been validated but this operation * must validate and update rx_filter. + * @set_mac_address: Set the MAC address of the device * @revision: Hardware architecture revision * @txd_ptr_tbl_base: TX descriptor ring base address * @rxd_ptr_tbl_base: RX descriptor ring base address @@ -1226,6 +1228,8 @@ struct efx_mtd_partition { * @hwtstamp_filters: Mask of hardware timestamp filter types supported */ struct efx_nic_type { + bool is_vf; + unsigned int mem_bar; unsigned int (*mem_map_size)(struct efx_nic *efx); int (*probe)(struct efx_nic *efx); void (*remove)(struct efx_nic *efx); @@ -1277,7 +1281,8 @@ struct efx_nic_type { void (*tx_init)(struct efx_tx_queue *tx_queue); void (*tx_remove)(struct efx_tx_queue *tx_queue); void (*tx_write)(struct efx_tx_queue *tx_queue); - void (*rx_push_rss_config)(struct efx_nic *efx); + int (*rx_push_rss_config)(struct efx_nic *efx, bool user, + const u32 *rx_indir_table); int (*rx_probe)(struct efx_rx_queue *rx_queue); void (*rx_init)(struct efx_rx_queue *rx_queue); void (*rx_remove)(struct efx_rx_queue *rx_queue); @@ -1330,11 +1335,26 @@ struct efx_nic_type { int (*ptp_set_ts_sync_events)(struct efx_nic *efx, bool en, bool temp); int (*ptp_set_ts_config)(struct efx_nic *efx, struct hwtstamp_config *init); + int (*sriov_configure)(struct efx_nic *efx, int num_vfs); int (*sriov_init)(struct efx_nic *efx); void (*sriov_fini)(struct efx_nic *efx); - void (*sriov_mac_address_changed)(struct efx_nic *efx); bool (*sriov_wanted)(struct efx_nic *efx); void (*sriov_reset)(struct efx_nic *efx); + void (*sriov_flr)(struct efx_nic *efx, unsigned vf_i); + int (*sriov_set_vf_mac)(struct efx_nic *efx, int vf_i, u8 *mac); + int (*sriov_set_vf_vlan)(struct efx_nic *efx, int vf_i, u16 vlan, + u8 qos); + int (*sriov_set_vf_spoofchk)(struct efx_nic *efx, int vf_i, + bool spoofchk); + int (*sriov_get_vf_config)(struct efx_nic *efx, int vf_i, + struct ifla_vf_info *ivi); + int (*sriov_set_vf_link_state)(struct efx_nic *efx, int vf_i, + int link_state); + int (*vswitching_probe)(struct efx_nic *efx); + int (*vswitching_restore)(struct efx_nic *efx); + void (*vswitching_remove)(struct efx_nic *efx); + int (*get_mac_address)(struct efx_nic *efx, unsigned char *perm_addr); + int (*set_mac_address)(struct efx_nic *efx); int revision; unsigned int txd_ptr_tbl_base; diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 93d10cbbd1cf..db8562ec586d 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -381,6 +381,7 @@ enum { * @efx: Pointer back to main interface structure * @wol_filter_id: Wake-on-LAN packet filter id * @stats: Hardware statistics + * @vf: Array of &struct siena_vf objects * @vf_buftbl_base: The zeroth buffer table index used to back VF queues. * @vfdi_status: Common VFDI status page to be dmad to VF address space. * @local_addr_list: List of local addresses. Protected by %local_lock. @@ -394,6 +395,7 @@ struct siena_nic_data { int wol_filter_id; u64 stats[SIENA_STAT_COUNT]; #ifdef CONFIG_SFC_SRIOV + struct siena_vf *vf; struct efx_channel *vfdi_channel; unsigned vf_buftbl_base; struct efx_buffer vfdi_status; @@ -483,12 +485,21 @@ enum { * @must_restore_piobufs: Flag: PIO buffers have yet to be restored after MC * reboot * @rx_rss_context: Firmware handle for our RSS context + * @rx_rss_context_exclusive: Whether our RSS context is exclusive or shared * @stats: Hardware statistics * @workaround_35388: Flag: firmware supports workaround for bug 35388 * @must_check_datapath_caps: Flag: @datapath_caps needs to be revalidated * after MC reboot * @datapath_caps: Capabilities of datapath firmware (FLAGS1 field of * %MC_CMD_GET_CAPABILITIES response) + * @rx_dpcpu_fw_id: Firmware ID of the RxDPCPU + * @tx_dpcpu_fw_id: Firmware ID of the TxDPCPU + * @vport_id: The function's vport ID, only relevant for PFs + * @must_probe_vswitching: Flag: vswitching has yet to be setup after MC reboot + * @pf_index: The number for this PF, or the parent PF if this is a VF +#ifdef CONFIG_SFC_SRIOV + * @vf: Pointer to VF data structure +#endif */ struct efx_ef10_nic_data { struct efx_buffer mcdi_buf; @@ -503,126 +514,26 @@ struct efx_ef10_nic_data { unsigned int piobuf_handle[EF10_TX_PIOBUF_COUNT]; bool must_restore_piobufs; u32 rx_rss_context; + bool rx_rss_context_exclusive; u64 stats[EF10_STAT_COUNT]; bool workaround_35388; bool must_check_datapath_caps; u32 datapath_caps; -}; - -/* - * On the SFC9000 family each port is associated with 1 PCI physical - * function (PF) handled by sfc and a configurable number of virtual - * functions (VFs) that may be handled by some other driver, often in - * a VM guest. The queue pointer registers are mapped in both PF and - * VF BARs such that an 8K region provides access to a single RX, TX - * and event queue (collectively a Virtual Interface, VI or VNIC). - * - * The PF has access to all 1024 VIs while VFs are mapped to VIs - * according to VI_BASE and VI_SCALE: VF i has access to VIs numbered - * in range [VI_BASE + i << VI_SCALE, VI_BASE + i + 1 << VI_SCALE). - * The number of VIs and the VI_SCALE value are configurable but must - * be established at boot time by firmware. - */ - -/* Maximum VI_SCALE parameter supported by Siena */ -#define EFX_VI_SCALE_MAX 6 -/* Base VI to use for SR-IOV. Must be aligned to (1 << EFX_VI_SCALE_MAX), - * so this is the smallest allowed value. */ -#define EFX_VI_BASE 128U -/* Maximum number of VFs allowed */ -#define EFX_VF_COUNT_MAX 127 -/* Limit EVQs on VFs to be only 8k to reduce buffer table reservation */ -#define EFX_MAX_VF_EVQ_SIZE 8192UL -/* The number of buffer table entries reserved for each VI on a VF */ -#define EFX_VF_BUFTBL_PER_VI \ - ((EFX_MAX_VF_EVQ_SIZE + 2 * EFX_MAX_DMAQ_SIZE) * \ - sizeof(efx_qword_t) / EFX_BUF_SIZE) - + unsigned int rx_dpcpu_fw_id; + unsigned int tx_dpcpu_fw_id; + unsigned int vport_id; + bool must_probe_vswitching; + unsigned int pf_index; #ifdef CONFIG_SFC_SRIOV - -/* SIENA */ -static inline bool efx_siena_sriov_wanted(struct efx_nic *efx) -{ - return efx->vf_count != 0; -} - -static inline bool efx_siena_sriov_enabled(struct efx_nic *efx) -{ - return efx->vf_init_count != 0; -} - -static inline unsigned int efx_vf_size(struct efx_nic *efx) -{ - return 1 << efx->vi_scale; -} + unsigned int vf_index; + struct ef10_vf *vf; +#endif + u8 vport_mac[ETH_ALEN]; +}; int efx_init_sriov(void); -void efx_siena_sriov_probe(struct efx_nic *efx); -int efx_siena_sriov_init(struct efx_nic *efx); -void efx_siena_sriov_mac_address_changed(struct efx_nic *efx); -void efx_siena_sriov_tx_flush_done(struct efx_nic *efx, efx_qword_t *event); -void efx_siena_sriov_rx_flush_done(struct efx_nic *efx, efx_qword_t *event); -void efx_siena_sriov_event(struct efx_channel *channel, efx_qword_t *event); -void efx_siena_sriov_desc_fetch_err(struct efx_nic *efx, unsigned dmaq); -void efx_siena_sriov_flr(struct efx_nic *efx, unsigned flr); -void efx_siena_sriov_reset(struct efx_nic *efx); -void efx_siena_sriov_fini(struct efx_nic *efx); void efx_fini_sriov(void); -/* EF10 */ -static inline bool efx_ef10_sriov_wanted(struct efx_nic *efx) { return false; } -static inline int efx_ef10_sriov_init(struct efx_nic *efx) { return -EOPNOTSUPP; } -static inline void efx_ef10_sriov_mac_address_changed(struct efx_nic *efx) {} -static inline void efx_ef10_sriov_reset(struct efx_nic *efx) {} -static inline void efx_ef10_sriov_fini(struct efx_nic *efx) {} - -#else - -/* SIENA */ -static inline bool efx_siena_sriov_wanted(struct efx_nic *efx) { return false; } -static inline bool efx_siena_sriov_enabled(struct efx_nic *efx) { return false; } -static inline unsigned int efx_vf_size(struct efx_nic *efx) { return 0; } -static inline int efx_init_sriov(void) { return 0; } -static inline void efx_siena_sriov_probe(struct efx_nic *efx) {} -static inline int efx_siena_sriov_init(struct efx_nic *efx) { return -EOPNOTSUPP; } -static inline void efx_siena_sriov_mac_address_changed(struct efx_nic *efx) {} -static inline void efx_siena_sriov_tx_flush_done(struct efx_nic *efx, - efx_qword_t *event) {} -static inline void efx_siena_sriov_rx_flush_done(struct efx_nic *efx, - efx_qword_t *event) {} -static inline void efx_siena_sriov_event(struct efx_channel *channel, - efx_qword_t *event) {} -static inline void efx_siena_sriov_desc_fetch_err(struct efx_nic *efx, - unsigned dmaq) {} -static inline void efx_siena_sriov_flr(struct efx_nic *efx, unsigned flr) {} -static inline void efx_siena_sriov_reset(struct efx_nic *efx) {} -static inline void efx_siena_sriov_fini(struct efx_nic *efx) {} -static inline void efx_fini_sriov(void) {} - -/* EF10 */ -static inline bool efx_ef10_sriov_wanted(struct efx_nic *efx) { return false; } -static inline int efx_ef10_sriov_init(struct efx_nic *efx) { return -EOPNOTSUPP; } -static inline void efx_ef10_sriov_mac_address_changed(struct efx_nic *efx) {} -static inline void efx_ef10_sriov_reset(struct efx_nic *efx) {} -static inline void efx_ef10_sriov_fini(struct efx_nic *efx) {} - -#endif - -/* FALCON */ -static inline bool efx_falcon_sriov_wanted(struct efx_nic *efx) { return false; } -static inline int efx_falcon_sriov_init(struct efx_nic *efx) { return -EOPNOTSUPP; } -static inline void efx_falcon_sriov_mac_address_changed(struct efx_nic *efx) {} -static inline void efx_falcon_sriov_reset(struct efx_nic *efx) {} -static inline void efx_falcon_sriov_fini(struct efx_nic *efx) {} - -int efx_siena_sriov_set_vf_mac(struct net_device *dev, int vf, u8 *mac); -int efx_siena_sriov_set_vf_vlan(struct net_device *dev, int vf, - u16 vlan, u8 qos); -int efx_siena_sriov_get_vf_config(struct net_device *dev, int vf, - struct ifla_vf_info *ivf); -int efx_siena_sriov_set_vf_spoofchk(struct net_device *net_dev, int vf, - bool spoofchk); - struct ethtool_ts_info; int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel); void efx_ptp_defer_probe_with_channel(struct efx_nic *efx); @@ -654,6 +565,7 @@ extern const struct efx_nic_type falcon_a1_nic_type; extern const struct efx_nic_type falcon_b0_nic_type; extern const struct efx_nic_type siena_a0_nic_type; extern const struct efx_nic_type efx_hunt_a0_nic_type; +extern const struct efx_nic_type efx_hunt_a0_vf_nic_type; /************************************************************************** * diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index a2e9aee05cdd..ad62615a93dc 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -306,7 +306,7 @@ struct efx_ptp_data { struct work_struct pps_work; struct workqueue_struct *pps_workwq; bool nic_ts_enabled; - MCDI_DECLARE_BUF(txbuf, MC_CMD_PTP_IN_TRANSMIT_LENMAX); + _MCDI_DECLARE_BUF(txbuf, MC_CMD_PTP_IN_TRANSMIT_LENMAX); unsigned int good_syncs; unsigned int fast_syncs; @@ -389,11 +389,8 @@ size_t efx_ptp_update_stats(struct efx_nic *efx, u64 *stats) MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0); rc = efx_mcdi_rpc(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), outbuf, sizeof(outbuf), NULL); - if (rc) { - netif_err(efx, hw, efx->net_dev, - "MC_CMD_PTP_OP_STATUS failed (%d)\n", rc); + if (rc) memset(outbuf, 0, sizeof(outbuf)); - } efx_nic_update_stats(efx_ptp_stat_desc, PTP_STAT_COUNT, efx_ptp_stat_mask, stats, _MCDI_PTR(outbuf, 0), false); @@ -490,14 +487,20 @@ static int efx_ptp_get_attributes(struct efx_nic *efx) */ MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_GET_ATTRIBUTES); MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0); - rc = efx_mcdi_rpc(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), - outbuf, sizeof(outbuf), &out_len); - if (rc == 0) + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &out_len); + if (rc == 0) { fmt = MCDI_DWORD(outbuf, PTP_OUT_GET_ATTRIBUTES_TIME_FORMAT); - else if (rc == -EINVAL) + } else if (rc == -EINVAL) { fmt = MC_CMD_PTP_OUT_GET_ATTRIBUTES_SECONDS_NANOSECONDS; - else + } else if (rc == -EPERM) { + netif_info(efx, probe, efx->net_dev, "no PTP support\n"); + return rc; + } else { + efx_mcdi_display_error(efx, MC_CMD_PTP, sizeof(inbuf), + outbuf, sizeof(outbuf), rc); return rc; + } if (fmt == MC_CMD_PTP_OUT_GET_ATTRIBUTES_SECONDS_27FRACTION) { ptp->ns_to_nic_time = efx_ptp_ns_to_s27; @@ -541,8 +544,8 @@ static int efx_ptp_get_timestamp_corrections(struct efx_nic *efx) MC_CMD_PTP_OP_GET_TIMESTAMP_CORRECTIONS); MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0); - rc = efx_mcdi_rpc(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), - outbuf, sizeof(outbuf), NULL); + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), NULL); if (rc == 0) { efx->ptp_data->ts_corrections.tx = MCDI_DWORD(outbuf, PTP_OUT_GET_TIMESTAMP_CORRECTIONS_TRANSMIT); @@ -558,6 +561,8 @@ static int efx_ptp_get_timestamp_corrections(struct efx_nic *efx) efx->ptp_data->ts_corrections.pps_out = 0; efx->ptp_data->ts_corrections.pps_in = 0; } else { + efx_mcdi_display_error(efx, MC_CMD_PTP, sizeof(inbuf), outbuf, + sizeof(outbuf), rc); return rc; } @@ -568,7 +573,7 @@ static int efx_ptp_get_timestamp_corrections(struct efx_nic *efx) static int efx_ptp_enable(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_PTP_IN_ENABLE_LEN); - MCDI_DECLARE_BUF_OUT_OR_ERR(outbuf, 0); + MCDI_DECLARE_BUF_ERR(outbuf); int rc; MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_ENABLE); @@ -596,7 +601,7 @@ static int efx_ptp_enable(struct efx_nic *efx) static int efx_ptp_disable(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_PTP_IN_DISABLE_LEN); - MCDI_DECLARE_BUF_OUT_OR_ERR(outbuf, 0); + MCDI_DECLARE_BUF_ERR(outbuf); int rc; MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_DISABLE); @@ -604,7 +609,12 @@ static int efx_ptp_disable(struct efx_nic *efx) rc = efx_mcdi_rpc_quiet(efx, MC_CMD_PTP, inbuf, sizeof(inbuf), outbuf, sizeof(outbuf), NULL); rc = (rc == -EALREADY) ? 0 : rc; - if (rc) + /* If we get ENOSYS, the NIC doesn't support PTP, and thus this function + * should only have been called during probe. + */ + if (rc == -ENOSYS || rc == -EPERM) + netif_info(efx, probe, efx->net_dev, "no PTP support\n"); + else if (rc) efx_mcdi_display_error(efx, MC_CMD_PTP, MC_CMD_PTP_IN_DISABLE_LEN, outbuf, sizeof(outbuf), rc); diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c index f12c811938d2..b323b9167526 100644 --- a/drivers/net/ethernet/sfc/siena.c +++ b/drivers/net/ethernet/sfc/siena.c @@ -25,6 +25,7 @@ #include "mcdi.h" #include "mcdi_pcol.h" #include "selftest.h" +#include "siena_sriov.h" /* Hardware control for SFC9000 family including SFL9021 (aka Siena). */ @@ -306,7 +307,9 @@ static int siena_probe_nic(struct efx_nic *efx) if (rc) goto fail5; +#ifdef CONFIG_SFC_SRIOV efx_siena_sriov_probe(efx); +#endif efx_ptp_defer_probe_with_channel(efx); return 0; @@ -321,7 +324,8 @@ fail1: return rc; } -static void siena_rx_push_rss_config(struct efx_nic *efx) +static int siena_rx_push_rss_config(struct efx_nic *efx, bool user, + const u32 *rx_indir_table) { efx_oword_t temp; @@ -343,7 +347,11 @@ static void siena_rx_push_rss_config(struct efx_nic *efx) FRF_CZ_RX_RSS_IPV6_TKEY_HI_WIDTH / 8); efx_writeo(efx, &temp, FR_CZ_RX_RSS_IPV6_REG3); + memcpy(efx->rx_indir_table, rx_indir_table, + sizeof(efx->rx_indir_table)); efx_farch_rx_push_indir_table(efx); + + return 0; } /* This call performs hardware-specific global initialisation, such as @@ -386,7 +394,7 @@ static int siena_init_nic(struct efx_nic *efx) EFX_RX_USR_BUF_SIZE >> 5); efx_writeo(efx, &temp, FR_AZ_RX_CFG); - siena_rx_push_rss_config(efx); + siena_rx_push_rss_config(efx, false, efx->rx_indir_table); /* Enable event logging */ rc = efx_mcdi_log_ctrl(efx, true, false, 0); @@ -909,6 +917,8 @@ fail: */ const struct efx_nic_type siena_a0_nic_type = { + .is_vf = false, + .mem_bar = EFX_MEM_BAR, .mem_map_size = siena_mem_map_size, .probe = siena_probe_nic, .remove = siena_remove_nic, @@ -996,11 +1006,22 @@ const struct efx_nic_type siena_a0_nic_type = { #endif .ptp_write_host_time = siena_ptp_write_host_time, .ptp_set_ts_config = siena_ptp_set_ts_config, +#ifdef CONFIG_SFC_SRIOV + .sriov_configure = efx_siena_sriov_configure, .sriov_init = efx_siena_sriov_init, .sriov_fini = efx_siena_sriov_fini, - .sriov_mac_address_changed = efx_siena_sriov_mac_address_changed, .sriov_wanted = efx_siena_sriov_wanted, .sriov_reset = efx_siena_sriov_reset, + .sriov_flr = efx_siena_sriov_flr, + .sriov_set_vf_mac = efx_siena_sriov_set_vf_mac, + .sriov_set_vf_vlan = efx_siena_sriov_set_vf_vlan, + .sriov_set_vf_spoofchk = efx_siena_sriov_set_vf_spoofchk, + .sriov_get_vf_config = efx_siena_sriov_get_vf_config, + .vswitching_probe = efx_port_dummy_op_int, + .vswitching_restore = efx_port_dummy_op_int, + .vswitching_remove = efx_port_dummy_op_void, + .set_mac_address = efx_siena_sriov_mac_address_changed, +#endif .revision = EFX_REV_SIENA_A0, .txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL, diff --git a/drivers/net/ethernet/sfc/siena_sriov.c b/drivers/net/ethernet/sfc/siena_sriov.c index fe83430796fd..da7b94f34604 100644 --- a/drivers/net/ethernet/sfc/siena_sriov.c +++ b/drivers/net/ethernet/sfc/siena_sriov.c @@ -16,6 +16,7 @@ #include "filter.h" #include "mcdi_pcol.h" #include "farch_regs.h" +#include "siena_sriov.h" #include "vfdi.h" /* Number of longs required to track all the VIs in a VF */ @@ -38,7 +39,7 @@ enum efx_vf_tx_filter_mode { }; /** - * struct efx_vf - Back-end resource and protocol state for a PCI VF + * struct siena_vf - Back-end resource and protocol state for a PCI VF * @efx: The Efx NIC owning this VF * @pci_rid: The PCI requester ID for this VF * @pci_name: The PCI name (formatted address) of this VF @@ -83,7 +84,7 @@ enum efx_vf_tx_filter_mode { * @rxq_retry_count: Number of receive queues in @rxq_retry_mask. * @reset_work: Work item to schedule a VF reset. */ -struct efx_vf { +struct siena_vf { struct efx_nic *efx; unsigned int pci_rid; char pci_name[13]; /* dddd:bb:dd.f */ @@ -189,7 +190,7 @@ MODULE_PARM_DESC(max_vfs, */ static struct workqueue_struct *vfdi_workqueue; -static unsigned abs_index(struct efx_vf *vf, unsigned index) +static unsigned abs_index(struct siena_vf *vf, unsigned index) { return EFX_VI_BASE + vf->index * efx_vf_size(vf->efx) + index; } @@ -207,8 +208,8 @@ static int efx_siena_sriov_cmd(struct efx_nic *efx, bool enable, MCDI_SET_DWORD(inbuf, SRIOV_IN_VI_BASE, EFX_VI_BASE); MCDI_SET_DWORD(inbuf, SRIOV_IN_VF_COUNT, efx->vf_count); - rc = efx_mcdi_rpc(efx, MC_CMD_SRIOV, inbuf, MC_CMD_SRIOV_IN_LEN, - outbuf, MC_CMD_SRIOV_OUT_LEN, &outlen); + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_SRIOV, inbuf, MC_CMD_SRIOV_IN_LEN, + outbuf, MC_CMD_SRIOV_OUT_LEN, &outlen); if (rc) return rc; if (outlen < MC_CMD_SRIOV_OUT_LEN) @@ -299,7 +300,7 @@ out: /* The TX filter is entirely controlled by this driver, and is modified * underneath the feet of the VF */ -static void efx_siena_sriov_reset_tx_filter(struct efx_vf *vf) +static void efx_siena_sriov_reset_tx_filter(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct efx_filter_spec filter; @@ -343,7 +344,7 @@ static void efx_siena_sriov_reset_tx_filter(struct efx_vf *vf) } /* The RX filter is managed here on behalf of the VF driver */ -static void efx_siena_sriov_reset_rx_filter(struct efx_vf *vf) +static void efx_siena_sriov_reset_rx_filter(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct efx_filter_spec filter; @@ -382,7 +383,7 @@ static void efx_siena_sriov_reset_rx_filter(struct efx_vf *vf) } } -static void __efx_siena_sriov_update_vf_addr(struct efx_vf *vf) +static void __efx_siena_sriov_update_vf_addr(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct siena_nic_data *nic_data = efx->nic_data; @@ -397,7 +398,7 @@ static void __efx_siena_sriov_update_vf_addr(struct efx_vf *vf) * local_page_list, either by acquiring local_lock or by running from * efx_siena_sriov_peer_work() */ -static void __efx_siena_sriov_push_vf_status(struct efx_vf *vf) +static void __efx_siena_sriov_push_vf_status(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct siena_nic_data *nic_data = efx->nic_data; @@ -509,8 +510,9 @@ static bool bad_buf_count(unsigned buf_count, unsigned max_entry_count) * Optionally set VF index and VI index within the VF. */ static bool map_vi_index(struct efx_nic *efx, unsigned abs_index, - struct efx_vf **vf_out, unsigned *rel_index_out) + struct siena_vf **vf_out, unsigned *rel_index_out) { + struct siena_nic_data *nic_data = efx->nic_data; unsigned vf_i; if (abs_index < EFX_VI_BASE) @@ -520,13 +522,13 @@ static bool map_vi_index(struct efx_nic *efx, unsigned abs_index, return true; if (vf_out) - *vf_out = efx->vf + vf_i; + *vf_out = nic_data->vf + vf_i; if (rel_index_out) *rel_index_out = abs_index % efx_vf_size(efx); return false; } -static int efx_vfdi_init_evq(struct efx_vf *vf) +static int efx_vfdi_init_evq(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct vfdi_req *req = vf->buf.addr; @@ -567,7 +569,7 @@ static int efx_vfdi_init_evq(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -static int efx_vfdi_init_rxq(struct efx_vf *vf) +static int efx_vfdi_init_rxq(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct vfdi_req *req = vf->buf.addr; @@ -608,7 +610,7 @@ static int efx_vfdi_init_rxq(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -static int efx_vfdi_init_txq(struct efx_vf *vf) +static int efx_vfdi_init_txq(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct vfdi_req *req = vf->buf.addr; @@ -655,7 +657,7 @@ static int efx_vfdi_init_txq(struct efx_vf *vf) } /* Returns true when efx_vfdi_fini_all_queues should wake */ -static bool efx_vfdi_flush_wake(struct efx_vf *vf) +static bool efx_vfdi_flush_wake(struct siena_vf *vf) { /* Ensure that all updates are visible to efx_vfdi_fini_all_queues() */ smp_mb(); @@ -664,7 +666,7 @@ static bool efx_vfdi_flush_wake(struct efx_vf *vf) atomic_read(&vf->rxq_retry_count); } -static void efx_vfdi_flush_clear(struct efx_vf *vf) +static void efx_vfdi_flush_clear(struct siena_vf *vf) { memset(vf->txq_mask, 0, sizeof(vf->txq_mask)); vf->txq_count = 0; @@ -674,7 +676,7 @@ static void efx_vfdi_flush_clear(struct efx_vf *vf) atomic_set(&vf->rxq_retry_count, 0); } -static int efx_vfdi_fini_all_queues(struct efx_vf *vf) +static int efx_vfdi_fini_all_queues(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; efx_oword_t reg; @@ -757,7 +759,7 @@ static int efx_vfdi_fini_all_queues(struct efx_vf *vf) return timeout ? 0 : VFDI_RC_ETIMEDOUT; } -static int efx_vfdi_insert_filter(struct efx_vf *vf) +static int efx_vfdi_insert_filter(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct siena_nic_data *nic_data = efx->nic_data; @@ -789,7 +791,7 @@ static int efx_vfdi_insert_filter(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -static int efx_vfdi_remove_all_filters(struct efx_vf *vf) +static int efx_vfdi_remove_all_filters(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct siena_nic_data *nic_data = efx->nic_data; @@ -801,7 +803,7 @@ static int efx_vfdi_remove_all_filters(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -static int efx_vfdi_set_status_page(struct efx_vf *vf) +static int efx_vfdi_set_status_page(struct siena_vf *vf) { struct efx_nic *efx = vf->efx; struct siena_nic_data *nic_data = efx->nic_data; @@ -846,7 +848,7 @@ static int efx_vfdi_set_status_page(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -static int efx_vfdi_clear_status_page(struct efx_vf *vf) +static int efx_vfdi_clear_status_page(struct siena_vf *vf) { mutex_lock(&vf->status_lock); vf->status_addr = 0; @@ -855,7 +857,7 @@ static int efx_vfdi_clear_status_page(struct efx_vf *vf) return VFDI_RC_SUCCESS; } -typedef int (*efx_vfdi_op_t)(struct efx_vf *vf); +typedef int (*efx_vfdi_op_t)(struct siena_vf *vf); static const efx_vfdi_op_t vfdi_ops[VFDI_OP_LIMIT] = { [VFDI_OP_INIT_EVQ] = efx_vfdi_init_evq, @@ -870,7 +872,7 @@ static const efx_vfdi_op_t vfdi_ops[VFDI_OP_LIMIT] = { static void efx_siena_sriov_vfdi(struct work_struct *work) { - struct efx_vf *vf = container_of(work, struct efx_vf, req); + struct siena_vf *vf = container_of(work, struct siena_vf, req); struct efx_nic *efx = vf->efx; struct vfdi_req *req = vf->buf.addr; struct efx_memcpy_req copy[2]; @@ -936,7 +938,8 @@ static void efx_siena_sriov_vfdi(struct work_struct *work) * event ring in guest memory with VFDI reset events, then (re-initialise) the * event queue to raise an interrupt. The guest driver will then recover. */ -static void efx_siena_sriov_reset_vf(struct efx_vf *vf, + +static void efx_siena_sriov_reset_vf(struct siena_vf *vf, struct efx_buffer *buffer) { struct efx_nic *efx = vf->efx; @@ -1006,7 +1009,7 @@ static void efx_siena_sriov_reset_vf(struct efx_vf *vf, static void efx_siena_sriov_reset_vf_work(struct work_struct *work) { - struct efx_vf *vf = container_of(work, struct efx_vf, req); + struct siena_vf *vf = container_of(work, struct siena_vf, req); struct efx_nic *efx = vf->efx; struct efx_buffer buf; @@ -1055,8 +1058,10 @@ void efx_siena_sriov_probe(struct efx_nic *efx) if (!max_vfs) return; - if (efx_siena_sriov_cmd(efx, false, &efx->vi_scale, &count)) + if (efx_siena_sriov_cmd(efx, false, &efx->vi_scale, &count)) { + netif_info(efx, probe, efx->net_dev, "no SR-IOV VFs probed\n"); return; + } if (count > 0 && count > max_vfs) count = max_vfs; @@ -1077,7 +1082,7 @@ static void efx_siena_sriov_peer_work(struct work_struct *data) peer_work); struct efx_nic *efx = nic_data->efx; struct vfdi_status *vfdi_status = nic_data->vfdi_status.addr; - struct efx_vf *vf; + struct siena_vf *vf; struct efx_local_addr *local_addr; struct vfdi_endpoint *peer; struct efx_endpoint_page *epp; @@ -1099,7 +1104,7 @@ static void efx_siena_sriov_peer_work(struct work_struct *data) peer_space = ARRAY_SIZE(vfdi_status->peers) - 1; peer_count = 1; for (pos = 0; pos < efx->vf_count; ++pos) { - vf = efx->vf + pos; + vf = nic_data->vf + pos; mutex_lock(&vf->status_lock); if (vf->rx_filtering && !is_zero_ether_addr(vf->addr.mac_addr)) { @@ -1155,7 +1160,7 @@ static void efx_siena_sriov_peer_work(struct work_struct *data) /* Finally, push the pages */ for (pos = 0; pos < efx->vf_count; ++pos) { - vf = efx->vf + pos; + vf = nic_data->vf + pos; mutex_lock(&vf->status_lock); if (vf->status_addr) @@ -1190,14 +1195,16 @@ static void efx_siena_sriov_free_local(struct efx_nic *efx) static int efx_siena_sriov_vf_alloc(struct efx_nic *efx) { unsigned index; - struct efx_vf *vf; + struct siena_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; - efx->vf = kzalloc(sizeof(struct efx_vf) * efx->vf_count, GFP_KERNEL); - if (!efx->vf) + nic_data->vf = kcalloc(efx->vf_count, sizeof(*nic_data->vf), + GFP_KERNEL); + if (!nic_data->vf) return -ENOMEM; for (index = 0; index < efx->vf_count; ++index) { - vf = efx->vf + index; + vf = nic_data->vf + index; vf->efx = efx; vf->index = index; @@ -1216,11 +1223,12 @@ static int efx_siena_sriov_vf_alloc(struct efx_nic *efx) static void efx_siena_sriov_vfs_fini(struct efx_nic *efx) { - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; unsigned int pos; for (pos = 0; pos < efx->vf_count; ++pos) { - vf = efx->vf + pos; + vf = nic_data->vf + pos; efx_nic_free_buffer(efx, &vf->buf); kfree(vf->peer_page_addrs); @@ -1237,7 +1245,7 @@ static int efx_siena_sriov_vfs_init(struct efx_nic *efx) struct siena_nic_data *nic_data = efx->nic_data; unsigned index, devfn, sriov, buftbl_base; u16 offset, stride; - struct efx_vf *vf; + struct siena_vf *vf; int rc; sriov = pci_find_ext_capability(pci_dev, PCI_EXT_CAP_ID_SRIOV); @@ -1250,7 +1258,7 @@ static int efx_siena_sriov_vfs_init(struct efx_nic *efx) buftbl_base = nic_data->vf_buftbl_base; devfn = pci_dev->devfn + offset; for (index = 0; index < efx->vf_count; ++index) { - vf = efx->vf + index; + vf = nic_data->vf + index; /* Reserve buffer entries */ vf->buftbl_base = buftbl_base; @@ -1350,7 +1358,7 @@ fail_pci: fail_vfs: cancel_work_sync(&nic_data->peer_work); efx_siena_sriov_free_local(efx); - kfree(efx->vf); + kfree(nic_data->vf); fail_alloc: efx_nic_free_buffer(efx, &nic_data->vfdi_status); fail_status: @@ -1361,7 +1369,7 @@ fail_cmd: void efx_siena_sriov_fini(struct efx_nic *efx) { - struct efx_vf *vf; + struct siena_vf *vf; unsigned int pos; struct siena_nic_data *nic_data = efx->nic_data; @@ -1377,7 +1385,7 @@ void efx_siena_sriov_fini(struct efx_nic *efx) /* Flush all reconfiguration work */ for (pos = 0; pos < efx->vf_count; ++pos) { - vf = efx->vf + pos; + vf = nic_data->vf + pos; cancel_work_sync(&vf->req); cancel_work_sync(&vf->reset_work); } @@ -1388,7 +1396,7 @@ void efx_siena_sriov_fini(struct efx_nic *efx) /* Tear down back-end state */ efx_siena_sriov_vfs_fini(efx); efx_siena_sriov_free_local(efx); - kfree(efx->vf); + kfree(nic_data->vf); efx_nic_free_buffer(efx, &nic_data->vfdi_status); efx_siena_sriov_cmd(efx, false, NULL, NULL); } @@ -1396,7 +1404,7 @@ void efx_siena_sriov_fini(struct efx_nic *efx) void efx_siena_sriov_event(struct efx_channel *channel, efx_qword_t *event) { struct efx_nic *efx = channel->efx; - struct efx_vf *vf; + struct siena_vf *vf; unsigned qid, seq, type, data; qid = EFX_QWORD_FIELD(*event, FSF_CZ_USER_QID); @@ -1452,11 +1460,12 @@ error: void efx_siena_sriov_flr(struct efx_nic *efx, unsigned vf_i) { - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; if (vf_i > efx->vf_init_count) return; - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; netif_info(efx, hw, efx->net_dev, "FLR on VF %s\n", vf->pci_name); @@ -1467,21 +1476,23 @@ void efx_siena_sriov_flr(struct efx_nic *efx, unsigned vf_i) vf->evq0_count = 0; } -void efx_siena_sriov_mac_address_changed(struct efx_nic *efx) +int efx_siena_sriov_mac_address_changed(struct efx_nic *efx) { struct siena_nic_data *nic_data = efx->nic_data; struct vfdi_status *vfdi_status = nic_data->vfdi_status.addr; if (!efx->vf_init_count) - return; + return 0; ether_addr_copy(vfdi_status->peers[0].mac_addr, efx->net_dev->dev_addr); queue_work(vfdi_workqueue, &nic_data->peer_work); + + return 0; } void efx_siena_sriov_tx_flush_done(struct efx_nic *efx, efx_qword_t *event) { - struct efx_vf *vf; + struct siena_vf *vf; unsigned queue, qid; queue = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_SUBDATA); @@ -1500,7 +1511,7 @@ void efx_siena_sriov_tx_flush_done(struct efx_nic *efx, efx_qword_t *event) void efx_siena_sriov_rx_flush_done(struct efx_nic *efx, efx_qword_t *event) { - struct efx_vf *vf; + struct siena_vf *vf; unsigned ev_failed, queue, qid; queue = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_RX_DESCQ_ID); @@ -1525,7 +1536,7 @@ void efx_siena_sriov_rx_flush_done(struct efx_nic *efx, efx_qword_t *event) /* Called from napi. Schedule the reset work item */ void efx_siena_sriov_desc_fetch_err(struct efx_nic *efx, unsigned dmaq) { - struct efx_vf *vf; + struct siena_vf *vf; unsigned int rel; if (map_vi_index(efx, dmaq, &vf, &rel)) @@ -1541,9 +1552,10 @@ void efx_siena_sriov_desc_fetch_err(struct efx_nic *efx, unsigned dmaq) /* Reset all VFs */ void efx_siena_sriov_reset(struct efx_nic *efx) { + struct siena_nic_data *nic_data = efx->nic_data; unsigned int vf_i; struct efx_buffer buf; - struct efx_vf *vf; + struct siena_vf *vf; ASSERT_RTNL(); @@ -1557,7 +1569,7 @@ void efx_siena_sriov_reset(struct efx_nic *efx) return; for (vf_i = 0; vf_i < efx->vf_init_count; ++vf_i) { - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; efx_siena_sriov_reset_vf(vf, &buf); } @@ -1573,7 +1585,6 @@ int efx_init_sriov(void) vfdi_workqueue = create_singlethread_workqueue("sfc_vfdi"); if (!vfdi_workqueue) return -ENOMEM; - return 0; } @@ -1582,14 +1593,14 @@ void efx_fini_sriov(void) destroy_workqueue(vfdi_workqueue); } -int efx_siena_sriov_set_vf_mac(struct net_device *net_dev, int vf_i, u8 *mac) +int efx_siena_sriov_set_vf_mac(struct efx_nic *efx, int vf_i, u8 *mac) { - struct efx_nic *efx = netdev_priv(net_dev); - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; if (vf_i >= efx->vf_init_count) return -EINVAL; - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; mutex_lock(&vf->status_lock); ether_addr_copy(vf->addr.mac_addr, mac); @@ -1599,16 +1610,16 @@ int efx_siena_sriov_set_vf_mac(struct net_device *net_dev, int vf_i, u8 *mac) return 0; } -int efx_siena_sriov_set_vf_vlan(struct net_device *net_dev, int vf_i, +int efx_siena_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, u16 vlan, u8 qos) { - struct efx_nic *efx = netdev_priv(net_dev); - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; u16 tci; if (vf_i >= efx->vf_init_count) return -EINVAL; - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; mutex_lock(&vf->status_lock); tci = (vlan & VLAN_VID_MASK) | ((qos & 0x7) << VLAN_PRIO_SHIFT); @@ -1619,16 +1630,16 @@ int efx_siena_sriov_set_vf_vlan(struct net_device *net_dev, int vf_i, return 0; } -int efx_siena_sriov_set_vf_spoofchk(struct net_device *net_dev, int vf_i, +int efx_siena_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf_i, bool spoofchk) { - struct efx_nic *efx = netdev_priv(net_dev); - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; int rc; if (vf_i >= efx->vf_init_count) return -EINVAL; - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; mutex_lock(&vf->txq_lock); if (vf->txq_count == 0) { @@ -1643,16 +1654,16 @@ int efx_siena_sriov_set_vf_spoofchk(struct net_device *net_dev, int vf_i, return rc; } -int efx_siena_sriov_get_vf_config(struct net_device *net_dev, int vf_i, +int efx_siena_sriov_get_vf_config(struct efx_nic *efx, int vf_i, struct ifla_vf_info *ivi) { - struct efx_nic *efx = netdev_priv(net_dev); - struct efx_vf *vf; + struct siena_nic_data *nic_data = efx->nic_data; + struct siena_vf *vf; u16 tci; if (vf_i >= efx->vf_init_count) return -EINVAL; - vf = efx->vf + vf_i; + vf = nic_data->vf + vf_i; ivi->vf = vf_i; ether_addr_copy(ivi->mac, vf->addr.mac_addr); @@ -1666,3 +1677,12 @@ int efx_siena_sriov_get_vf_config(struct net_device *net_dev, int vf_i, return 0; } +bool efx_siena_sriov_wanted(struct efx_nic *efx) +{ + return efx->vf_count != 0; +} + +int efx_siena_sriov_configure(struct efx_nic *efx, int num_vfs) +{ + return 0; +} diff --git a/drivers/net/ethernet/sfc/siena_sriov.h b/drivers/net/ethernet/sfc/siena_sriov.h new file mode 100644 index 000000000000..d88d4dab170a --- /dev/null +++ b/drivers/net/ethernet/sfc/siena_sriov.h @@ -0,0 +1,79 @@ +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2015 Solarflare Communications Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ + +#ifndef SIENA_SRIOV_H +#define SIENA_SRIOV_H + +#include "net_driver.h" + +/* On the SFC9000 family each port is associated with 1 PCI physical + * function (PF) handled by sfc and a configurable number of virtual + * functions (VFs) that may be handled by some other driver, often in + * a VM guest. The queue pointer registers are mapped in both PF and + * VF BARs such that an 8K region provides access to a single RX, TX + * and event queue (collectively a Virtual Interface, VI or VNIC). + * + * The PF has access to all 1024 VIs while VFs are mapped to VIs + * according to VI_BASE and VI_SCALE: VF i has access to VIs numbered + * in range [VI_BASE + i << VI_SCALE, VI_BASE + i + 1 << VI_SCALE). + * The number of VIs and the VI_SCALE value are configurable but must + * be established at boot time by firmware. + */ + +/* Maximum VI_SCALE parameter supported by Siena */ +#define EFX_VI_SCALE_MAX 6 +/* Base VI to use for SR-IOV. Must be aligned to (1 << EFX_VI_SCALE_MAX), + * so this is the smallest allowed value. + */ +#define EFX_VI_BASE 128U +/* Maximum number of VFs allowed */ +#define EFX_VF_COUNT_MAX 127 +/* Limit EVQs on VFs to be only 8k to reduce buffer table reservation */ +#define EFX_MAX_VF_EVQ_SIZE 8192UL +/* The number of buffer table entries reserved for each VI on a VF */ +#define EFX_VF_BUFTBL_PER_VI \ + ((EFX_MAX_VF_EVQ_SIZE + 2 * EFX_MAX_DMAQ_SIZE) * \ + sizeof(efx_qword_t) / EFX_BUF_SIZE) + +int efx_siena_sriov_configure(struct efx_nic *efx, int num_vfs); +int efx_siena_sriov_init(struct efx_nic *efx); +void efx_siena_sriov_fini(struct efx_nic *efx); +int efx_siena_sriov_mac_address_changed(struct efx_nic *efx); +bool efx_siena_sriov_wanted(struct efx_nic *efx); +void efx_siena_sriov_reset(struct efx_nic *efx); +void efx_siena_sriov_flr(struct efx_nic *efx, unsigned flr); + +int efx_siena_sriov_set_vf_mac(struct efx_nic *efx, int vf, u8 *mac); +int efx_siena_sriov_set_vf_vlan(struct efx_nic *efx, int vf, + u16 vlan, u8 qos); +int efx_siena_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf, + bool spoofchk); +int efx_siena_sriov_get_vf_config(struct efx_nic *efx, int vf, + struct ifla_vf_info *ivf); + +#ifdef CONFIG_SFC_SRIOV + +static inline bool efx_siena_sriov_enabled(struct efx_nic *efx) +{ + return efx->vf_init_count != 0; +} +#else /* !CONFIG_SFC_SRIOV */ +static inline bool efx_siena_sriov_enabled(struct efx_nic *efx) +{ + return false; +} +#endif /* CONFIG_SFC_SRIOV */ + +void efx_siena_sriov_probe(struct efx_nic *efx); +void efx_siena_sriov_tx_flush_done(struct efx_nic *efx, efx_qword_t *event); +void efx_siena_sriov_rx_flush_done(struct efx_nic *efx, efx_qword_t *event); +void efx_siena_sriov_event(struct efx_channel *channel, efx_qword_t *event); +void efx_siena_sriov_desc_fetch_err(struct efx_nic *efx, unsigned dmaq); + +#endif /* SIENA_SRIOV_H */ diff --git a/drivers/net/ethernet/sfc/sriov.c b/drivers/net/ethernet/sfc/sriov.c new file mode 100644 index 000000000000..6c5edbdbfa27 --- /dev/null +++ b/drivers/net/ethernet/sfc/sriov.c @@ -0,0 +1,72 @@ +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2014-2015 Solarflare Communications Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ +#include <linux/module.h> +#include "net_driver.h" +#include "nic.h" +#include "sriov.h" + +int efx_sriov_set_vf_mac(struct net_device *net_dev, int vf_i, u8 *mac) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->sriov_set_vf_mac) + return efx->type->sriov_set_vf_mac(efx, vf_i, mac); + else + return -EOPNOTSUPP; +} + +int efx_sriov_set_vf_vlan(struct net_device *net_dev, int vf_i, u16 vlan, + u8 qos) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->sriov_set_vf_vlan) { + if ((vlan & ~VLAN_VID_MASK) || + (qos & ~(VLAN_PRIO_MASK >> VLAN_PRIO_SHIFT))) + return -EINVAL; + + return efx->type->sriov_set_vf_vlan(efx, vf_i, vlan, qos); + } else { + return -EOPNOTSUPP; + } +} + +int efx_sriov_set_vf_spoofchk(struct net_device *net_dev, int vf_i, + bool spoofchk) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->sriov_set_vf_spoofchk) + return efx->type->sriov_set_vf_spoofchk(efx, vf_i, spoofchk); + else + return -EOPNOTSUPP; +} + +int efx_sriov_get_vf_config(struct net_device *net_dev, int vf_i, + struct ifla_vf_info *ivi) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->sriov_get_vf_config) + return efx->type->sriov_get_vf_config(efx, vf_i, ivi); + else + return -EOPNOTSUPP; +} + +int efx_sriov_set_vf_link_state(struct net_device *net_dev, int vf_i, + int link_state) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->sriov_set_vf_link_state) + return efx->type->sriov_set_vf_link_state(efx, vf_i, + link_state); + else + return -EOPNOTSUPP; +} diff --git a/drivers/net/ethernet/sfc/sriov.h b/drivers/net/ethernet/sfc/sriov.h new file mode 100644 index 000000000000..3be15a54c562 --- /dev/null +++ b/drivers/net/ethernet/sfc/sriov.h @@ -0,0 +1,29 @@ +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2014-2015 Solarflare Communications Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ + +#ifndef EFX_SRIOV_H +#define EFX_SRIOV_H + +#include "net_driver.h" + +#ifdef CONFIG_SFC_SRIOV + +int efx_sriov_set_vf_mac(struct net_device *net_dev, int vf_i, u8 *mac); +int efx_sriov_set_vf_vlan(struct net_device *net_dev, int vf_i, u16 vlan, + u8 qos); +int efx_sriov_set_vf_spoofchk(struct net_device *net_dev, int vf_i, + bool spoofchk); +int efx_sriov_get_vf_config(struct net_device *net_dev, int vf_i, + struct ifla_vf_info *ivi); +int efx_sriov_set_vf_link_state(struct net_device *net_dev, int vf_i, + int link_state); + +#endif /* CONFIG_SFC_SRIOV */ + +#endif /* EFX_SRIOV_H */ diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 7d3af190be55..cec147d1d34f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -16,6 +16,7 @@ if STMMAC_ETH config STMMAC_PLATFORM tristate "STMMAC Platform bus support" depends on STMMAC_ETH + select MFD_SYSCON default y ---help--- This selects the platform specific bus support for the stmmac driver. @@ -26,6 +27,95 @@ config STMMAC_PLATFORM If unsure, say N. +if STMMAC_PLATFORM + +config DWMAC_GENERIC + tristate "Generic driver for DWMAC" + default STMMAC_PLATFORM + ---help--- + Generic DWMAC driver for platforms that don't require any + platform specific code to function or is using platform + data for setup. + +config DWMAC_IPQ806X + tristate "QCA IPQ806x DWMAC support" + default ARCH_QCOM + depends on OF + select MFD_SYSCON + help + Support for QCA IPQ806X DWMAC Ethernet. + + This selects the IPQ806x SoC glue layer support for the stmmac + device driver. This driver does not use any of the hardware + acceleration features available on this SoC. Network devices + will behave like standard non-accelerated ethernet interfaces. + +config DWMAC_LPC18XX + tristate "NXP LPC18xx/43xx DWMAC support" + default ARCH_LPC18XX + depends on OF + select MFD_SYSCON + ---help--- + Support for NXP LPC18xx/43xx DWMAC Ethernet. + +config DWMAC_MESON + tristate "Amlogic Meson dwmac support" + default ARCH_MESON + depends on OF + help + Support for Ethernet controller on Amlogic Meson SoCs. + + This selects the Amlogic Meson SoC glue layer support for + the stmmac device driver. This driver is used for Meson6 and + Meson8 SoCs. + +config DWMAC_ROCKCHIP + tristate "Rockchip dwmac support" + default ARCH_ROCKCHIP + depends on OF + select MFD_SYSCON + help + Support for Ethernet controller on Rockchip RK3288 SoC. + + This selects the Rockchip RK3288 SoC glue layer support for + the stmmac device driver. + +config DWMAC_SOCFPGA + tristate "SOCFPGA dwmac support" + default ARCH_SOCFPGA + depends on OF + select MFD_SYSCON + help + Support for ethernet controller on Altera SOCFPGA + + This selects the Altera SOCFPGA SoC glue layer support + for the stmmac device driver. This driver is used for + arria5 and cyclone5 FPGA SoCs. + +config DWMAC_STI + tristate "STi GMAC support" + default ARCH_STI + depends on OF + select MFD_SYSCON + ---help--- + Support for ethernet controller on STi SOCs. + + This selects STi SoC glue layer support for the stmmac + device driver. This driver is used on for the STi series + SOCs GMAC ethernet controller. + +config DWMAC_SUNXI + tristate "Allwinner GMAC support" + default ARCH_SUNXI + depends on OF + ---help--- + Support for Allwinner A20/A31 GMAC ethernet controllers. + + This selects Allwinner SoC glue layer support for the + stmmac device driver. This driver is used for A20/A31 + GMAC ethernet controller. +endif + config STMMAC_PCI tristate "STMMAC PCI bus support" depends on STMMAC_ETH && PCI diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 73c2715a27f3..b3901616f4f6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -4,9 +4,17 @@ stmmac-objs:= stmmac_main.o stmmac_ethtool.o stmmac_mdio.o ring_mode.o \ dwmac100_core.o dwmac100_dma.o enh_desc.o norm_desc.o \ mmc_core.o stmmac_hwtstamp.o stmmac_ptp.o $(stmmac-y) -obj-$(CONFIG_STMMAC_PLATFORM) += stmmac-platform.o -stmmac-platform-objs:= stmmac_platform.o dwmac-meson.o dwmac-sunxi.o \ - dwmac-sti.o dwmac-socfpga.o dwmac-rk.o +# Ordering matters. Generic driver must be last. +obj-$(CONFIG_STMMAC_PLATFORM) += stmmac-platform.o +obj-$(CONFIG_DWMAC_IPQ806X) += dwmac-ipq806x.o +obj-$(CONFIG_DWMAC_LPC18XX) += dwmac-lpc18xx.o +obj-$(CONFIG_DWMAC_MESON) += dwmac-meson.o +obj-$(CONFIG_DWMAC_ROCKCHIP) += dwmac-rk.o +obj-$(CONFIG_DWMAC_SOCFPGA) += dwmac-socfpga.o +obj-$(CONFIG_DWMAC_STI) += dwmac-sti.o +obj-$(CONFIG_DWMAC_SUNXI) += dwmac-sunxi.o +obj-$(CONFIG_DWMAC_GENERIC) += dwmac-generic.o +stmmac-platform-objs:= stmmac_platform.o obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o stmmac-pci-objs:= stmmac_pci.o diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c new file mode 100644 index 000000000000..e817a1a44379 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-generic.c @@ -0,0 +1,41 @@ +/* + * Generic DWMAC platform driver + * + * Copyright (C) 2007-2011 STMicroelectronics Ltd + * Copyright (C) 2015 Joachim Eastwood <manabian@gmail.com> + * + * This file is licensed under the terms of the GNU General Public + * License version 2. This program is licensed "as is" without any + * warranty of any kind, whether express or implied. + */ + +#include <linux/module.h> +#include <linux/of.h> +#include <linux/platform_device.h> + +#include "stmmac.h" +#include "stmmac_platform.h" + +static const struct of_device_id dwmac_generic_match[] = { + { .compatible = "st,spear600-gmac"}, + { .compatible = "snps,dwmac-3.610"}, + { .compatible = "snps,dwmac-3.70a"}, + { .compatible = "snps,dwmac-3.710"}, + { .compatible = "snps,dwmac"}, + { } +}; +MODULE_DEVICE_TABLE(of, dwmac_generic_match); + +static struct platform_driver dwmac_generic_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = STMMAC_RESOURCE_NAME, + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = of_match_ptr(dwmac_generic_match), + }, +}; +module_platform_driver(dwmac_generic_driver); + +MODULE_DESCRIPTION("Generic dwmac driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c new file mode 100644 index 000000000000..7e3129e7f143 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c @@ -0,0 +1,365 @@ +/* + * Qualcomm Atheros IPQ806x GMAC glue layer + * + * Copyright (C) 2015 The Linux Foundation + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include <linux/device.h> +#include <linux/platform_device.h> +#include <linux/phy.h> +#include <linux/regmap.h> +#include <linux/clk.h> +#include <linux/reset.h> +#include <linux/of_net.h> +#include <linux/mfd/syscon.h> +#include <linux/stmmac.h> +#include <linux/of_mdio.h> +#include <linux/module.h> + +#include "stmmac_platform.h" + +#define NSS_COMMON_CLK_GATE 0x8 +#define NSS_COMMON_CLK_GATE_PTP_EN(x) BIT(0x10 + x) +#define NSS_COMMON_CLK_GATE_RGMII_RX_EN(x) BIT(0x9 + (x * 2)) +#define NSS_COMMON_CLK_GATE_RGMII_TX_EN(x) BIT(0x8 + (x * 2)) +#define NSS_COMMON_CLK_GATE_GMII_RX_EN(x) BIT(0x4 + x) +#define NSS_COMMON_CLK_GATE_GMII_TX_EN(x) BIT(0x0 + x) + +#define NSS_COMMON_CLK_DIV0 0xC +#define NSS_COMMON_CLK_DIV_OFFSET(x) (x * 8) +#define NSS_COMMON_CLK_DIV_MASK 0x7f + +#define NSS_COMMON_CLK_SRC_CTRL 0x14 +#define NSS_COMMON_CLK_SRC_CTRL_OFFSET(x) (1 << x) +/* Mode is coded on 1 bit but is different depending on the MAC ID: + * MAC0: QSGMII=0 RGMII=1 + * MAC1: QSGMII=0 SGMII=0 RGMII=1 + * MAC2 & MAC3: QSGMII=0 SGMII=1 + */ +#define NSS_COMMON_CLK_SRC_CTRL_RGMII(x) 1 +#define NSS_COMMON_CLK_SRC_CTRL_SGMII(x) ((x >= 2) ? 1 : 0) + +#define NSS_COMMON_MACSEC_CTL 0x28 +#define NSS_COMMON_MACSEC_CTL_EXT_BYPASS_EN(x) (1 << x) + +#define NSS_COMMON_GMAC_CTL(x) (0x30 + (x * 4)) +#define NSS_COMMON_GMAC_CTL_CSYS_REQ BIT(19) +#define NSS_COMMON_GMAC_CTL_PHY_IFACE_SEL BIT(16) +#define NSS_COMMON_GMAC_CTL_IFG_LIMIT_OFFSET 8 +#define NSS_COMMON_GMAC_CTL_IFG_OFFSET 0 +#define NSS_COMMON_GMAC_CTL_IFG_MASK 0x3f + +#define NSS_COMMON_CLK_DIV_RGMII_1000 1 +#define NSS_COMMON_CLK_DIV_RGMII_100 9 +#define NSS_COMMON_CLK_DIV_RGMII_10 99 +#define NSS_COMMON_CLK_DIV_SGMII_1000 0 +#define NSS_COMMON_CLK_DIV_SGMII_100 4 +#define NSS_COMMON_CLK_DIV_SGMII_10 49 + +#define QSGMII_PCS_MODE_CTL 0x68 +#define QSGMII_PCS_MODE_CTL_AUTONEG_EN(x) BIT((x * 8) + 7) + +#define QSGMII_PCS_CAL_LCKDT_CTL 0x120 +#define QSGMII_PCS_CAL_LCKDT_CTL_RST BIT(19) + +/* Only GMAC1/2/3 support SGMII and their CTL register are not contiguous */ +#define QSGMII_PHY_SGMII_CTL(x) ((x == 1) ? 0x134 : \ + (0x13c + (4 * (x - 2)))) +#define QSGMII_PHY_CDR_EN BIT(0) +#define QSGMII_PHY_RX_FRONT_EN BIT(1) +#define QSGMII_PHY_RX_SIGNAL_DETECT_EN BIT(2) +#define QSGMII_PHY_TX_DRIVER_EN BIT(3) +#define QSGMII_PHY_QSGMII_EN BIT(7) +#define QSGMII_PHY_PHASE_LOOP_GAIN_OFFSET 12 +#define QSGMII_PHY_PHASE_LOOP_GAIN_MASK 0x7 +#define QSGMII_PHY_RX_DC_BIAS_OFFSET 18 +#define QSGMII_PHY_RX_DC_BIAS_MASK 0x3 +#define QSGMII_PHY_RX_INPUT_EQU_OFFSET 20 +#define QSGMII_PHY_RX_INPUT_EQU_MASK 0x3 +#define QSGMII_PHY_CDR_PI_SLEW_OFFSET 22 +#define QSGMII_PHY_CDR_PI_SLEW_MASK 0x3 +#define QSGMII_PHY_TX_DRV_AMP_OFFSET 28 +#define QSGMII_PHY_TX_DRV_AMP_MASK 0xf + +struct ipq806x_gmac { + struct platform_device *pdev; + struct regmap *nss_common; + struct regmap *qsgmii_csr; + uint32_t id; + struct clk *core_clk; + phy_interface_t phy_mode; +}; + +static int get_clk_div_sgmii(struct ipq806x_gmac *gmac, unsigned int speed) +{ + struct device *dev = &gmac->pdev->dev; + int div; + + switch (speed) { + case SPEED_1000: + div = NSS_COMMON_CLK_DIV_SGMII_1000; + break; + + case SPEED_100: + div = NSS_COMMON_CLK_DIV_SGMII_100; + break; + + case SPEED_10: + div = NSS_COMMON_CLK_DIV_SGMII_10; + break; + + default: + dev_err(dev, "Speed %dMbps not supported in SGMII\n", speed); + return -EINVAL; + } + + return div; +} + +static int get_clk_div_rgmii(struct ipq806x_gmac *gmac, unsigned int speed) +{ + struct device *dev = &gmac->pdev->dev; + int div; + + switch (speed) { + case SPEED_1000: + div = NSS_COMMON_CLK_DIV_RGMII_1000; + break; + + case SPEED_100: + div = NSS_COMMON_CLK_DIV_RGMII_100; + break; + + case SPEED_10: + div = NSS_COMMON_CLK_DIV_RGMII_10; + break; + + default: + dev_err(dev, "Speed %dMbps not supported in RGMII\n", speed); + return -EINVAL; + } + + return div; +} + +static int ipq806x_gmac_set_speed(struct ipq806x_gmac *gmac, unsigned int speed) +{ + uint32_t clk_bits, val; + int div; + + switch (gmac->phy_mode) { + case PHY_INTERFACE_MODE_RGMII: + div = get_clk_div_rgmii(gmac, speed); + clk_bits = NSS_COMMON_CLK_GATE_RGMII_RX_EN(gmac->id) | + NSS_COMMON_CLK_GATE_RGMII_TX_EN(gmac->id); + break; + + case PHY_INTERFACE_MODE_SGMII: + div = get_clk_div_sgmii(gmac, speed); + clk_bits = NSS_COMMON_CLK_GATE_GMII_RX_EN(gmac->id) | + NSS_COMMON_CLK_GATE_GMII_TX_EN(gmac->id); + break; + + default: + dev_err(&gmac->pdev->dev, "Unsupported PHY mode: \"%s\"\n", + phy_modes(gmac->phy_mode)); + return -EINVAL; + } + + /* Disable the clocks */ + regmap_read(gmac->nss_common, NSS_COMMON_CLK_GATE, &val); + val &= ~clk_bits; + regmap_write(gmac->nss_common, NSS_COMMON_CLK_GATE, val); + + /* Set the divider */ + regmap_read(gmac->nss_common, NSS_COMMON_CLK_DIV0, &val); + val &= ~(NSS_COMMON_CLK_DIV_MASK + << NSS_COMMON_CLK_DIV_OFFSET(gmac->id)); + val |= div << NSS_COMMON_CLK_DIV_OFFSET(gmac->id); + regmap_write(gmac->nss_common, NSS_COMMON_CLK_DIV0, val); + + /* Enable the clock back */ + regmap_read(gmac->nss_common, NSS_COMMON_CLK_GATE, &val); + val |= clk_bits; + regmap_write(gmac->nss_common, NSS_COMMON_CLK_GATE, val); + + return 0; +} + +static void *ipq806x_gmac_of_parse(struct ipq806x_gmac *gmac) +{ + struct device *dev = &gmac->pdev->dev; + + gmac->phy_mode = of_get_phy_mode(dev->of_node); + if (gmac->phy_mode < 0) { + dev_err(dev, "missing phy mode property\n"); + return ERR_PTR(-EINVAL); + } + + if (of_property_read_u32(dev->of_node, "qcom,id", &gmac->id) < 0) { + dev_err(dev, "missing qcom id property\n"); + return ERR_PTR(-EINVAL); + } + + /* The GMACs are called 1 to 4 in the documentation, but to simplify the + * code and keep it consistent with the Linux convention, we'll number + * them from 0 to 3 here. + */ + if (gmac->id < 0 || gmac->id > 3) { + dev_err(dev, "invalid gmac id\n"); + return ERR_PTR(-EINVAL); + } + + gmac->core_clk = devm_clk_get(dev, "stmmaceth"); + if (IS_ERR(gmac->core_clk)) { + dev_err(dev, "missing stmmaceth clk property\n"); + return gmac->core_clk; + } + clk_set_rate(gmac->core_clk, 266000000); + + /* Setup the register map for the nss common registers */ + gmac->nss_common = syscon_regmap_lookup_by_phandle(dev->of_node, + "qcom,nss-common"); + if (IS_ERR(gmac->nss_common)) { + dev_err(dev, "missing nss-common node\n"); + return gmac->nss_common; + } + + /* Setup the register map for the qsgmii csr registers */ + gmac->qsgmii_csr = syscon_regmap_lookup_by_phandle(dev->of_node, + "qcom,qsgmii-csr"); + if (IS_ERR(gmac->qsgmii_csr)) { + dev_err(dev, "missing qsgmii-csr node\n"); + return gmac->qsgmii_csr; + } + + return NULL; +} + +static void *ipq806x_gmac_setup(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct ipq806x_gmac *gmac; + int val; + void *err; + + gmac = devm_kzalloc(dev, sizeof(*gmac), GFP_KERNEL); + if (!gmac) + return ERR_PTR(-ENOMEM); + + gmac->pdev = pdev; + + err = ipq806x_gmac_of_parse(gmac); + if (err) { + dev_err(dev, "device tree parsing error\n"); + return err; + } + + regmap_write(gmac->qsgmii_csr, QSGMII_PCS_CAL_LCKDT_CTL, + QSGMII_PCS_CAL_LCKDT_CTL_RST); + + /* Inter frame gap is set to 12 */ + val = 12 << NSS_COMMON_GMAC_CTL_IFG_OFFSET | + 12 << NSS_COMMON_GMAC_CTL_IFG_LIMIT_OFFSET; + /* We also initiate an AXI low power exit request */ + val |= NSS_COMMON_GMAC_CTL_CSYS_REQ; + switch (gmac->phy_mode) { + case PHY_INTERFACE_MODE_RGMII: + val |= NSS_COMMON_GMAC_CTL_PHY_IFACE_SEL; + break; + case PHY_INTERFACE_MODE_SGMII: + val &= ~NSS_COMMON_GMAC_CTL_PHY_IFACE_SEL; + break; + default: + dev_err(&pdev->dev, "Unsupported PHY mode: \"%s\"\n", + phy_modes(gmac->phy_mode)); + return NULL; + } + regmap_write(gmac->nss_common, NSS_COMMON_GMAC_CTL(gmac->id), val); + + /* Configure the clock src according to the mode */ + regmap_read(gmac->nss_common, NSS_COMMON_CLK_SRC_CTRL, &val); + val &= ~NSS_COMMON_CLK_SRC_CTRL_OFFSET(gmac->id); + switch (gmac->phy_mode) { + case PHY_INTERFACE_MODE_RGMII: + val |= NSS_COMMON_CLK_SRC_CTRL_RGMII(gmac->id) << + NSS_COMMON_CLK_SRC_CTRL_OFFSET(gmac->id); + break; + case PHY_INTERFACE_MODE_SGMII: + val |= NSS_COMMON_CLK_SRC_CTRL_SGMII(gmac->id) << + NSS_COMMON_CLK_SRC_CTRL_OFFSET(gmac->id); + break; + default: + dev_err(&pdev->dev, "Unsupported PHY mode: \"%s\"\n", + phy_modes(gmac->phy_mode)); + return NULL; + } + regmap_write(gmac->nss_common, NSS_COMMON_CLK_SRC_CTRL, val); + + /* Enable PTP clock */ + regmap_read(gmac->nss_common, NSS_COMMON_CLK_GATE, &val); + val |= NSS_COMMON_CLK_GATE_PTP_EN(gmac->id); + regmap_write(gmac->nss_common, NSS_COMMON_CLK_GATE, val); + + if (gmac->phy_mode == PHY_INTERFACE_MODE_SGMII) { + regmap_write(gmac->qsgmii_csr, QSGMII_PHY_SGMII_CTL(gmac->id), + QSGMII_PHY_CDR_EN | + QSGMII_PHY_RX_FRONT_EN | + QSGMII_PHY_RX_SIGNAL_DETECT_EN | + QSGMII_PHY_TX_DRIVER_EN | + QSGMII_PHY_QSGMII_EN | + 0x4 << QSGMII_PHY_PHASE_LOOP_GAIN_OFFSET | + 0x3 << QSGMII_PHY_RX_DC_BIAS_OFFSET | + 0x1 << QSGMII_PHY_RX_INPUT_EQU_OFFSET | + 0x2 << QSGMII_PHY_CDR_PI_SLEW_OFFSET | + 0xC << QSGMII_PHY_TX_DRV_AMP_OFFSET); + } + + return gmac; +} + +static void ipq806x_gmac_fix_mac_speed(void *priv, unsigned int speed) +{ + struct ipq806x_gmac *gmac = priv; + + ipq806x_gmac_set_speed(gmac, speed); +} + +static const struct stmmac_of_data ipq806x_gmac_data = { + .has_gmac = 1, + .setup = ipq806x_gmac_setup, + .fix_mac_speed = ipq806x_gmac_fix_mac_speed, +}; + +static const struct of_device_id ipq806x_gmac_dwmac_match[] = { + { .compatible = "qcom,ipq806x-gmac", .data = &ipq806x_gmac_data }, + { } +}; +MODULE_DEVICE_TABLE(of, ipq806x_gmac_dwmac_match); + +static struct platform_driver ipq806x_gmac_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "ipq806x-gmac-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = ipq806x_gmac_dwmac_match, + }, +}; +module_platform_driver(ipq806x_gmac_dwmac_driver); + +MODULE_AUTHOR("Mathieu Olivari <mathieu@codeaurora.org>"); +MODULE_DESCRIPTION("Qualcomm Atheros IPQ806x DWMAC specific glue layer"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c new file mode 100644 index 000000000000..cb888d3ebbdc --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -0,0 +1,99 @@ +/* + * DWMAC glue for NXP LPC18xx/LPC43xx Ethernet + * + * Copyright (C) 2015 Joachim Eastwood <manabian@gmail.com> + * + * This file is licensed under the terms of the GNU General Public + * License version 2. This program is licensed "as is" without any + * warranty of any kind, whether express or implied. + */ + +#include <linux/mfd/syscon.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_net.h> +#include <linux/phy.h> +#include <linux/platform_device.h> +#include <linux/regmap.h> +#include <linux/stmmac.h> + +#include "stmmac_platform.h" + +/* Register defines for CREG syscon */ +#define LPC18XX_CREG_CREG6 0x12c +# define LPC18XX_CREG_CREG6_ETHMODE_MASK 0x7 +# define LPC18XX_CREG_CREG6_ETHMODE_MII 0x0 +# define LPC18XX_CREG_CREG6_ETHMODE_RMII 0x4 + +struct lpc18xx_dwmac_priv_data { + struct regmap *reg; + int interface; +}; + +static void *lpc18xx_dwmac_setup(struct platform_device *pdev) +{ + struct lpc18xx_dwmac_priv_data *dwmac; + + dwmac = devm_kzalloc(&pdev->dev, sizeof(*dwmac), GFP_KERNEL); + if (!dwmac) + return ERR_PTR(-ENOMEM); + + dwmac->interface = of_get_phy_mode(pdev->dev.of_node); + if (dwmac->interface < 0) + return ERR_PTR(dwmac->interface); + + dwmac->reg = syscon_regmap_lookup_by_compatible("nxp,lpc1850-creg"); + if (IS_ERR(dwmac->reg)) { + dev_err(&pdev->dev, "Syscon lookup failed\n"); + return dwmac->reg; + } + + return dwmac; +} + +static int lpc18xx_dwmac_init(struct platform_device *pdev, void *priv) +{ + struct lpc18xx_dwmac_priv_data *dwmac = priv; + u8 ethmode; + + if (dwmac->interface == PHY_INTERFACE_MODE_MII) { + ethmode = LPC18XX_CREG_CREG6_ETHMODE_MII; + } else if (dwmac->interface == PHY_INTERFACE_MODE_RMII) { + ethmode = LPC18XX_CREG_CREG6_ETHMODE_RMII; + } else { + dev_err(&pdev->dev, "Only MII and RMII mode supported\n"); + return -EINVAL; + } + + regmap_update_bits(dwmac->reg, LPC18XX_CREG_CREG6, + LPC18XX_CREG_CREG6_ETHMODE_MASK, ethmode); + + return 0; +} + +static const struct stmmac_of_data lpc18xx_dwmac_data = { + .has_gmac = 1, + .setup = lpc18xx_dwmac_setup, + .init = lpc18xx_dwmac_init, +}; + +static const struct of_device_id lpc18xx_dwmac_match[] = { + { .compatible = "nxp,lpc1850-dwmac", .data = &lpc18xx_dwmac_data }, + { } +}; +MODULE_DEVICE_TABLE(of, lpc18xx_dwmac_match); + +static struct platform_driver lpc18xx_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "lpc18xx-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = lpc18xx_dwmac_match, + }, +}; +module_platform_driver(lpc18xx_dwmac_driver); + +MODULE_AUTHOR("Joachim Eastwood <manabian@gmail.com>"); +MODULE_DESCRIPTION("DWMAC glue for LPC18xx/43xx Ethernet"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson.c index cca028d632f6..61a324a87d09 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson.c @@ -15,6 +15,7 @@ #include <linux/ethtool.h> #include <linux/io.h> #include <linux/ioport.h> +#include <linux/module.h> #include <linux/platform_device.h> #include <linux/stmmac.h> @@ -63,7 +64,28 @@ static void *meson6_dwmac_setup(struct platform_device *pdev) return dwmac; } -const struct stmmac_of_data meson6_dwmac_data = { +static const struct stmmac_of_data meson6_dwmac_data = { .setup = meson6_dwmac_setup, .fix_mac_speed = meson6_dwmac_fix_mac_speed, }; + +static const struct of_device_id meson6_dwmac_match[] = { + { .compatible = "amlogic,meson6-dwmac", .data = &meson6_dwmac_data}, + { } +}; +MODULE_DEVICE_TABLE(of, meson6_dwmac_match); + +static struct platform_driver meson6_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "meson6-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = meson6_dwmac_match, + }, +}; +module_platform_driver(meson6_dwmac_driver); + +MODULE_AUTHOR("Beniamino Galvani <b.galvani@gmail.com>"); +MODULE_DESCRIPTION("Amlogic Meson DWMAC glue layer"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 6249a4ec08f0..30e28f0d9a60 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -22,13 +22,17 @@ #include <linux/phy.h> #include <linux/of_net.h> #include <linux/gpio.h> +#include <linux/module.h> #include <linux/of_gpio.h> #include <linux/of_device.h> +#include <linux/platform_device.h> #include <linux/regulator/consumer.h> #include <linux/delay.h> #include <linux/mfd/syscon.h> #include <linux/regmap.h> +#include "stmmac_platform.h" + struct rk_priv_data { struct platform_device *pdev; int phy_iface; @@ -428,10 +432,31 @@ static void rk_fix_speed(void *priv, unsigned int speed) dev_err(dev, "unsupported interface %d", bsp_priv->phy_iface); } -const struct stmmac_of_data rk3288_gmac_data = { +static const struct stmmac_of_data rk3288_gmac_data = { .has_gmac = 1, .fix_mac_speed = rk_fix_speed, .setup = rk_gmac_setup, .init = rk_gmac_init, .exit = rk_gmac_exit, }; + +static const struct of_device_id rk_gmac_dwmac_match[] = { + { .compatible = "rockchip,rk3288-gmac", .data = &rk3288_gmac_data}, + { } +}; +MODULE_DEVICE_TABLE(of, rk_gmac_dwmac_match); + +static struct platform_driver rk_gmac_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "rk_gmac-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = rk_gmac_dwmac_match, + }, +}; +module_platform_driver(rk_gmac_dwmac_driver); + +MODULE_AUTHOR("Chen-Zhi (Roger Chen) <roger.chen@rock-chips.com>"); +MODULE_DESCRIPTION("Rockchip RK3288 DWMAC specific glue layer"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 5a36bd2c7837..8141c5b844ae 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -257,9 +257,28 @@ static int socfpga_dwmac_init(struct platform_device *pdev, void *priv) return ret; } -const struct stmmac_of_data socfpga_gmac_data = { +static const struct stmmac_of_data socfpga_gmac_data = { .setup = socfpga_dwmac_probe, .init = socfpga_dwmac_init, .exit = socfpga_dwmac_exit, .fix_mac_speed = socfpga_dwmac_fix_mac_speed, }; + +static const struct of_device_id socfpga_dwmac_match[] = { + { .compatible = "altr,socfpga-stmmac", .data = &socfpga_gmac_data }, + { } +}; +MODULE_DEVICE_TABLE(of, socfpga_dwmac_match); + +static struct platform_driver socfpga_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "socfpga-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = socfpga_dwmac_match, + }, +}; +module_platform_driver(socfpga_dwmac_driver); + +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index bb6e2dc61bec..a2e8111c5d14 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -17,6 +17,7 @@ #include <linux/stmmac.h> #include <linux/phy.h> #include <linux/mfd/syscon.h> +#include <linux/module.h> #include <linux/regmap.h> #include <linux/clk.h> #include <linux/of.h> @@ -351,16 +352,40 @@ static void *sti_dwmac_setup(struct platform_device *pdev) return dwmac; } -const struct stmmac_of_data stih4xx_dwmac_data = { +static const struct stmmac_of_data stih4xx_dwmac_data = { .fix_mac_speed = stih4xx_fix_retime_src, .setup = sti_dwmac_setup, .init = stix4xx_init, .exit = sti_dwmac_exit, }; -const struct stmmac_of_data stid127_dwmac_data = { +static const struct stmmac_of_data stid127_dwmac_data = { .fix_mac_speed = stid127_fix_retime_src, .setup = sti_dwmac_setup, .init = stid127_init, .exit = sti_dwmac_exit, }; + +static const struct of_device_id sti_dwmac_match[] = { + { .compatible = "st,stih415-dwmac", .data = &stih4xx_dwmac_data}, + { .compatible = "st,stih416-dwmac", .data = &stih4xx_dwmac_data}, + { .compatible = "st,stid127-dwmac", .data = &stid127_dwmac_data}, + { .compatible = "st,stih407-dwmac", .data = &stih4xx_dwmac_data}, + { } +}; +MODULE_DEVICE_TABLE(of, sti_dwmac_match); + +static struct platform_driver sti_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "sti-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = sti_dwmac_match, + }, +}; +module_platform_driver(sti_dwmac_driver); + +MODULE_AUTHOR("Srinivas Kandagatla <srinivas.kandagatla@st.com>"); +MODULE_DESCRIPTION("STMicroelectronics DWMAC Specific Glue layer"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c index c5ea9ab75b03..15048ca39759 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c @@ -18,7 +18,9 @@ #include <linux/stmmac.h> #include <linux/clk.h> +#include <linux/module.h> #include <linux/phy.h> +#include <linux/platform_device.h> #include <linux/of_net.h> #include <linux/regulator/consumer.h> @@ -132,7 +134,7 @@ static void sun7i_fix_speed(void *priv, unsigned int speed) /* of_data specifying hardware features and callbacks. * hardware features were copied from Allwinner drivers. */ -const struct stmmac_of_data sun7i_gmac_data = { +static const struct stmmac_of_data sun7i_gmac_data = { .has_gmac = 1, .tx_coe = 1, .fix_mac_speed = sun7i_fix_speed, @@ -140,3 +142,24 @@ const struct stmmac_of_data sun7i_gmac_data = { .init = sun7i_gmac_init, .exit = sun7i_gmac_exit, }; + +static const struct of_device_id sun7i_dwmac_match[] = { + { .compatible = "allwinner,sun7i-a20-gmac", .data = &sun7i_gmac_data}, + { } +}; +MODULE_DEVICE_TABLE(of, sun7i_dwmac_match); + +static struct platform_driver sun7i_dwmac_driver = { + .probe = stmmac_pltfr_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "sun7i-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = sun7i_dwmac_match, + }, +}; +module_platform_driver(sun7i_dwmac_driver); + +MODULE_AUTHOR("Chen-Yu Tsai <wens@csie.org>"); +MODULE_DESCRIPTION("Allwinner sunxi DWMAC specific glue layer"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 73bab983edd9..1f3b33a6c6a8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -34,6 +34,14 @@ #include <linux/ptp_clock_kernel.h> #include <linux/reset.h> +struct stmmac_resources { + void __iomem *addr; + const char *mac; + int wol_irq; + int lpi_irq; + int irq; +}; + struct stmmac_tx_info { dma_addr_t buf; bool map_as_page; @@ -135,9 +143,9 @@ void stmmac_ptp_unregister(struct stmmac_priv *priv); int stmmac_resume(struct net_device *ndev); int stmmac_suspend(struct net_device *ndev); int stmmac_dvr_remove(struct net_device *ndev); -struct stmmac_priv *stmmac_dvr_probe(struct device *device, - struct plat_stmmacenet_data *plat_dat, - void __iomem *addr); +int stmmac_dvr_probe(struct device *device, + struct plat_stmmacenet_data *plat_dat, + struct stmmac_resources *res); void stmmac_disable_eee_mode(struct stmmac_priv *priv); bool stmmac_eee_init(struct stmmac_priv *priv); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 2c5ce2baca87..a5156739e1e1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -52,6 +52,7 @@ #include "stmmac_ptp.h" #include "stmmac.h" #include <linux/reset.h> +#include <linux/of_mdio.h> #define STMMAC_ALIGN(x) L1_CACHE_ALIGN(x) @@ -816,18 +817,25 @@ static int stmmac_init_phy(struct net_device *dev) priv->speed = 0; priv->oldduplex = -1; - if (priv->plat->phy_bus_name) - snprintf(bus_id, MII_BUS_ID_SIZE, "%s-%x", - priv->plat->phy_bus_name, priv->plat->bus_id); - else - snprintf(bus_id, MII_BUS_ID_SIZE, "stmmac-%x", - priv->plat->bus_id); + if (priv->plat->phy_node) { + phydev = of_phy_connect(dev, priv->plat->phy_node, + &stmmac_adjust_link, 0, interface); + } else { + if (priv->plat->phy_bus_name) + snprintf(bus_id, MII_BUS_ID_SIZE, "%s-%x", + priv->plat->phy_bus_name, priv->plat->bus_id); + else + snprintf(bus_id, MII_BUS_ID_SIZE, "stmmac-%x", + priv->plat->bus_id); - snprintf(phy_id_fmt, MII_BUS_ID_SIZE + 3, PHY_ID_FMT, bus_id, - priv->plat->phy_addr); - pr_debug("stmmac_init_phy: trying to attach to %s\n", phy_id_fmt); + snprintf(phy_id_fmt, MII_BUS_ID_SIZE + 3, PHY_ID_FMT, bus_id, + priv->plat->phy_addr); + pr_debug("stmmac_init_phy: trying to attach to %s\n", + phy_id_fmt); - phydev = phy_connect(dev, phy_id_fmt, &stmmac_adjust_link, interface); + phydev = phy_connect(dev, phy_id_fmt, &stmmac_adjust_link, + interface); + } if (IS_ERR(phydev)) { pr_err("%s: Could not attach to PHY\n", dev->name); @@ -848,7 +856,7 @@ static int stmmac_init_phy(struct net_device *dev) * device as well. * Note: phydev->phy_id is the result of reading the UID PHY registers. */ - if (phydev->phy_id == 0) { + if (!priv->plat->phy_node && phydev->phy_id == 0) { phy_disconnect(phydev); return -ENODEV; } @@ -975,13 +983,11 @@ static int stmmac_init_rx_buffers(struct stmmac_priv *priv, struct dma_desc *p, { struct sk_buff *skb; - skb = __netdev_alloc_skb(priv->dev, priv->dma_buf_sz + NET_IP_ALIGN, - flags); + skb = __netdev_alloc_skb_ip_align(priv->dev, priv->dma_buf_sz, flags); if (!skb) { pr_err("%s: Rx init fails; skb is NULL\n", __func__); return -ENOMEM; } - skb_reserve(skb, NET_IP_ALIGN); priv->rx_skbuff[i] = skb; priv->rx_skbuff_dma[i] = dma_map_single(priv->device, skb->data, priv->dma_buf_sz, @@ -2800,16 +2806,15 @@ static int stmmac_hw_init(struct stmmac_priv *priv) * stmmac_dvr_probe * @device: device pointer * @plat_dat: platform data pointer - * @addr: iobase memory address + * @res: stmmac resource pointer * Description: this is the main probe function used to * call the alloc_etherdev, allocate the priv structure. * Return: - * on success the new private structure is returned, otherwise the error - * pointer. + * returns 0 on success, otherwise errno. */ -struct stmmac_priv *stmmac_dvr_probe(struct device *device, - struct plat_stmmacenet_data *plat_dat, - void __iomem *addr) +int stmmac_dvr_probe(struct device *device, + struct plat_stmmacenet_data *plat_dat, + struct stmmac_resources *res) { int ret = 0; struct net_device *ndev = NULL; @@ -2817,7 +2822,7 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, ndev = alloc_etherdev(sizeof(struct stmmac_priv)); if (!ndev) - return ERR_PTR(-ENOMEM); + return -ENOMEM; SET_NETDEV_DEV(ndev, device); @@ -2828,8 +2833,17 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, stmmac_set_ethtool_ops(ndev); priv->pause = pause; priv->plat = plat_dat; - priv->ioaddr = addr; - priv->dev->base_addr = (unsigned long)addr; + priv->ioaddr = res->addr; + priv->dev->base_addr = (unsigned long)res->addr; + + priv->dev->irq = res->irq; + priv->wol_irq = res->wol_irq; + priv->lpi_irq = res->lpi_irq; + + if (res->mac) + memcpy(priv->dev->dev_addr, res->mac, ETH_ALEN); + + dev_set_drvdata(device, priv); /* Verify driver arguments */ stmmac_verify_args(); @@ -2944,7 +2958,7 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, } } - return priv; + return 0; error_mdio_register: unregister_netdev(ndev); @@ -2957,7 +2971,7 @@ error_pclk_get: error_clk_get: free_netdev(ndev); - return ERR_PTR(ret); + return ret; } EXPORT_SYMBOL_GPL(stmmac_dvr_probe); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 3bca908716e2..d71a721ea61c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -163,7 +163,7 @@ static int stmmac_pci_probe(struct pci_dev *pdev, { struct stmmac_pci_info *info = (struct stmmac_pci_info *)id->driver_data; struct plat_stmmacenet_data *plat; - struct stmmac_priv *priv; + struct stmmac_resources res; int i; int ret; @@ -214,19 +214,12 @@ static int stmmac_pci_probe(struct pci_dev *pdev, pci_enable_msi(pdev); - priv = stmmac_dvr_probe(&pdev->dev, plat, pcim_iomap_table(pdev)[i]); - if (IS_ERR(priv)) { - dev_err(&pdev->dev, "%s: main driver probe failed\n", __func__); - return PTR_ERR(priv); - } - priv->dev->irq = pdev->irq; - priv->wol_irq = pdev->irq; - - pci_set_drvdata(pdev, priv->dev); + memset(&res, 0, sizeof(res)); + res.addr = pcim_iomap_table(pdev)[i]; + res.wol_irq = pdev->irq; + res.irq = pdev->irq; - dev_dbg(&pdev->dev, "STMMAC PCI driver registration completed\n"); - - return 0; + return stmmac_dvr_probe(&pdev->dev, plat, &res); } /** diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 68aec5c460db..f3918c7e7eeb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -28,29 +28,11 @@ #include <linux/of.h> #include <linux/of_net.h> #include <linux/of_device.h> +#include <linux/of_mdio.h> #include "stmmac.h" #include "stmmac_platform.h" -static const struct of_device_id stmmac_dt_ids[] = { - /* SoC specific glue layers should come before generic bindings */ - { .compatible = "rockchip,rk3288-gmac", .data = &rk3288_gmac_data}, - { .compatible = "amlogic,meson6-dwmac", .data = &meson6_dwmac_data}, - { .compatible = "allwinner,sun7i-a20-gmac", .data = &sun7i_gmac_data}, - { .compatible = "st,stih415-dwmac", .data = &stih4xx_dwmac_data}, - { .compatible = "st,stih416-dwmac", .data = &stih4xx_dwmac_data}, - { .compatible = "st,stid127-dwmac", .data = &stid127_dwmac_data}, - { .compatible = "st,stih407-dwmac", .data = &stih4xx_dwmac_data}, - { .compatible = "altr,socfpga-stmmac", .data = &socfpga_gmac_data }, - { .compatible = "st,spear600-gmac"}, - { .compatible = "snps,dwmac-3.610"}, - { .compatible = "snps,dwmac-3.70a"}, - { .compatible = "snps,dwmac-3.710"}, - { .compatible = "snps,dwmac"}, - { /* sentinel */ } -}; -MODULE_DEVICE_TABLE(of, stmmac_dt_ids); - #ifdef CONFIG_OF /** @@ -129,14 +111,9 @@ static int stmmac_probe_config_dt(struct platform_device *pdev, struct device_node *np = pdev->dev.of_node; struct stmmac_dma_cfg *dma_cfg; const struct of_device_id *device; + struct device *dev = &pdev->dev; - if (!np) - return -ENODEV; - - device = of_match_device(stmmac_dt_ids, &pdev->dev); - if (!device) - return -ENODEV; - + device = of_match_device(dev->driver->of_match_table, dev); if (device->data) { const struct stmmac_of_data *data = device->data; plat->has_gmac = data->has_gmac; @@ -168,13 +145,24 @@ static int stmmac_probe_config_dt(struct platform_device *pdev, /* Default to phy auto-detection */ plat->phy_addr = -1; + /* If we find a phy-handle property, use it as the PHY */ + plat->phy_node = of_parse_phandle(np, "phy-handle", 0); + + /* If phy-handle is not specified, check if we have a fixed-phy */ + if (!plat->phy_node && of_phy_is_fixed_link(np)) { + if ((of_phy_register_fixed_link(np) < 0)) + return -ENODEV; + + plat->phy_node = of_node_get(np); + } + /* "snps,phy-addr" is not a standard property. Mark it as deprecated * and warn of its use. Remove this when phy node support is added. */ if (of_property_read_u32(np, "snps,phy-addr", &plat->phy_addr) == 0) dev_warn(&pdev->dev, "snps,phy-addr property is deprecated\n"); - if (plat->phy_bus_name) + if (plat->phy_node || plat->phy_bus_name) plat->mdio_bus_data = NULL; else plat->mdio_bus_data = @@ -232,8 +220,10 @@ static int stmmac_probe_config_dt(struct platform_device *pdev, if (of_find_property(np, "snps,pbl", NULL)) { dma_cfg = devm_kzalloc(&pdev->dev, sizeof(*dma_cfg), GFP_KERNEL); - if (!dma_cfg) + if (!dma_cfg) { + of_node_put(np); return -ENOMEM; + } plat->dma_cfg = dma_cfg; of_property_read_u32(np, "snps,pbl", &dma_cfg->pbl); dma_cfg->fixed_burst = @@ -268,27 +258,26 @@ static int stmmac_probe_config_dt(struct platform_device *pdev, * the necessary platform resources, invoke custom helper (if required) and * invoke the main probe function. */ -static int stmmac_pltfr_probe(struct platform_device *pdev) +int stmmac_pltfr_probe(struct platform_device *pdev) { + struct stmmac_resources stmmac_res; int ret = 0; struct resource *res; struct device *dev = &pdev->dev; - void __iomem *addr = NULL; - struct stmmac_priv *priv = NULL; struct plat_stmmacenet_data *plat_dat = NULL; - const char *mac = NULL; - int irq, wol_irq, lpi_irq; + + memset(&stmmac_res, 0, sizeof(stmmac_res)); /* Get IRQ information early to have an ability to ask for deferred * probe if needed before we went too far with resource allocation. */ - irq = platform_get_irq_byname(pdev, "macirq"); - if (irq < 0) { - if (irq != -EPROBE_DEFER) { + stmmac_res.irq = platform_get_irq_byname(pdev, "macirq"); + if (stmmac_res.irq < 0) { + if (stmmac_res.irq != -EPROBE_DEFER) { dev_err(dev, "MAC IRQ configuration information not found\n"); } - return irq; + return stmmac_res.irq; } /* On some platforms e.g. SPEAr the wake up irq differs from the mac irq @@ -298,21 +287,21 @@ static int stmmac_pltfr_probe(struct platform_device *pdev) * In case the wake up interrupt is not passed from the platform * so the driver will continue to use the mac irq (ndev->irq) */ - wol_irq = platform_get_irq_byname(pdev, "eth_wake_irq"); - if (wol_irq < 0) { - if (wol_irq == -EPROBE_DEFER) + stmmac_res.wol_irq = platform_get_irq_byname(pdev, "eth_wake_irq"); + if (stmmac_res.wol_irq < 0) { + if (stmmac_res.wol_irq == -EPROBE_DEFER) return -EPROBE_DEFER; - wol_irq = irq; + stmmac_res.wol_irq = stmmac_res.irq; } - lpi_irq = platform_get_irq_byname(pdev, "eth_lpi"); - if (lpi_irq == -EPROBE_DEFER) + stmmac_res.lpi_irq = platform_get_irq_byname(pdev, "eth_lpi"); + if (stmmac_res.lpi_irq == -EPROBE_DEFER) return -EPROBE_DEFER; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - addr = devm_ioremap_resource(dev, res); - if (IS_ERR(addr)) - return PTR_ERR(addr); + stmmac_res.addr = devm_ioremap_resource(dev, res); + if (IS_ERR(stmmac_res.addr)) + return PTR_ERR(stmmac_res.addr); plat_dat = dev_get_platdata(&pdev->dev); @@ -332,7 +321,7 @@ static int stmmac_pltfr_probe(struct platform_device *pdev) plat_dat->unicast_filter_entries = 1; if (pdev->dev.of_node) { - ret = stmmac_probe_config_dt(pdev, plat_dat, &mac); + ret = stmmac_probe_config_dt(pdev, plat_dat, &stmmac_res.mac); if (ret) { pr_err("%s: main dt probe failed", __func__); return ret; @@ -353,27 +342,9 @@ static int stmmac_pltfr_probe(struct platform_device *pdev) return ret; } - priv = stmmac_dvr_probe(&(pdev->dev), plat_dat, addr); - if (IS_ERR(priv)) { - pr_err("%s: main driver probe failed", __func__); - return PTR_ERR(priv); - } - - /* Copy IRQ values to priv structure which is now avaialble */ - priv->dev->irq = irq; - priv->wol_irq = wol_irq; - priv->lpi_irq = lpi_irq; - - /* Get MAC address if available (DT) */ - if (mac) - memcpy(priv->dev->dev_addr, mac, ETH_ALEN); - - platform_set_drvdata(pdev, priv->dev); - - pr_debug("STMMAC platform driver registration completed"); - - return 0; + return stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); } +EXPORT_SYMBOL_GPL(stmmac_pltfr_probe); /** * stmmac_pltfr_remove @@ -381,7 +352,7 @@ static int stmmac_pltfr_probe(struct platform_device *pdev) * Description: this function calls the main to free the net resources * and calls the platforms hook and release the resources (e.g. mem). */ -static int stmmac_pltfr_remove(struct platform_device *pdev) +int stmmac_pltfr_remove(struct platform_device *pdev) { struct net_device *ndev = platform_get_drvdata(pdev); struct stmmac_priv *priv = netdev_priv(ndev); @@ -395,6 +366,7 @@ static int stmmac_pltfr_remove(struct platform_device *pdev) return ret; } +EXPORT_SYMBOL_GPL(stmmac_pltfr_remove); #ifdef CONFIG_PM_SLEEP /** @@ -438,21 +410,6 @@ static int stmmac_pltfr_resume(struct device *dev) } #endif /* CONFIG_PM_SLEEP */ -static SIMPLE_DEV_PM_OPS(stmmac_pltfr_pm_ops, - stmmac_pltfr_suspend, stmmac_pltfr_resume); - -static struct platform_driver stmmac_pltfr_driver = { - .probe = stmmac_pltfr_probe, - .remove = stmmac_pltfr_remove, - .driver = { - .name = STMMAC_RESOURCE_NAME, - .pm = &stmmac_pltfr_pm_ops, - .of_match_table = of_match_ptr(stmmac_dt_ids), - }, -}; - -module_platform_driver(stmmac_pltfr_driver); - -MODULE_DESCRIPTION("STMMAC 10/100/1000 Ethernet PLATFORM driver"); -MODULE_AUTHOR("Giuseppe Cavallaro <peppe.cavallaro@st.com>"); -MODULE_LICENSE("GPL"); +SIMPLE_DEV_PM_OPS(stmmac_pltfr_pm_ops, stmmac_pltfr_suspend, + stmmac_pltfr_resume); +EXPORT_SYMBOL_GPL(stmmac_pltfr_pm_ops); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h index 093eb99e5ffd..71da86d7bd00 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h @@ -19,11 +19,8 @@ #ifndef __STMMAC_PLATFORM_H__ #define __STMMAC_PLATFORM_H__ -extern const struct stmmac_of_data meson6_dwmac_data; -extern const struct stmmac_of_data sun7i_gmac_data; -extern const struct stmmac_of_data stih4xx_dwmac_data; -extern const struct stmmac_of_data stid127_dwmac_data; -extern const struct stmmac_of_data socfpga_gmac_data; -extern const struct stmmac_of_data rk3288_gmac_data; +int stmmac_pltfr_probe(struct platform_device *pdev); +int stmmac_pltfr_remove(struct platform_device *pdev); +extern const struct dev_pm_ops stmmac_pltfr_pm_ops; #endif /* __STMMAC_PLATFORM_H__ */ diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index b536b4c82752..462820514fae 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1361,7 +1361,6 @@ static int cpsw_ndo_stop(struct net_device *ndev) if (cpsw_common_res_usage_state(priv) <= 1) { cpts_unregister(priv->cpts); cpsw_intr_disable(priv); - cpdma_ctlr_int_ctrl(priv->dma, false); cpdma_ctlr_stop(priv->dma); cpsw_ale_stop(priv->ale); } @@ -1456,7 +1455,7 @@ static void cpsw_hwtstamp_v2(struct cpsw_priv *priv) if (priv->cpts->rx_enable) ctrl |= CTRL_V2_RX_TS_BITS; - break; + break; case CPSW_VERSION_3: default: ctrl &= ~CTRL_V3_ALL_TS_MASK; @@ -1466,7 +1465,7 @@ static void cpsw_hwtstamp_v2(struct cpsw_priv *priv) if (priv->cpts->rx_enable) ctrl |= CTRL_V3_RX_TS_BITS; - break; + break; } mtype = (30 << TS_SEQ_ID_OFFSET_SHIFT) | EVENT_MSG_BITS; @@ -1589,10 +1588,8 @@ static void cpsw_ndo_tx_timeout(struct net_device *ndev) cpsw_err(priv, tx_err, "transmit timeout, restarting dma\n"); ndev->stats.tx_errors++; cpsw_intr_disable(priv); - cpdma_ctlr_int_ctrl(priv->dma, false); cpdma_chan_stop(priv->txch); cpdma_chan_start(priv->txch); - cpdma_ctlr_int_ctrl(priv->dma, true); cpsw_intr_enable(priv); } @@ -1629,10 +1626,8 @@ static void cpsw_ndo_poll_controller(struct net_device *ndev) struct cpsw_priv *priv = netdev_priv(ndev); cpsw_intr_disable(priv); - cpdma_ctlr_int_ctrl(priv->dma, false); cpsw_rx_interrupt(priv->irqs_table[0], priv); cpsw_tx_interrupt(priv->irqs_table[1], priv); - cpdma_ctlr_int_ctrl(priv->dma, true); cpsw_intr_enable(priv); } #endif diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 6e927b4583aa..43b061bd8e07 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -268,39 +268,6 @@ int cpsw_ale_flush_multicast(struct cpsw_ale *ale, int port_mask, int vid) } EXPORT_SYMBOL_GPL(cpsw_ale_flush_multicast); -static void cpsw_ale_flush_ucast(struct cpsw_ale *ale, u32 *ale_entry, - int port_mask) -{ - int port; - - port = cpsw_ale_get_port_num(ale_entry); - if ((BIT(port) & port_mask) == 0) - return; /* ports dont intersect, not interested */ - cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE); -} - -int cpsw_ale_flush(struct cpsw_ale *ale, int port_mask) -{ - u32 ale_entry[ALE_ENTRY_WORDS]; - int ret, idx; - - for (idx = 0; idx < ale->params.ale_entries; idx++) { - cpsw_ale_read(ale, idx, ale_entry); - ret = cpsw_ale_get_entry_type(ale_entry); - if (ret != ALE_TYPE_ADDR && ret != ALE_TYPE_VLAN_ADDR) - continue; - - if (cpsw_ale_get_mcast(ale_entry)) - cpsw_ale_flush_mcast(ale, ale_entry, port_mask); - else - cpsw_ale_flush_ucast(ale, ale_entry, port_mask); - - cpsw_ale_write(ale, idx, ale_entry); - } - return 0; -} -EXPORT_SYMBOL_GPL(cpsw_ale_flush); - static inline void cpsw_ale_set_vlan_entry_type(u32 *ale_entry, int flags, u16 vid) { @@ -752,18 +719,6 @@ static void cpsw_ale_timer(unsigned long arg) } } -int cpsw_ale_set_ageout(struct cpsw_ale *ale, int ageout) -{ - del_timer_sync(&ale->timer); - ale->ageout = ageout * HZ; - if (ale->ageout) { - ale->timer.expires = jiffies + ale->ageout; - add_timer(&ale->timer); - } - return 0; -} -EXPORT_SYMBOL_GPL(cpsw_ale_set_ageout); - void cpsw_ale_start(struct cpsw_ale *ale) { u32 rev; diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index af1e7ecd87c6..a7001894f3da 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -90,8 +90,6 @@ int cpsw_ale_destroy(struct cpsw_ale *ale); void cpsw_ale_start(struct cpsw_ale *ale); void cpsw_ale_stop(struct cpsw_ale *ale); -int cpsw_ale_set_ageout(struct cpsw_ale *ale, int ageout); -int cpsw_ale_flush(struct cpsw_ale *ale, int port_mask); int cpsw_ale_flush_multicast(struct cpsw_ale *ale, int port_mask, int vid); int cpsw_ale_add_ucast(struct cpsw_ale *ale, u8 *addr, int port, int flags, u16 vid); diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 43efc3a0cda5..5ec4ed3f6c8d 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -537,7 +537,7 @@ int netcp_unregister_rxhook(struct netcp_intf *netcp_priv, int order, static void netcp_frag_free(bool is_frag, void *ptr) { if (is_frag) - put_page(virt_to_head_page(ptr)); + skb_free_frag(ptr); else kfree(ptr); } @@ -698,7 +698,6 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp) } } - netcp->ndev->last_rx = jiffies; netcp->ndev->stats.rx_packets++; netcp->ndev->stats.rx_bytes += skb->len; diff --git a/drivers/net/ethernet/tile/tilepro.c b/drivers/net/ethernet/tile/tilepro.c index 3d8f60d9643e..6f0a4495c7f3 100644 --- a/drivers/net/ethernet/tile/tilepro.c +++ b/drivers/net/ethernet/tile/tilepro.c @@ -721,9 +721,6 @@ static bool tile_net_poll_aux(struct tile_net_cpu *info, int index) if (!hash_default) __inv_buffer(buf, len); - /* ISSUE: Is this needed? */ - dev->last_rx = jiffies; - #ifdef TILE_NET_DUMP_PACKETS dump_packet(buf, len, "rx"); #endif /* TILE_NET_DUMP_PACKETS */ diff --git a/drivers/net/ethernet/toshiba/spider_net.c b/drivers/net/ethernet/toshiba/spider_net.c index 8e9371a3388a..3c54a2cae5df 100644 --- a/drivers/net/ethernet/toshiba/spider_net.c +++ b/drivers/net/ethernet/toshiba/spider_net.c @@ -604,8 +604,7 @@ spider_net_set_multi(struct net_device *netdev) int i; u32 reg; struct spider_net_card *card = netdev_priv(netdev); - unsigned long bitmask[SPIDER_NET_MULTICAST_HASHES / BITS_PER_LONG] = - {0, }; + DECLARE_BITMAP(bitmask, SPIDER_NET_MULTICAST_HASHES) = {}; spider_net_set_promisc(card); diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c index de2850497c09..725106f75d42 100644 --- a/drivers/net/ethernet/via/via-rhine.c +++ b/drivers/net/ethernet/via/via-rhine.c @@ -472,8 +472,7 @@ struct rhine_private { /* Frequently used values: keep some adjacent for cache effect. */ u32 quirks; - struct rx_desc *rx_head_desc; - unsigned int cur_rx, dirty_rx; /* Producer/consumer ring indices */ + unsigned int cur_rx; unsigned int cur_tx, dirty_tx; unsigned int rx_buf_sz; /* Based on MTU+slack. */ struct rhine_stats rx_stats; @@ -1213,17 +1212,61 @@ static void free_ring(struct net_device* dev) } -static void alloc_rbufs(struct net_device *dev) +struct rhine_skb_dma { + struct sk_buff *skb; + dma_addr_t dma; +}; + +static inline int rhine_skb_dma_init(struct net_device *dev, + struct rhine_skb_dma *sd) { struct rhine_private *rp = netdev_priv(dev); struct device *hwdev = dev->dev.parent; - dma_addr_t next; + const int size = rp->rx_buf_sz; + + sd->skb = netdev_alloc_skb(dev, size); + if (!sd->skb) + return -ENOMEM; + + sd->dma = dma_map_single(hwdev, sd->skb->data, size, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(hwdev, sd->dma))) { + netif_err(rp, drv, dev, "Rx DMA mapping failure\n"); + dev_kfree_skb_any(sd->skb); + return -EIO; + } + + return 0; +} + +static void rhine_reset_rbufs(struct rhine_private *rp) +{ int i; - rp->dirty_rx = rp->cur_rx = 0; + rp->cur_rx = 0; + + for (i = 0; i < RX_RING_SIZE; i++) + rp->rx_ring[i].rx_status = cpu_to_le32(DescOwn); +} + +static inline void rhine_skb_dma_nic_store(struct rhine_private *rp, + struct rhine_skb_dma *sd, int entry) +{ + rp->rx_skbuff_dma[entry] = sd->dma; + rp->rx_skbuff[entry] = sd->skb; + + rp->rx_ring[entry].addr = cpu_to_le32(sd->dma); + dma_wmb(); +} + +static void free_rbufs(struct net_device* dev); + +static int alloc_rbufs(struct net_device *dev) +{ + struct rhine_private *rp = netdev_priv(dev); + dma_addr_t next; + int rc, i; rp->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32); - rp->rx_head_desc = &rp->rx_ring[0]; next = rp->rx_ring_dma; /* Init the ring entries */ @@ -1239,23 +1282,20 @@ static void alloc_rbufs(struct net_device *dev) /* Fill in the Rx buffers. Handle allocation failure gracefully. */ for (i = 0; i < RX_RING_SIZE; i++) { - struct sk_buff *skb = netdev_alloc_skb(dev, rp->rx_buf_sz); - rp->rx_skbuff[i] = skb; - if (skb == NULL) - break; + struct rhine_skb_dma sd; - rp->rx_skbuff_dma[i] = - dma_map_single(hwdev, skb->data, rp->rx_buf_sz, - DMA_FROM_DEVICE); - if (dma_mapping_error(hwdev, rp->rx_skbuff_dma[i])) { - rp->rx_skbuff_dma[i] = 0; - dev_kfree_skb(skb); - break; + rc = rhine_skb_dma_init(dev, &sd); + if (rc < 0) { + free_rbufs(dev); + goto out; } - rp->rx_ring[i].addr = cpu_to_le32(rp->rx_skbuff_dma[i]); - rp->rx_ring[i].rx_status = cpu_to_le32(DescOwn); + + rhine_skb_dma_nic_store(rp, &sd, i); } - rp->dirty_rx = (unsigned int)(i - RX_RING_SIZE); + + rhine_reset_rbufs(rp); +out: + return rc; } static void free_rbufs(struct net_device* dev) @@ -1659,16 +1699,18 @@ static int rhine_open(struct net_device *dev) rc = request_irq(rp->irq, rhine_interrupt, IRQF_SHARED, dev->name, dev); if (rc) - return rc; + goto out; netif_dbg(rp, ifup, dev, "%s() irq %d\n", __func__, rp->irq); rc = alloc_ring(dev); - if (rc) { - free_irq(rp->irq, dev); - return rc; - } - alloc_rbufs(dev); + if (rc < 0) + goto out_free_irq; + + rc = alloc_rbufs(dev); + if (rc < 0) + goto out_free_ring; + alloc_tbufs(dev); rhine_chip_reset(dev); rhine_task_enable(rp); @@ -1680,7 +1722,14 @@ static int rhine_open(struct net_device *dev) netif_start_queue(dev); - return 0; +out: + return rc; + +out_free_ring: + free_ring(dev); +out_free_irq: + free_irq(rp->irq, dev); + goto out; } static void rhine_reset_task(struct work_struct *work) @@ -1700,9 +1749,9 @@ static void rhine_reset_task(struct work_struct *work) /* clear all descriptors */ free_tbufs(dev); - free_rbufs(dev); alloc_tbufs(dev); - alloc_rbufs(dev); + + rhine_reset_rbufs(rp); /* Reinitialize the hardware. */ rhine_chip_reset(dev); @@ -1730,6 +1779,11 @@ static void rhine_tx_timeout(struct net_device *dev) schedule_work(&rp->reset_task); } +static inline bool rhine_tx_queue_full(struct rhine_private *rp) +{ + return (rp->cur_tx - rp->dirty_tx) >= TX_QUEUE_LEN; +} + static netdev_tx_t rhine_start_tx(struct sk_buff *skb, struct net_device *dev) { @@ -1800,11 +1854,17 @@ static netdev_tx_t rhine_start_tx(struct sk_buff *skb, netdev_sent_queue(dev, skb->len); /* lock eth irq */ - wmb(); + dma_wmb(); rp->tx_ring[entry].tx_status |= cpu_to_le32(DescOwn); wmb(); rp->cur_tx++; + /* + * Nobody wants cur_tx write to rot for ages after the NIC will have + * seen the transmit request, especially as the transmit completion + * handler could miss it. + */ + smp_wmb(); /* Non-x86 Todo: explicitly flush cache lines here. */ @@ -1817,8 +1877,14 @@ static netdev_tx_t rhine_start_tx(struct sk_buff *skb, ioaddr + ChipCmd1); IOSYNC; - if (rp->cur_tx == rp->dirty_tx + TX_QUEUE_LEN) + /* dirty_tx may be pessimistically out-of-sync. See rhine_tx. */ + if (rhine_tx_queue_full(rp)) { netif_stop_queue(dev); + smp_rmb(); + /* Rejuvenate. */ + if (!rhine_tx_queue_full(rp)) + netif_wake_queue(dev); + } netif_dbg(rp, tx_queued, dev, "Transmit frame #%d queued in slot %d\n", rp->cur_tx - 1, entry); @@ -1866,13 +1932,24 @@ static void rhine_tx(struct net_device *dev) { struct rhine_private *rp = netdev_priv(dev); struct device *hwdev = dev->dev.parent; - int txstatus = 0, entry = rp->dirty_tx % TX_RING_SIZE; unsigned int pkts_compl = 0, bytes_compl = 0; + unsigned int dirty_tx = rp->dirty_tx; + unsigned int cur_tx; struct sk_buff *skb; + /* + * The race with rhine_start_tx does not matter here as long as the + * driver enforces a value of cur_tx that was relevant when the + * packet was scheduled to the network chipset. + * Executive summary: smp_rmb() balances smp_wmb() in rhine_start_tx. + */ + smp_rmb(); + cur_tx = rp->cur_tx; /* find and cleanup dirty tx descriptors */ - while (rp->dirty_tx != rp->cur_tx) { - txstatus = le32_to_cpu(rp->tx_ring[entry].tx_status); + while (dirty_tx != cur_tx) { + unsigned int entry = dirty_tx % TX_RING_SIZE; + u32 txstatus = le32_to_cpu(rp->tx_ring[entry].tx_status); + netif_dbg(rp, tx_done, dev, "Tx scavenge %d status %08x\n", entry, txstatus); if (txstatus & DescOwn) @@ -1921,12 +1998,23 @@ static void rhine_tx(struct net_device *dev) pkts_compl++; dev_consume_skb_any(skb); rp->tx_skbuff[entry] = NULL; - entry = (++rp->dirty_tx) % TX_RING_SIZE; + dirty_tx++; } + rp->dirty_tx = dirty_tx; + /* Pity we can't rely on the nearby BQL completion implicit barrier. */ + smp_wmb(); + netdev_completed_queue(dev, pkts_compl, bytes_compl); - if ((rp->cur_tx - rp->dirty_tx) < TX_QUEUE_LEN - 4) + + /* cur_tx may be optimistically out-of-sync. See rhine_start_tx. */ + if (!rhine_tx_queue_full(rp) && netif_queue_stopped(dev)) { netif_wake_queue(dev); + smp_rmb(); + /* Rejuvenate. */ + if (rhine_tx_queue_full(rp)) + netif_stop_queue(dev); + } } /** @@ -1944,22 +2032,33 @@ static inline u16 rhine_get_vlan_tci(struct sk_buff *skb, int data_size) return be16_to_cpup((__be16 *)trailer); } +static inline void rhine_rx_vlan_tag(struct sk_buff *skb, struct rx_desc *desc, + int data_size) +{ + dma_rmb(); + if (unlikely(desc->desc_length & cpu_to_le32(DescTag))) { + u16 vlan_tci; + + vlan_tci = rhine_get_vlan_tci(skb, data_size); + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci); + } +} + /* Process up to limit frames from receive ring */ static int rhine_rx(struct net_device *dev, int limit) { struct rhine_private *rp = netdev_priv(dev); struct device *hwdev = dev->dev.parent; - int count; int entry = rp->cur_rx % RX_RING_SIZE; + int count; netif_dbg(rp, rx_status, dev, "%s(), entry %d status %08x\n", __func__, - entry, le32_to_cpu(rp->rx_head_desc->rx_status)); + entry, le32_to_cpu(rp->rx_ring[entry].rx_status)); /* If EOP is set on the next entry, it's a new packet. Send it up. */ for (count = 0; count < limit; ++count) { - struct rx_desc *desc = rp->rx_head_desc; + struct rx_desc *desc = rp->rx_ring + entry; u32 desc_status = le32_to_cpu(desc->rx_status); - u32 desc_length = le32_to_cpu(desc->desc_length); int data_size = desc_status >> 16; if (desc_status & DescOwn) @@ -1975,10 +2074,6 @@ static int rhine_rx(struct net_device *dev, int limit) "entry %#x length %d status %08x!\n", entry, data_size, desc_status); - netdev_warn(dev, - "Oversized Ethernet frame %p vs %p\n", - rp->rx_head_desc, - &rp->rx_ring[entry]); dev->stats.rx_length_errors++; } else if (desc_status & RxErr) { /* There was a error. */ @@ -2000,16 +2095,17 @@ static int rhine_rx(struct net_device *dev, int limit) } } } else { - struct sk_buff *skb = NULL; /* Length should omit the CRC */ int pkt_len = data_size - 4; - u16 vlan_tci = 0; + struct sk_buff *skb; /* Check if the packet is long enough to accept without copying to a minimally-sized skbuff. */ - if (pkt_len < rx_copybreak) + if (pkt_len < rx_copybreak) { skb = netdev_alloc_skb_ip_align(dev, pkt_len); - if (skb) { + if (unlikely(!skb)) + goto drop; + dma_sync_single_for_cpu(hwdev, rp->rx_skbuff_dma[entry], rp->rx_buf_sz, @@ -2018,32 +2114,31 @@ static int rhine_rx(struct net_device *dev, int limit) skb_copy_to_linear_data(skb, rp->rx_skbuff[entry]->data, pkt_len); - skb_put(skb, pkt_len); + dma_sync_single_for_device(hwdev, rp->rx_skbuff_dma[entry], rp->rx_buf_sz, DMA_FROM_DEVICE); } else { + struct rhine_skb_dma sd; + + if (unlikely(rhine_skb_dma_init(dev, &sd) < 0)) + goto drop; + skb = rp->rx_skbuff[entry]; - if (skb == NULL) { - netdev_err(dev, "Inconsistent Rx descriptor chain\n"); - break; - } - rp->rx_skbuff[entry] = NULL; - skb_put(skb, pkt_len); + dma_unmap_single(hwdev, rp->rx_skbuff_dma[entry], rp->rx_buf_sz, DMA_FROM_DEVICE); + rhine_skb_dma_nic_store(rp, &sd, entry); } - if (unlikely(desc_length & DescTag)) - vlan_tci = rhine_get_vlan_tci(skb, data_size); - + skb_put(skb, pkt_len); skb->protocol = eth_type_trans(skb, dev); - if (unlikely(desc_length & DescTag)) - __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci); + rhine_rx_vlan_tag(skb, desc, data_size); + netif_receive_skb(skb); u64_stats_update_begin(&rp->rx_stats.syncp); @@ -2051,35 +2146,16 @@ static int rhine_rx(struct net_device *dev, int limit) rp->rx_stats.packets++; u64_stats_update_end(&rp->rx_stats.syncp); } +give_descriptor_to_nic: + desc->rx_status = cpu_to_le32(DescOwn); entry = (++rp->cur_rx) % RX_RING_SIZE; - rp->rx_head_desc = &rp->rx_ring[entry]; - } - - /* Refill the Rx ring buffers. */ - for (; rp->cur_rx - rp->dirty_rx > 0; rp->dirty_rx++) { - struct sk_buff *skb; - entry = rp->dirty_rx % RX_RING_SIZE; - if (rp->rx_skbuff[entry] == NULL) { - skb = netdev_alloc_skb(dev, rp->rx_buf_sz); - rp->rx_skbuff[entry] = skb; - if (skb == NULL) - break; /* Better luck next round. */ - rp->rx_skbuff_dma[entry] = - dma_map_single(hwdev, skb->data, - rp->rx_buf_sz, - DMA_FROM_DEVICE); - if (dma_mapping_error(hwdev, - rp->rx_skbuff_dma[entry])) { - dev_kfree_skb(skb); - rp->rx_skbuff_dma[entry] = 0; - break; - } - rp->rx_ring[entry].addr = cpu_to_le32(rp->rx_skbuff_dma[entry]); - } - rp->rx_ring[entry].rx_status = cpu_to_le32(DescOwn); } return count; + +drop: + dev->stats.rx_dropped++; + goto give_descriptor_to_nic; } static void rhine_restart_tx(struct net_device *dev) { @@ -2484,9 +2560,8 @@ static int rhine_resume(struct device *device) enable_mmio(rp->pioaddr, rp->quirks); rhine_power_init(dev); free_tbufs(dev); - free_rbufs(dev); alloc_tbufs(dev); - alloc_rbufs(dev); + rhine_reset_rbufs(rp); rhine_task_enable(rp); spin_lock_bh(&rp->lock); init_registers(dev); diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index af2694dc6f90..3b99a4df71f8 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -688,10 +688,8 @@ static int temac_start_xmit(struct sk_buff *skb, struct net_device *ndev) cur_p = &lp->tx_bd_v[lp->tx_bd_tail]; if (temac_check_tx_bd_space(lp, num_frag)) { - if (!netif_queue_stopped(ndev)) { + if (!netif_queue_stopped(ndev)) netif_stop_queue(ndev); - return NETDEV_TX_BUSY; - } return NETDEV_TX_BUSY; } diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h index 4c9b4fa1d3c1..7cb9abac95c8 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h @@ -11,16 +11,16 @@ #include <linux/netdevice.h> #include <linux/spinlock.h> #include <linux/interrupt.h> +#include <linux/if_vlan.h> /* Packet size info */ #define XAE_HDR_SIZE 14 /* Size of Ethernet header */ -#define XAE_HDR_VLAN_SIZE 18 /* Size of an Ethernet hdr + VLAN */ #define XAE_TRL_SIZE 4 /* Size of Ethernet trailer (FCS) */ #define XAE_MTU 1500 /* Max MTU of an Ethernet frame */ #define XAE_JUMBO_MTU 9000 /* Max MTU of a jumbo Eth. frame */ #define XAE_MAX_FRAME_SIZE (XAE_MTU + XAE_HDR_SIZE + XAE_TRL_SIZE) -#define XAE_MAX_VLAN_FRAME_SIZE (XAE_MTU + XAE_HDR_VLAN_SIZE + XAE_TRL_SIZE) +#define XAE_MAX_VLAN_FRAME_SIZE (XAE_MTU + VLAN_ETH_HLEN + XAE_TRL_SIZE) #define XAE_MAX_JUMBO_FRAME_SIZE (XAE_JUMBO_MTU + XAE_HDR_SIZE + XAE_TRL_SIZE) /* Configuration options */ @@ -38,18 +38,21 @@ #define XAE_OPTION_FLOW_CONTROL (1 << 4) /* Strip FCS and PAD from incoming frames. Note: PAD from VLAN frames is not - * stripped. Default: disabled (set) */ + * stripped. Default: disabled (set) + */ #define XAE_OPTION_FCS_STRIP (1 << 5) /* Generate FCS field and add PAD automatically for outgoing frames. - * Default: enabled (set) */ + * Default: enabled (set) + */ #define XAE_OPTION_FCS_INSERT (1 << 6) /* Enable Length/Type error checking for incoming frames. When this option is * set, the MAC will filter frames that have a mismatched type/length field * and if XAE_OPTION_REPORT_RXERR is set, the user is notified when these * types of frames are encountered. When this option is cleared, the MAC will - * allow these types of frames to be received. Default: enabled (set) */ + * allow these types of frames to be received. Default: enabled (set) + */ #define XAE_OPTION_LENTYPE_ERR (1 << 7) /* Enable the transmitter. Default: enabled (set) */ @@ -159,12 +162,12 @@ #define XAE_MDIO_MWD_OFFSET 0x00000508 /* MII Management Write Data */ #define XAE_MDIO_MRD_OFFSET 0x0000050C /* MII Management Read Data */ #define XAE_MDIO_MIS_OFFSET 0x00000600 /* MII Management Interrupt Status */ -#define XAE_MDIO_MIP_OFFSET 0x00000620 /* MII Mgmt Interrupt Pending - * register offset */ -#define XAE_MDIO_MIE_OFFSET 0x00000640 /* MII Management Interrupt Enable - * register offset */ -#define XAE_MDIO_MIC_OFFSET 0x00000660 /* MII Management Interrupt Clear - * register offset. */ +/* MII Mgmt Interrupt Pending register offset */ +#define XAE_MDIO_MIP_OFFSET 0x00000620 +/* MII Management Interrupt Enable register offset */ +#define XAE_MDIO_MIE_OFFSET 0x00000640 +/* MII Management Interrupt Clear register offset. */ +#define XAE_MDIO_MIC_OFFSET 0x00000660 #define XAE_UAW0_OFFSET 0x00000700 /* Unicast address word 0 */ #define XAE_UAW1_OFFSET 0x00000704 /* Unicast address word 1 */ #define XAE_FMI_OFFSET 0x00000708 /* Filter Mask Index */ @@ -176,18 +179,17 @@ #define XAE_MCAST_TABLE_OFFSET 0x00020000 /* Multicast table address */ /* Bit Masks for Axi Ethernet RAF register */ -#define XAE_RAF_MCSTREJ_MASK 0x00000002 /* Reject receive multicast - * destination address */ -#define XAE_RAF_BCSTREJ_MASK 0x00000004 /* Reject receive broadcast - * destination address */ +/* Reject receive multicast destination address */ +#define XAE_RAF_MCSTREJ_MASK 0x00000002 +/* Reject receive broadcast destination address */ +#define XAE_RAF_BCSTREJ_MASK 0x00000004 #define XAE_RAF_TXVTAGMODE_MASK 0x00000018 /* Tx VLAN TAG mode */ #define XAE_RAF_RXVTAGMODE_MASK 0x00000060 /* Rx VLAN TAG mode */ #define XAE_RAF_TXVSTRPMODE_MASK 0x00000180 /* Tx VLAN STRIP mode */ #define XAE_RAF_RXVSTRPMODE_MASK 0x00000600 /* Rx VLAN STRIP mode */ #define XAE_RAF_NEWFNCENBL_MASK 0x00000800 /* New function mode */ -#define XAE_RAF_EMULTIFLTRENBL_MASK 0x00001000 /* Exteneded Multicast - * Filtering mode - */ +/* Exteneded Multicast Filtering mode */ +#define XAE_RAF_EMULTIFLTRENBL_MASK 0x00001000 #define XAE_RAF_STATSRST_MASK 0x00002000 /* Stats. Counter Reset */ #define XAE_RAF_RXBADFRMEN_MASK 0x00004000 /* Recv Bad Frame Enable */ #define XAE_RAF_TXVTAGMODE_SHIFT 3 /* Tx Tag mode shift bits */ @@ -197,15 +199,16 @@ /* Bit Masks for Axi Ethernet TPF and IFGP registers */ #define XAE_TPF_TPFV_MASK 0x0000FFFF /* Tx pause frame value */ -#define XAE_IFGP0_IFGP_MASK 0x0000007F /* Transmit inter-frame - * gap adjustment value */ +/* Transmit inter-frame gap adjustment value */ +#define XAE_IFGP0_IFGP_MASK 0x0000007F /* Bit Masks for Axi Ethernet IS, IE and IP registers, Same masks apply - * for all 3 registers. */ -#define XAE_INT_HARDACSCMPLT_MASK 0x00000001 /* Hard register access - * complete */ -#define XAE_INT_AUTONEG_MASK 0x00000002 /* Auto negotiation - * complete */ + * for all 3 registers. + */ +/* Hard register access complete */ +#define XAE_INT_HARDACSCMPLT_MASK 0x00000001 +/* Auto negotiation complete */ +#define XAE_INT_AUTONEG_MASK 0x00000002 #define XAE_INT_RXCMPIT_MASK 0x00000004 /* Rx complete */ #define XAE_INT_RXRJECT_MASK 0x00000008 /* Rx frame rejected */ #define XAE_INT_RXFIFOOVR_MASK 0x00000010 /* Rx fifo overrun */ @@ -215,10 +218,9 @@ #define XAE_INT_PHYRSTCMPLT_MASK 0x00000100 /* Phy Reset complete */ #define XAE_INT_ALL_MASK 0x0000003F /* All the ints */ +/* INT bits that indicate receive errors */ #define XAE_INT_RECV_ERROR_MASK \ - (XAE_INT_RXRJECT_MASK | XAE_INT_RXFIFOOVR_MASK) /* INT bits that - * indicate receive - * errors */ + (XAE_INT_RXRJECT_MASK | XAE_INT_RXFIFOOVR_MASK) /* Bit masks for Axi Ethernet VLAN TPID Word 0 register */ #define XAE_TPID_0_MASK 0x0000FFFF /* TPID 0 */ @@ -231,27 +233,28 @@ /* Bit masks for Axi Ethernet RCW1 register */ #define XAE_RCW1_RST_MASK 0x80000000 /* Reset */ #define XAE_RCW1_JUM_MASK 0x40000000 /* Jumbo frame enable */ -#define XAE_RCW1_FCS_MASK 0x20000000 /* In-Band FCS enable - * (FCS not stripped) */ +/* In-Band FCS enable (FCS not stripped) */ +#define XAE_RCW1_FCS_MASK 0x20000000 #define XAE_RCW1_RX_MASK 0x10000000 /* Receiver enable */ #define XAE_RCW1_VLAN_MASK 0x08000000 /* VLAN frame enable */ -#define XAE_RCW1_LT_DIS_MASK 0x02000000 /* Length/type field valid check - * disable */ -#define XAE_RCW1_CL_DIS_MASK 0x01000000 /* Control frame Length check - * disable */ -#define XAE_RCW1_PAUSEADDR_MASK 0x0000FFFF /* Pause frame source address - * bits [47:32]. Bits [31:0] are - * stored in register RCW0 */ +/* Length/type field valid check disable */ +#define XAE_RCW1_LT_DIS_MASK 0x02000000 +/* Control frame Length check disable */ +#define XAE_RCW1_CL_DIS_MASK 0x01000000 +/* Pause frame source address bits [47:32]. Bits [31:0] are + * stored in register RCW0 + */ +#define XAE_RCW1_PAUSEADDR_MASK 0x0000FFFF /* Bit masks for Axi Ethernet TC register */ #define XAE_TC_RST_MASK 0x80000000 /* Reset */ #define XAE_TC_JUM_MASK 0x40000000 /* Jumbo frame enable */ -#define XAE_TC_FCS_MASK 0x20000000 /* In-Band FCS enable - * (FCS not generated) */ +/* In-Band FCS enable (FCS not generated) */ +#define XAE_TC_FCS_MASK 0x20000000 #define XAE_TC_TX_MASK 0x10000000 /* Transmitter enable */ #define XAE_TC_VLAN_MASK 0x08000000 /* VLAN frame enable */ -#define XAE_TC_IFG_MASK 0x02000000 /* Inter-frame gap adjustment - * enable */ +/* Inter-frame gap adjustment enable */ +#define XAE_TC_IFG_MASK 0x02000000 /* Bit masks for Axi Ethernet FCC register */ #define XAE_FCC_FCRX_MASK 0x20000000 /* Rx flow control enable */ @@ -301,10 +304,10 @@ #define XAE_MDIO_INT_MIIM_RDY_MASK 0x00000001 /* MIIM Interrupt */ /* Bit masks for Axi Ethernet UAW1 register */ -#define XAE_UAW1_UNICASTADDR_MASK 0x0000FFFF /* Station address bits - * [47:32]; Station address - * bits [31:0] are stored in - * register UAW0 */ +/* Station address bits [47:32]; Station address + * bits [31:0] are stored in register UAW0 + */ +#define XAE_UAW1_UNICASTADDR_MASK 0x0000FFFF /* Bit masks for Axi Ethernet FMI register */ #define XAE_FMI_PM_MASK 0x80000000 /* Promis. mode enable */ @@ -320,8 +323,8 @@ #define XAE_PHY_TYPE_SGMII 4 #define XAE_PHY_TYPE_1000BASE_X 5 -#define XAE_MULTICAST_CAM_TABLE_NUM 4 /* Total number of entries in the - * hardware multicast table. */ + /* Total number of entries in the hardware multicast table. */ +#define XAE_MULTICAST_CAM_TABLE_NUM 4 /* Axi Ethernet Synthesis features */ #define XAE_FEATURE_PARTIAL_RX_CSUM (1 << 0) @@ -407,8 +410,11 @@ struct axidma_bd { * Txed/Rxed in the existing hardware. If jumbo option is * supported, the maximum frame size would be 9k. Else it is * 1522 bytes (assuming support for basic VLAN) - * @jumbo_support: Stores hardware configuration for jumbo support. If hardware - * can handle jumbo packets, this entry will be 1, else 0. + * @rxmem: Stores rx memory size for jumbo frame handling. + * @csum_offload_on_tx_path: Stores the checksum selection on TX side. + * @csum_offload_on_rx_path: Stores the checksum selection on RX side. + * @coalesce_count_rx: Store the irq coalesce on RX side. + * @coalesce_count_tx: Store the irq coalesce on TX side. */ struct axienet_local { struct net_device *ndev; @@ -446,7 +452,7 @@ struct axienet_local { u32 rx_bd_ci; u32 max_frm_size; - u32 jumbo_support; + u32 rxmem; int csum_offload_on_tx_path; int csum_offload_on_rx_path; @@ -472,7 +478,7 @@ struct axienet_option { * @lp: Pointer to axienet local structure * @offset: Address offset from the base address of Axi Ethernet core * - * returns: The contents of the Axi Ethernet register + * Return: The contents of the Axi Ethernet register * * This function returns the contents of the corresponding register. */ diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 28b7e7d9c272..4208dd7ef101 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -117,7 +117,7 @@ static struct axienet_option axienet_options[] = { * @lp: Pointer to axienet local structure * @reg: Address offset from the base address of the Axi DMA core * - * returns: The contents of the Axi DMA register + * Return: The contents of the Axi DMA register * * This function returns the contents of the corresponding Axi DMA register. */ @@ -179,8 +179,7 @@ static void axienet_dma_bd_release(struct net_device *ndev) * axienet_dma_bd_init - Setup buffer descriptor rings for Axi DMA * @ndev: Pointer to the net_device structure * - * returns: 0, on success - * -ENOMEM, on failure + * Return: 0, on success -ENOMEM, on failure * * This function is called to initialize the Rx and Tx DMA descriptor * rings. This initializes the descriptors with required default values @@ -198,9 +197,7 @@ static int axienet_dma_bd_init(struct net_device *ndev) lp->tx_bd_tail = 0; lp->rx_bd_ci = 0; - /* - * Allocate the Tx and Rx buffer descriptors. - */ + /* Allocate the Tx and Rx buffer descriptors. */ lp->tx_bd_v = dma_zalloc_coherent(ndev->dev.parent, sizeof(*lp->tx_bd_v) * TX_BD_NUM, &lp->tx_bd_p, GFP_KERNEL); @@ -263,7 +260,8 @@ static int axienet_dma_bd_init(struct net_device *ndev) axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, cr); /* Populate the tail pointer and bring the Rx Axi DMA engine out of - * halted state. This will make the Rx side ready for reception.*/ + * halted state. This will make the Rx side ready for reception. + */ axienet_dma_out32(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_RX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, @@ -273,7 +271,8 @@ static int axienet_dma_bd_init(struct net_device *ndev) /* Write to the RS (Run-stop) bit in the Tx channel control register. * Tx channel is now ready to run. But only after we write to the - * tail pointer register that the Tx channel will start transmitting */ + * tail pointer register that the Tx channel will start transmitting. + */ axienet_dma_out32(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, @@ -320,7 +319,7 @@ static void axienet_set_mac_address(struct net_device *ndev, void *address) * @ndev: Pointer to the net_device structure * @p: 6 byte Address to be written as MAC address * - * returns: 0 for all conditions. Presently, there is no failure case. + * Return: 0 for all conditions. Presently, there is no failure case. * * This function is called to initialize the MAC address of the Axi Ethernet * core. It calls the core specific axienet_set_mac_address. This is the @@ -354,7 +353,8 @@ static void axienet_set_multicast_list(struct net_device *ndev) netdev_mc_count(ndev) > XAE_MULTICAST_CAM_TABLE_NUM) { /* We must make the kernel realize we had to move into * promiscuous mode. If it was a promiscuous mode request - * the flag is already set. If not we set it. */ + * the flag is already set. If not we set it. + */ ndev->flags |= IFF_PROMISC; reg = axienet_ior(lp, XAE_FMI_OFFSET); reg |= XAE_FMI_PM_MASK; @@ -438,14 +438,15 @@ static void __axienet_device_reset(struct axienet_local *lp, /* Reset Axi DMA. This would reset Axi Ethernet core as well. The reset * process of Axi DMA takes a while to complete as all pending * commands/transfers will be flushed or completed during this - * reset process. */ + * reset process. + */ axienet_dma_out32(lp, offset, XAXIDMA_CR_RESET_MASK); timeout = DELAY_OF_ONE_MILLISEC; while (axienet_dma_in32(lp, offset) & XAXIDMA_CR_RESET_MASK) { udelay(1); if (--timeout == 0) { - dev_err(dev, "axienet_device_reset DMA " - "reset timeout!\n"); + netdev_err(lp->ndev, "%s: DMA reset timeout!\n", + __func__); break; } } @@ -471,19 +472,21 @@ static void axienet_device_reset(struct net_device *ndev) __axienet_device_reset(lp, &ndev->dev, XAXIDMA_RX_CR_OFFSET); lp->max_frm_size = XAE_MAX_VLAN_FRAME_SIZE; + lp->options |= XAE_OPTION_VLAN; lp->options &= (~XAE_OPTION_JUMBO); if ((ndev->mtu > XAE_MTU) && - (ndev->mtu <= XAE_JUMBO_MTU) && - (lp->jumbo_support)) { - lp->max_frm_size = ndev->mtu + XAE_HDR_VLAN_SIZE + - XAE_TRL_SIZE; - lp->options |= XAE_OPTION_JUMBO; + (ndev->mtu <= XAE_JUMBO_MTU)) { + lp->max_frm_size = ndev->mtu + VLAN_ETH_HLEN + + XAE_TRL_SIZE; + + if (lp->max_frm_size <= lp->rxmem) + lp->options |= XAE_OPTION_JUMBO; } if (axienet_dma_bd_init(ndev)) { - dev_err(&ndev->dev, "axienet_device_reset descriptor " - "allocation failed\n"); + netdev_err(ndev, "%s: descriptor allocation failed\n", + __func__); } axienet_status = axienet_ior(lp, XAE_RCW1_OFFSET); @@ -497,7 +500,8 @@ static void axienet_device_reset(struct net_device *ndev) axienet_iow(lp, XAE_FCC_OFFSET, XAE_FCC_FCRX_MASK); /* Sync default options with HW but leave receiver and - * transmitter disabled.*/ + * transmitter disabled. + */ axienet_setoptions(ndev, lp->options & ~(XAE_OPTION_TXEN | XAE_OPTION_RXEN)); axienet_set_mac_address(ndev, NULL); @@ -558,8 +562,8 @@ static void axienet_adjust_link(struct net_device *ndev) lp->last_link = link_state; phy_print_status(phy); } else { - dev_err(&ndev->dev, "Error setting Axi Ethernet " - "mac speed\n"); + netdev_err(ndev, + "Error setting Axi Ethernet mac speed\n"); } } } @@ -617,7 +621,7 @@ static void axienet_start_xmit_done(struct net_device *ndev) * @lp: Pointer to the axienet_local structure * @num_frag: The number of BDs to check for * - * returns: 0, on success + * Return: 0, on success * NETDEV_TX_BUSY, if any of the descriptors are not free * * This function is invoked before BDs are allocated and transmission starts. @@ -640,7 +644,7 @@ static inline int axienet_check_tx_bd_space(struct axienet_local *lp, * @skb: sk_buff pointer that contains data to be Txed. * @ndev: Pointer to net_device structure. * - * returns: NETDEV_TX_OK, on success + * Return: NETDEV_TX_OK, on success * NETDEV_TX_BUSY, if any of the descriptors are not free * * This function is invoked from upper layers to initiate transmission. The @@ -726,15 +730,15 @@ static void axienet_recv(struct net_device *ndev) u32 csumstatus; u32 size = 0; u32 packets = 0; - dma_addr_t tail_p; + dma_addr_t tail_p = 0; struct axienet_local *lp = netdev_priv(ndev); struct sk_buff *skb, *new_skb; struct axidma_bd *cur_p; - tail_p = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_ci; cur_p = &lp->rx_bd_v[lp->rx_bd_ci]; while ((cur_p->status & XAXIDMA_BD_STS_COMPLETE_MASK)) { + tail_p = lp->rx_bd_p + sizeof(*lp->rx_bd_v) * lp->rx_bd_ci; skb = (struct sk_buff *) (cur_p->sw_id_offset); length = cur_p->app4 & 0x0000FFFF; @@ -786,7 +790,8 @@ static void axienet_recv(struct net_device *ndev) ndev->stats.rx_packets += packets; ndev->stats.rx_bytes += size; - axienet_dma_out32(lp, XAXIDMA_RX_TDESC_OFFSET, tail_p); + if (tail_p) + axienet_dma_out32(lp, XAXIDMA_RX_TDESC_OFFSET, tail_p); } /** @@ -794,7 +799,7 @@ static void axienet_recv(struct net_device *ndev) * @irq: irq number * @_ndev: net_device pointer * - * returns: IRQ_HANDLED for all cases. + * Return: IRQ_HANDLED for all cases. * * This is the Axi DMA Tx done Isr. It invokes "axienet_start_xmit_done" * to complete the BD processing. @@ -808,6 +813,7 @@ static irqreturn_t axienet_tx_irq(int irq, void *_ndev) status = axienet_dma_in32(lp, XAXIDMA_TX_SR_OFFSET); if (status & (XAXIDMA_IRQ_IOC_MASK | XAXIDMA_IRQ_DELAY_MASK)) { + axienet_dma_out32(lp, XAXIDMA_TX_SR_OFFSET, status); axienet_start_xmit_done(lp->ndev); goto out; } @@ -831,9 +837,9 @@ static irqreturn_t axienet_tx_irq(int irq, void *_ndev) axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr); tasklet_schedule(&lp->dma_err_tasklet); + axienet_dma_out32(lp, XAXIDMA_TX_SR_OFFSET, status); } out: - axienet_dma_out32(lp, XAXIDMA_TX_SR_OFFSET, status); return IRQ_HANDLED; } @@ -842,7 +848,7 @@ out: * @irq: irq number * @_ndev: net_device pointer * - * returns: IRQ_HANDLED for all cases. + * Return: IRQ_HANDLED for all cases. * * This is the Axi DMA Rx Isr. It invokes "axienet_recv" to complete the BD * processing. @@ -856,6 +862,7 @@ static irqreturn_t axienet_rx_irq(int irq, void *_ndev) status = axienet_dma_in32(lp, XAXIDMA_RX_SR_OFFSET); if (status & (XAXIDMA_IRQ_IOC_MASK | XAXIDMA_IRQ_DELAY_MASK)) { + axienet_dma_out32(lp, XAXIDMA_RX_SR_OFFSET, status); axienet_recv(lp->ndev); goto out; } @@ -879,9 +886,9 @@ static irqreturn_t axienet_rx_irq(int irq, void *_ndev) axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, cr); tasklet_schedule(&lp->dma_err_tasklet); + axienet_dma_out32(lp, XAXIDMA_RX_SR_OFFSET, status); } out: - axienet_dma_out32(lp, XAXIDMA_RX_SR_OFFSET, status); return IRQ_HANDLED; } @@ -891,7 +898,7 @@ static void axienet_dma_err_handler(unsigned long data); * axienet_open - Driver open routine. * @ndev: Pointer to net_device structure * - * returns: 0, on success. + * Return: 0, on success. * -ENODEV, if PHY cannot be connected to * non-zero error value on failure * @@ -914,7 +921,8 @@ static int axienet_open(struct net_device *ndev) /* Disable the MDIO interface till Axi Ethernet Reset is completed. * When we do an Axi Ethernet reset, it resets the complete core * including the MDIO. If MDIO is not disabled when the reset - * process is started, MDIO will be broken afterwards. */ + * process is started, MDIO will be broken afterwards. + */ axienet_iow(lp, XAE_MDIO_MC_OFFSET, (mdio_mcreg & (~XAE_MDIO_MC_MDIOEN_MASK))); axienet_device_reset(ndev); @@ -925,14 +933,20 @@ static int axienet_open(struct net_device *ndev) return ret; if (lp->phy_node) { - lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, + if (lp->phy_type == XAE_PHY_TYPE_GMII) { + lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, axienet_adjust_link, 0, PHY_INTERFACE_MODE_GMII); - if (!lp->phy_dev) { - dev_err(lp->dev, "of_phy_connect() failed\n"); - return -ENODEV; + } else if (lp->phy_type == XAE_PHY_TYPE_RGMII_2_0) { + lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, + axienet_adjust_link, 0, + PHY_INTERFACE_MODE_RGMII_ID); } - phy_start(lp->phy_dev); + + if (!lp->phy_dev) + dev_err(lp->dev, "of_phy_connect() failed\n"); + else + phy_start(lp->phy_dev); } /* Enable tasklets for Axi DMA error handling */ @@ -965,7 +979,7 @@ err_tx_irq: * axienet_stop - Driver stop routine. * @ndev: Pointer to net_device structure * - * returns: 0, on success. + * Return: 0, on success. * * This is the driver stop routine. It calls phy_disconnect to stop the PHY * device. It also removes the interrupt handlers and disables the interrupts. @@ -1005,7 +1019,7 @@ static int axienet_stop(struct net_device *ndev) * @ndev: Pointer to net_device structure * @new_mtu: New mtu value to be applied * - * returns: Always returns 0 (success). + * Return: Always returns 0 (success). * * This is the change mtu driver routine. It checks if the Axi Ethernet * hardware supports jumbo frames before changing the mtu. This can be @@ -1017,15 +1031,15 @@ static int axienet_change_mtu(struct net_device *ndev, int new_mtu) if (netif_running(ndev)) return -EBUSY; - if (lp->jumbo_support) { - if ((new_mtu > XAE_JUMBO_MTU) || (new_mtu < 64)) - return -EINVAL; - ndev->mtu = new_mtu; - } else { - if ((new_mtu > XAE_MTU) || (new_mtu < 64)) - return -EINVAL; - ndev->mtu = new_mtu; - } + + if ((new_mtu + VLAN_ETH_HLEN + + XAE_TRL_SIZE) > lp->rxmem) + return -EINVAL; + + if ((new_mtu > XAE_JUMBO_MTU) || (new_mtu < 64)) + return -EINVAL; + + ndev->mtu = new_mtu; return 0; } @@ -1072,6 +1086,8 @@ static const struct net_device_ops axienet_netdev_ops = { * not be found, the function returns -ENODEV. This function calls the * relevant PHY ethtool API to get the PHY settings. * Issue "ethtool ethX" under linux prompt to execute this function. + * + * Return: 0 on success, -ENODEV if PHY doesn't exist */ static int axienet_ethtools_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) @@ -1093,6 +1109,8 @@ static int axienet_ethtools_get_settings(struct net_device *ndev, * relevant PHY ethtool API to set the PHY. * Issue e.g. "ethtool -s ethX speed 1000" under linux prompt to execute this * function. + * + * Return: 0 on success, -ENODEV if PHY doesn't exist */ static int axienet_ethtools_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) @@ -1127,6 +1145,8 @@ static void axienet_ethtools_get_drvinfo(struct net_device *ndev, * * This implements ethtool command for getting the total register length * information. + * + * Return: the total regs length */ static int axienet_ethtools_get_regs_len(struct net_device *ndev) { @@ -1213,11 +1233,13 @@ axienet_ethtools_get_pauseparam(struct net_device *ndev, * axienet_ethtools_set_pauseparam - Set device pause parameter(flow control) * settings. * @ndev: Pointer to net_device structure - * @epauseparam:Pointer to ethtool_pauseparam structure + * @epauseparm:Pointer to ethtool_pauseparam structure * * This implements ethtool command for enabling flow control on Rx and Tx * paths. Issue "ethtool -A ethX tx on|off" under linux prompt to execute this * function. + * + * Return: 0 on success, -EFAULT if device is running */ static int axienet_ethtools_set_pauseparam(struct net_device *ndev, @@ -1227,8 +1249,8 @@ axienet_ethtools_set_pauseparam(struct net_device *ndev, struct axienet_local *lp = netdev_priv(ndev); if (netif_running(ndev)) { - printk(KERN_ERR "%s: Please stop netif before applying " - "configruation\n", ndev->name); + netdev_err(ndev, + "Please stop netif before applying configuration\n"); return -EFAULT; } @@ -1254,6 +1276,8 @@ axienet_ethtools_set_pauseparam(struct net_device *ndev, * This implements ethtool command for getting the DMA interrupt coalescing * count on Tx and Rx paths. Issue "ethtool -c ethX" under linux prompt to * execute this function. + * + * Return: 0 always */ static int axienet_ethtools_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *ecoalesce) @@ -1277,6 +1301,8 @@ static int axienet_ethtools_get_coalesce(struct net_device *ndev, * This implements ethtool command for setting the DMA interrupt coalescing * count on Tx and Rx paths. Issue "ethtool -C ethX rx-frames 5" under linux * prompt to execute this function. + * + * Return: 0, on success, Non-zero error value on failure. */ static int axienet_ethtools_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ecoalesce) @@ -1284,8 +1310,8 @@ static int axienet_ethtools_set_coalesce(struct net_device *ndev, struct axienet_local *lp = netdev_priv(ndev); if (netif_running(ndev)) { - printk(KERN_ERR "%s: Please stop netif before applying " - "configruation\n", ndev->name); + netdev_err(ndev, + "Please stop netif before applying configuration\n"); return -EFAULT; } @@ -1354,7 +1380,8 @@ static void axienet_dma_err_handler(unsigned long data) /* Disable the MDIO interface till Axi Ethernet Reset is completed. * When we do an Axi Ethernet reset, it resets the complete core * including the MDIO. So if MDIO is not disabled when the reset - * process is started, MDIO will be broken afterwards. */ + * process is started, MDIO will be broken afterwards. + */ axienet_iow(lp, XAE_MDIO_MC_OFFSET, (mdio_mcreg & ~XAE_MDIO_MC_MDIOEN_MASK)); @@ -1425,7 +1452,8 @@ static void axienet_dma_err_handler(unsigned long data) axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, cr); /* Populate the tail pointer and bring the Rx Axi DMA engine out of - * halted state. This will make the Rx side ready for reception.*/ + * halted state. This will make the Rx side ready for reception. + */ axienet_dma_out32(lp, XAXIDMA_RX_CDESC_OFFSET, lp->rx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_RX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, @@ -1435,7 +1463,8 @@ static void axienet_dma_err_handler(unsigned long data) /* Write to the RS (Run-stop) bit in the Tx channel control register. * Tx channel is now ready to run. But only after we write to the - * tail pointer register that the Tx channel will start transmitting */ + * tail pointer register that the Tx channel will start transmitting + */ axienet_dma_out32(lp, XAXIDMA_TX_CDESC_OFFSET, lp->tx_bd_p); cr = axienet_dma_in32(lp, XAXIDMA_TX_CR_OFFSET); axienet_dma_out32(lp, XAXIDMA_TX_CR_OFFSET, @@ -1451,7 +1480,8 @@ static void axienet_dma_err_handler(unsigned long data) axienet_iow(lp, XAE_FCC_OFFSET, XAE_FCC_FCRX_MASK); /* Sync default options with HW but leave receiver and - * transmitter disabled.*/ + * transmitter disabled. + */ axienet_setoptions(ndev, lp->options & ~(XAE_OPTION_TXEN | XAE_OPTION_RXEN)); axienet_set_mac_address(ndev, NULL); @@ -1460,11 +1490,10 @@ static void axienet_dma_err_handler(unsigned long data) } /** - * axienet_of_probe - Axi Ethernet probe function. - * @op: Pointer to platform device structure. - * @match: Pointer to device id structure + * axienet_probe - Axi Ethernet probe function. + * @pdev: Pointer to platform device structure. * - * returns: 0, on success + * Return: 0, on success * Non-zero error value on failure. * * This is the probe routine for Axi Ethernet driver. This is called before @@ -1472,22 +1501,23 @@ static void axienet_dma_err_handler(unsigned long data) * device. Parses through device tree and populates fields of * axienet_local. It registers the Ethernet device. */ -static int axienet_of_probe(struct platform_device *op) +static int axienet_probe(struct platform_device *pdev) { - __be32 *p; - int size, ret = 0; + int ret; struct device_node *np; struct axienet_local *lp; struct net_device *ndev; - const void *addr; + u8 mac_addr[6]; + struct resource *ethres, dmares; + u32 value; ndev = alloc_etherdev(sizeof(*lp)); if (!ndev) return -ENOMEM; - platform_set_drvdata(op, ndev); + platform_set_drvdata(pdev, ndev); - SET_NETDEV_DEV(ndev, &op->dev); + SET_NETDEV_DEV(ndev, &pdev->dev); ndev->flags &= ~IFF_MULTICAST; /* clear multicast */ ndev->features = NETIF_F_SG; ndev->netdev_ops = &axienet_netdev_ops; @@ -1495,21 +1525,23 @@ static int axienet_of_probe(struct platform_device *op) lp = netdev_priv(ndev); lp->ndev = ndev; - lp->dev = &op->dev; + lp->dev = &pdev->dev; lp->options = XAE_OPTION_DEFAULTS; /* Map device registers */ - lp->regs = of_iomap(op->dev.of_node, 0); + ethres = platform_get_resource(pdev, IORESOURCE_MEM, 0); + lp->regs = devm_ioremap_resource(&pdev->dev, ethres); if (!lp->regs) { - dev_err(&op->dev, "could not map Axi Ethernet regs.\n"); + dev_err(&pdev->dev, "could not map Axi Ethernet regs.\n"); ret = -ENOMEM; - goto nodev; + goto free_netdev; } + /* Setup checksum offload, but default to off if not specified */ lp->features = 0; - p = (__be32 *) of_get_property(op->dev.of_node, "xlnx,txcsum", NULL); - if (p) { - switch (be32_to_cpup(p)) { + ret = of_property_read_u32(pdev->dev.of_node, "xlnx,txcsum", &value); + if (!ret) { + switch (value) { case 1: lp->csum_offload_on_tx_path = XAE_FEATURE_PARTIAL_TX_CSUM; @@ -1528,9 +1560,9 @@ static int axienet_of_probe(struct platform_device *op) lp->csum_offload_on_tx_path = XAE_NO_CSUM_OFFLOAD; } } - p = (__be32 *) of_get_property(op->dev.of_node, "xlnx,rxcsum", NULL); - if (p) { - switch (be32_to_cpup(p)) { + ret = of_property_read_u32(pdev->dev.of_node, "xlnx,rxcsum", &value); + if (!ret) { + switch (value) { case 1: lp->csum_offload_on_rx_path = XAE_FEATURE_PARTIAL_RX_CSUM; @@ -1546,82 +1578,77 @@ static int axienet_of_probe(struct platform_device *op) } } /* For supporting jumbo frames, the Axi Ethernet hardware must have - * a larger Rx/Tx Memory. Typically, the size must be more than or - * equal to 16384 bytes, so that we can enable jumbo option and start - * supporting jumbo frames. Here we check for memory allocated for - * Rx/Tx in the hardware from the device-tree and accordingly set - * flags. */ - p = (__be32 *) of_get_property(op->dev.of_node, "xlnx,rxmem", NULL); - if (p) { - if ((be32_to_cpup(p)) >= 0x4000) - lp->jumbo_support = 1; - } - p = (__be32 *) of_get_property(op->dev.of_node, "xlnx,phy-type", NULL); - if (p) - lp->phy_type = be32_to_cpup(p); + * a larger Rx/Tx Memory. Typically, the size must be large so that + * we can enable jumbo option and start supporting jumbo frames. + * Here we check for memory allocated for Rx/Tx in the hardware from + * the device-tree and accordingly set flags. + */ + of_property_read_u32(pdev->dev.of_node, "xlnx,rxmem", &lp->rxmem); + of_property_read_u32(pdev->dev.of_node, "xlnx,phy-type", &lp->phy_type); /* Find the DMA node, map the DMA registers, and decode the DMA IRQs */ - np = of_parse_phandle(op->dev.of_node, "axistream-connected", 0); - if (!np) { - dev_err(&op->dev, "could not find DMA node\n"); - ret = -ENODEV; - goto err_iounmap; + np = of_parse_phandle(pdev->dev.of_node, "axistream-connected", 0); + if (IS_ERR(np)) { + dev_err(&pdev->dev, "could not find DMA node\n"); + ret = PTR_ERR(np); + goto free_netdev; } - lp->dma_regs = of_iomap(np, 0); - if (lp->dma_regs) { - dev_dbg(&op->dev, "MEM base: %p\n", lp->dma_regs); - } else { - dev_err(&op->dev, "unable to map DMA registers\n"); - of_node_put(np); + ret = of_address_to_resource(np, 0, &dmares); + if (ret) { + dev_err(&pdev->dev, "unable to get DMA resource\n"); + goto free_netdev; + } + lp->dma_regs = devm_ioremap_resource(&pdev->dev, &dmares); + if (!lp->dma_regs) { + dev_err(&pdev->dev, "could not map DMA regs\n"); + ret = -ENOMEM; + goto free_netdev; } lp->rx_irq = irq_of_parse_and_map(np, 1); lp->tx_irq = irq_of_parse_and_map(np, 0); of_node_put(np); if ((lp->rx_irq <= 0) || (lp->tx_irq <= 0)) { - dev_err(&op->dev, "could not determine irqs\n"); + dev_err(&pdev->dev, "could not determine irqs\n"); ret = -ENOMEM; - goto err_iounmap_2; + goto free_netdev; } /* Retrieve the MAC address */ - addr = of_get_property(op->dev.of_node, "local-mac-address", &size); - if ((!addr) || (size != 6)) { - dev_err(&op->dev, "could not find MAC address\n"); - ret = -ENODEV; - goto err_iounmap_2; + ret = of_property_read_u8_array(pdev->dev.of_node, + "local-mac-address", mac_addr, 6); + if (ret) { + dev_err(&pdev->dev, "could not find MAC address\n"); + goto free_netdev; } - axienet_set_mac_address(ndev, (void *) addr); + axienet_set_mac_address(ndev, (void *)mac_addr); lp->coalesce_count_rx = XAXIDMA_DFT_RX_THRESHOLD; lp->coalesce_count_tx = XAXIDMA_DFT_TX_THRESHOLD; - lp->phy_node = of_parse_phandle(op->dev.of_node, "phy-handle", 0); - ret = axienet_mdio_setup(lp, op->dev.of_node); - if (ret) - dev_warn(&op->dev, "error registering MDIO bus\n"); + lp->phy_node = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0); + if (lp->phy_node) { + ret = axienet_mdio_setup(lp, pdev->dev.of_node); + if (ret) + dev_warn(&pdev->dev, "error registering MDIO bus\n"); + } ret = register_netdev(lp->ndev); if (ret) { dev_err(lp->dev, "register_netdev() error (%i)\n", ret); - goto err_iounmap_2; + goto free_netdev; } return 0; -err_iounmap_2: - if (lp->dma_regs) - iounmap(lp->dma_regs); -err_iounmap: - iounmap(lp->regs); -nodev: +free_netdev: free_netdev(ndev); - ndev = NULL; + return ret; } -static int axienet_of_remove(struct platform_device *op) +static int axienet_remove(struct platform_device *pdev) { - struct net_device *ndev = platform_get_drvdata(op); + struct net_device *ndev = platform_get_drvdata(pdev); struct axienet_local *lp = netdev_priv(ndev); axienet_mdio_teardown(lp); @@ -1630,24 +1657,21 @@ static int axienet_of_remove(struct platform_device *op) of_node_put(lp->phy_node); lp->phy_node = NULL; - iounmap(lp->regs); - if (lp->dma_regs) - iounmap(lp->dma_regs); free_netdev(ndev); return 0; } -static struct platform_driver axienet_of_driver = { - .probe = axienet_of_probe, - .remove = axienet_of_remove, +static struct platform_driver axienet_driver = { + .probe = axienet_probe, + .remove = axienet_remove, .driver = { .name = "xilinx_axienet", .of_match_table = axienet_of_match, }, }; -module_platform_driver(axienet_of_driver); +module_platform_driver(axienet_driver); MODULE_DESCRIPTION("Xilinx Axi Ethernet driver"); MODULE_AUTHOR("Xilinx"); diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c index 3b67d60d4378..2a5a16834c01 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_mdio.c @@ -37,7 +37,7 @@ int axienet_mdio_wait_until_ready(struct axienet_local *lp) * @phy_id: Address of the PHY device * @reg: PHY register to read * - * returns: The register contents on success, -ETIMEDOUT on a timeout + * Return: The register contents on success, -ETIMEDOUT on a timeout * * Reads the contents of the requested register from the requested PHY * address by first writing the details into MCR register. After a while @@ -80,7 +80,7 @@ static int axienet_mdio_read(struct mii_bus *bus, int phy_id, int reg) * @reg: PHY register to write to * @val: Value to be written into the register * - * returns: 0 on success, -ETIMEDOUT on a timeout + * Return: 0 on success, -ETIMEDOUT on a timeout * * Writes the value to the requested register by first writing the value * into MWD register. The the MCR register is then appropriately setup @@ -119,7 +119,7 @@ static int axienet_mdio_write(struct mii_bus *bus, int phy_id, int reg, * @lp: Pointer to axienet local data structure. * @np: Pointer to device node * - * returns: 0 on success, -ETIMEDOUT on a timeout, -ENOMEM when + * Return: 0 on success, -ETIMEDOUT on a timeout, -ENOMEM when * mdiobus_alloc (to allocate memory for mii bus structure) fails. * * Sets up the MDIO interface by initializing the MDIO clock and enabling the @@ -161,19 +161,19 @@ int axienet_mdio_setup(struct axienet_local *lp, struct device_node *np) np1 = of_find_node_by_name(NULL, "cpu"); if (!np1) { - printk(KERN_WARNING "%s(): Could not find CPU device node.", - __func__); - printk(KERN_WARNING "Setting MDIO clock divisor to " - "default %d\n", DEFAULT_CLOCK_DIVISOR); + netdev_warn(lp->ndev, "Could not find CPU device node.\n"); + netdev_warn(lp->ndev, + "Setting MDIO clock divisor to default %d\n", + DEFAULT_CLOCK_DIVISOR); clk_div = DEFAULT_CLOCK_DIVISOR; goto issue; } property_p = (u32 *) of_get_property(np1, "clock-frequency", NULL); if (!property_p) { - printk(KERN_WARNING "%s(): Could not find CPU property: " - "clock-frequency.", __func__); - printk(KERN_WARNING "Setting MDIO clock divisor to " - "default %d\n", DEFAULT_CLOCK_DIVISOR); + netdev_warn(lp->ndev, "clock-frequency property not found.\n"); + netdev_warn(lp->ndev, + "Setting MDIO clock divisor to default %d\n", + DEFAULT_CLOCK_DIVISOR); clk_div = DEFAULT_CLOCK_DIVISOR; of_node_put(np1); goto issue; @@ -183,12 +183,14 @@ int axienet_mdio_setup(struct axienet_local *lp, struct device_node *np) clk_div = (host_clock / (MAX_MDIO_FREQ * 2)) - 1; /* If there is any remainder from the division of * fHOST / (MAX_MDIO_FREQ * 2), then we need to add - * 1 to the clock divisor or we will surely be above 2.5 MHz */ + * 1 to the clock divisor or we will surely be above 2.5 MHz + */ if (host_clock % (MAX_MDIO_FREQ * 2)) clk_div++; - printk(KERN_DEBUG "%s(): Setting MDIO clock divisor to %u based " - "on %u Hz host clock.\n", __func__, clk_div, host_clock); + netdev_dbg(lp->ndev, + "Setting MDIO clock divisor to %u/%u Hz host clock.\n", + clk_div, host_clock); of_node_put(np1); issue: |