diff options
Diffstat (limited to 'drivers/net/ethernet')
81 files changed, 4276 insertions, 1648 deletions
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..084a50a555de 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); @@ -1641,6 +1702,8 @@ 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..42a4b4a0bc14 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 diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index ec56a9b65dc3..2ef202d10948 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -662,7 +662,7 @@ 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/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 61aa570aad9a..e7c10b0addb5 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) @@ -782,7 +784,7 @@ 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); @@ -828,7 +830,7 @@ 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), @@ -1642,7 +1644,10 @@ 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) @@ -1651,8 +1656,13 @@ 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); @@ -1856,6 +1866,26 @@ 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; @@ -2132,7 +2162,7 @@ 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, @@ -2693,6 +2723,15 @@ 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 of_device_id macb_dt_ids[] = { { .compatible = "cdns,at32ap7000-macb" }, { .compatible = "cdns,at91sam9260-macb", .data = &at91sam9260_config }, @@ -2703,6 +2742,7 @@ 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}, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, macb_dt_ids); @@ -2771,6 +2811,10 @@ 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 eb7d76f7bf6a..7d4ef513df75 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 */ @@ -397,6 +398,7 @@ #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) \ @@ -514,6 +516,9 @@ 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 @@ -757,6 +762,7 @@ 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 { @@ -826,6 +832,9 @@ 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/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 524d11098c56..932ab3b72a4d 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -328,6 +328,17 @@ 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 +641,7 @@ 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; @@ -1055,7 +1061,7 @@ 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); @@ -1195,12 +1201,15 @@ 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); @@ -1215,6 +1224,7 @@ 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, @@ -1310,4 +1320,9 @@ 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_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index 10d82b51d7ef..401272a2691e 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -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; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 803d91beec6f..5aecf69efe56 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -856,23 +856,39 @@ 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->fn, 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 +901,15 @@ 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 +997,7 @@ 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 +1027,7 @@ 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 +1047,7 @@ 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 +1064,9 @@ 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; @@ -1398,7 +1420,7 @@ 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; } @@ -2432,6 +2454,7 @@ 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)) { @@ -4340,7 +4363,12 @@ 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 +4376,6 @@ 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; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h index 78ab4d406ce2..df34293f35e8 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 { diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 0d2eddab04ef..dd18fcb644f9 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. */ @@ -540,6 +530,10 @@ static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q) val = PIDX_T5_V(q->pend_cred / 8) | DBTYPE_F; val |= DBPRIO_F; + + /* 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 @@ -930,7 +924,10 @@ 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. @@ -1047,7 +1044,7 @@ nocsum: /* * unknown protocol, disable HW csum * and hope a bad packet is detected */ - return TXPKT_L4CSUM_DIS; + return TXPKT_L4CSUM_DIS_F; } } else { /* @@ -1064,14 +1061,15 @@ 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); + return TXPKT_CSUM_TYPE_V(csum_type) | + TXPKT_IPHDR_LEN_V(skb_network_header_len(skb)) | + TXPKT_ETHHDR_LEN_V(skb_network_offset(skb) - ETH_HLEN); 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 +1110,11 @@ 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 +1128,6 @@ 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 +1140,7 @@ 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 +1154,20 @@ 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 +1218,23 @@ 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); + cntrl = TXPKT_CSUM_TYPE_V(v6 ? TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | + TXPKT_IPHDR_LEN_V(l3hdr_len) | + TXPKT_ETHHDR_LEN_V(eth_xtra_len); q->tso++; q->tx_cso += ssi->gso_segs; } else { @@ -1238,23 +1243,24 @@ 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(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->fn)); cpl->pack = htons(0); cpl->len = htons(skb->len); cpl->ctrl1 = cpu_to_be64(cntrl); @@ -1964,7 +1970,7 @@ 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 +2017,19 @@ 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 +2064,7 @@ 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 +2072,7 @@ 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 +2096,7 @@ 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 +2143,7 @@ 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 +2153,16 @@ 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 +2210,7 @@ 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 +2286,7 @@ 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 +2307,16 @@ 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); } @@ -2437,9 +2393,12 @@ 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; @@ -2462,7 +2421,8 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, 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 +2431,19 @@ 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) { + /* 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, @@ -2481,12 +2452,18 @@ 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(FETCHBURSTMAX_512B_X)); c.fl0size = htons(flsz); c.fl0addr = cpu_to_be64(fl->addr); } @@ -2532,6 +2509,41 @@ 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->fn, 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: @@ -2595,14 +2607,15 @@ int t4_sge_alloc_eth_txq(struct adapter *adap, struct sge_eth_txq *txq, 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); @@ -2637,7 +2650,7 @@ 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; @@ -2649,14 +2662,15 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq, 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); @@ -2701,14 +2715,15 @@ int t4_sge_alloc_ofld_txq(struct adapter *adap, struct sge_ofld_txq *txq, 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); @@ -3023,7 +3038,11 @@ 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 +3086,14 @@ 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..c626252e51d4 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -3014,6 +3014,31 @@ int t4_config_glbl_rss(struct adapter *adapter, int mbox, unsigned int mode, 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) { @@ -3401,7 +3426,7 @@ 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 +3434,7 @@ 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 +3485,7 @@ 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, \ @@ -5373,6 +5398,28 @@ 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 = htonl(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 = htonl(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); + } + return 0; +} + int t4_port_init(struct adapter *adap, int mbox, int pf, int vf) { u8 addr[6]; @@ -5717,3 +5764,130 @@ 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..d90f8a03e378 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 +646,66 @@ 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 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_values.h b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h index 19b2dcf6acde..72ec1f91d29f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h @@ -61,6 +61,29 @@ #define SGE_TIMERREGS 6 #define TIMERREG_COUNTER0_X 0 +#define FETCHBURSTMIN_64B_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..16c6d67370ee 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -1123,6 +1123,7 @@ 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_diag { @@ -1377,6 +1378,7 @@ 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 +1401,7 @@ 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 +1592,7 @@ 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) 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..2e41d1541d73 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c @@ -1100,7 +1100,7 @@ nocsum: * unknown protocol, disable HW csum * and hope a bad packet is detected */ - return TXPKT_L4CSUM_DIS; + return TXPKT_L4CSUM_DIS_F; } } else { /* @@ -1117,15 +1117,15 @@ 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); + return TXPKT_CSUM_TYPE_V(csum_type) | + TXPKT_IPHDR_LEN_V(skb_network_header_len(skb)) | + TXPKT_ETHHDR_LEN_V(skb_network_offset(skb) - ETH_HLEN); 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 +1160,7 @@ 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 +1183,13 @@ 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 +1288,30 @@ 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)); + cntrl = (TXPKT_CSUM_TYPE_V(v6 ? + TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | + TXPKT_IPHDR_LEN_V(l3hdr_len) | + TXPKT_ETHHDR_LEN_V(eth_xtra_len)); txq->tso++; txq->tx_cso += ssi->gso_segs; } else { @@ -1320,10 +1328,10 @@ 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(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 +1340,15 @@ 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 +1671,7 @@ 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 +1760,8 @@ 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 +1772,7 @@ 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 +1783,7 @@ 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 +1826,10 @@ 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 +1841,7 @@ 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 +1883,7 @@ 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 +1944,10 @@ 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 +1959,7 @@ 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, @@ -2243,8 +2251,12 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq, * 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), diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 1bf1cdce74ac..1eafc075edae 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 @@ -35,7 +35,7 @@ #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" @@ -366,6 +366,7 @@ struct be_vf_cfg { u32 tx_rate; u32 plink_tracking; u32 privileges; + bool spoofchk; }; enum vf_state { @@ -804,6 +805,7 @@ 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..b419bde29a6f 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 @@ -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))) @@ -1457,7 +1458,7 @@ 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 +3154,7 @@ 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 +3190,14 @@ 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 +3208,7 @@ 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 +3256,10 @@ 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 +3271,7 @@ 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..c713d514fcd1 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 @@ -1109,10 +1110,6 @@ 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 +1742,24 @@ 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 +1777,13 @@ 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 +2339,9 @@ 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..9124a93eb474 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 @@ -368,6 +368,14 @@ 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 a6dcbf850c1f..f15a3cfeb217 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 @@ -211,7 +211,8 @@ 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; @@ -227,6 +228,7 @@ 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); } @@ -662,6 +664,8 @@ 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) @@ -1265,7 +1269,8 @@ 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 +1471,7 @@ 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 +1484,7 @@ 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 +1513,7 @@ 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 +1648,39 @@ 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 +1689,110 @@ 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 +2300,7 @@ 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) @@ -2573,7 +2661,7 @@ 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 +2677,7 @@ 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 +2926,7 @@ 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 +2952,18 @@ 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; } @@ -3015,7 +3112,7 @@ 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 +3159,7 @@ 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 +3326,12 @@ 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 +3363,7 @@ 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 +3663,7 @@ 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 +3710,7 @@ 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 +3758,12 @@ 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 +3840,9 @@ 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}; @@ -4182,7 +4290,7 @@ 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 +4774,11 @@ 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 +4790,22 @@ 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 +4913,7 @@ 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 +4945,8 @@ 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 +5099,7 @@ 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 @@ -5182,7 +5268,9 @@ 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 +5290,7 @@ 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; 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/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..35357ae2fe75 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); 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/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..9f6fb19062a0 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -3053,7 +3053,7 @@ 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 +3061,14 @@ 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 +3095,7 @@ 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 +3112,7 @@ 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/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/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 ec251531bd9f..f0a9cb44be6b 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; }; @@ -325,16 +326,83 @@ static bool rocker_port_is_bridged(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(struct rocker_port *rocker_port, + 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 +411,23 @@ 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(struct rocker_port *rocker_port, + enum switchdev_trans trans, + struct rocker_wait *wait) { - kfree(work); + rocker_port_kfree(rocker_port, trans, wait); } static bool rocker_wait_event_timeout(struct rocker_wait *wait, @@ -1317,12 +1388,7 @@ static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id) 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); @@ -1374,22 +1440,44 @@ 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); +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) +{ + 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(struct rocker *rocker, const struct rocker_tlv *info) { + struct rocker_mac_vlan_seen_work *sw; 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; + int flags = ROCKER_OP_FLAG_LEARNED; __be16 vlan_id; rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info); @@ -1411,7 +1499,20 @@ 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, @@ -1494,41 +1595,44 @@ typedef int (*rocker_cmd_cb_t)(struct rocker *rocker, static int rocker_cmd_exec(struct rocker *rocker, struct rocker_port *rocker_port, + enum switchdev_trans trans, rocker_cmd_cb_t prepare, void *prepare_priv, - rocker_cmd_cb_t process, void *process_priv, - bool nowait) + rocker_cmd_cb_t process, void *process_priv) { 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); 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) @@ -1539,7 +1643,7 @@ static int rocker_cmd_exec(struct rocker *rocker, rocker_desc_gen_clear(desc_info); out: - rocker_wait_destroy(wait); + rocker_wait_destroy(rocker_port, trans, wait); return err; } @@ -1762,41 +1866,46 @@ 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, + 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, + 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, + 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, + 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->rocker, 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, @@ -2308,8 +2417,8 @@ 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 +2433,9 @@ 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(rocker_port, trans, found); found = match; found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD; } else { @@ -2334,18 +2444,19 @@ 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, + return rocker_cmd_exec(rocker, rocker_port, trans, rocker_cmd_flow_tbl_add, - found, NULL, NULL, nowait); + 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 +2471,43 @@ 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(rocker_port, trans, match); if (found) { - err = rocker_cmd_exec(rocker, rocker_port, + err = rocker_cmd_exec(rocker, rocker_port, trans, rocker_cmd_flow_tbl_del, - found, NULL, NULL, nowait); - kfree(found); + found, NULL, NULL); + rocker_port_kfree(rocker_port, 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 +2517,19 @@ 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 +2543,11 @@ 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 +2556,7 @@ 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 +2580,11 @@ 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 +2596,7 @@ 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 +2609,7 @@ 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 +2634,11 @@ 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 +2646,7 @@ 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 +2660,29 @@ 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; @@ -2602,7 +2711,7 @@ 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 * @@ -2620,22 +2729,24 @@ 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(struct rocker_port *rocker_port, + 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(rocker_port, trans, entry->group_ids); break; default: break; } - kfree(entry); + rocker_port_kfree(rocker_port, 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 +2757,9 @@ 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(rocker_port, trans, found); found = match; found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD; } else { @@ -2655,18 +2767,19 @@ 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, + return rocker_cmd_exec(rocker, rocker_port, trans, rocker_cmd_group_tbl_add, - found, NULL, NULL, nowait); + 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 +2791,95 @@ 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(rocker_port, trans, match); if (found) { - err = rocker_cmd_exec(rocker, rocker_port, + err = rocker_cmd_exec(rocker, 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(rocker_port, 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) { 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(rocker_port, 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, + 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, u8 *src_mac, 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; @@ -2777,7 +2892,7 @@ 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 * @@ -2802,17 +2917,17 @@ static void _rocker_neigh_add(struct rocker *rocker, be32_to_cpu(entry->ip_addr)); } -static void _rocker_neigh_del(struct rocker *rocker, +static void _rocker_neigh_del(struct rocker_port *rocker_port, + enum switchdev_trans trans, struct rocker_neigh_tbl_entry *entry) { if (--entry->ref_count == 0) { hash_del(&entry->entry); - kfree(entry); + rocker_port_kfree(rocker_port, trans, entry); } } -static void _rocker_neigh_update(struct rocker *rocker, - struct rocker_neigh_tbl_entry *entry, +static void _rocker_neigh_update(struct rocker_neigh_tbl_entry *entry, u8 *eth_dst, bool ttl_check) { if (eth_dst) { @@ -2824,6 +2939,7 @@ static void _rocker_neigh_update(struct rocker *rocker, } static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, __be32 ip_addr, u8 *eth_dst) { struct rocker *rocker = rocker_port->rocker; @@ -2840,7 +2956,7 @@ 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; @@ -2860,9 +2976,9 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, _rocker_neigh_add(rocker, entry); } else if (removing) { memcpy(entry, found, sizeof(*entry)); - _rocker_neigh_del(rocker, found); + _rocker_neigh_del(rocker_port, trans, found); } else if (updating) { - _rocker_neigh_update(rocker, found, eth_dst, true); + _rocker_neigh_update(found, eth_dst, true); memcpy(entry, found, sizeof(*entry)); } else { err = -ENOENT; @@ -2879,7 +2995,7 @@ 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 +3011,7 @@ 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 +3025,13 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port, err_out: if (!adding) - kfree(entry); + rocker_port_kfree(rocker_port, 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); @@ -2932,14 +3048,16 @@ 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); 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; @@ -2952,7 +3070,7 @@ 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; @@ -2973,9 +3091,9 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port, int flags, *index = entry->index; resolved = false; } else if (removing) { - _rocker_neigh_del(rocker, found); + _rocker_neigh_del(rocker_port, trans, found); } else if (updating) { - _rocker_neigh_update(rocker, found, NULL, false); + _rocker_neigh_update(found, NULL, false); resolved = !is_zero_ether_addr(found->eth_dst); } else { err = -ENOENT; @@ -2984,7 +3102,7 @@ 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(rocker_port, trans, entry); if (err) return err; @@ -2992,12 +3110,13 @@ 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; @@ -3008,8 +3127,8 @@ static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port, 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; @@ -3032,21 +3151,20 @@ 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(rocker_port, 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; struct rocker_port *p; @@ -3063,9 +3181,8 @@ 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", @@ -3089,9 +3206,8 @@ 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); @@ -3147,8 +3263,8 @@ 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, + struct rocker_ctrl *ctrl, __be16 vlan_id) { u32 in_pport = rocker_port->pport; u32 in_pport_mask = 0xffffffff; @@ -3163,7 +3279,7 @@ 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, @@ -3180,7 +3296,8 @@ 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, + struct rocker_ctrl *ctrl, __be16 vlan_id) { enum rocker_of_dpa_table_id goto_tbl = @@ -3192,7 +3309,7 @@ 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); @@ -3204,8 +3321,8 @@ 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, + struct rocker_ctrl *ctrl, __be16 vlan_id) { u32 in_pport_mask = 0xffffffff; __be16 vlan_id_mask = htons(0xffff); @@ -3214,7 +3331,7 @@ 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, @@ -3227,32 +3344,34 @@ 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, +static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, 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; @@ -3262,7 +3381,8 @@ 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, +static int rocker_port_ctrl(struct rocker_port *rocker_port, + enum switchdev_trans trans, int flags, struct rocker_ctrl *ctrl) { u16 vid; @@ -3271,7 +3391,7 @@ 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; @@ -3280,8 +3400,8 @@ 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; @@ -3295,50 +3415,57 @@ 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; @@ -3353,7 +3480,7 @@ 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) @@ -3365,7 +3492,8 @@ 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; @@ -3377,23 +3505,24 @@ static void rocker_port_fdb_learn_work(struct work_struct *work) 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->rocker_port, 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 = @@ -3409,8 +3538,8 @@ 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; @@ -3422,18 +3551,22 @@ 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(rocker_port, trans, lw); + else + schedule_work(&lw->work); return 0; } @@ -3451,6 +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) { @@ -3460,7 +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; @@ -3475,7 +3609,7 @@ static int rocker_port_fdb(struct rocker_port *rocker_port, found = rocker_fdb_tbl_find(rocker, fdb); if (removing && found) { - kfree(fdb); + rocker_port_kfree(rocker_port, trans, fdb); hash_del(&found->entry); } else if (!removing && !found) { hash_add(rocker->fdb_tbl, &fdb->entry, fdb->key_crc32); @@ -3485,22 +3619,23 @@ static int rocker_port_fdb(struct rocker_port *rocker_port, /* Check if adding and already exists, or removing and can't find */ if (!found != !removing) { - kfree(fdb); + rocker_port_kfree(rocker_port, 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; @@ -3516,7 +3651,7 @@ 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) @@ -3531,7 +3666,8 @@ 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; @@ -3544,7 +3680,7 @@ 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, @@ -3553,7 +3689,7 @@ 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, @@ -3562,13 +3698,14 @@ 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 @@ -3588,9 +3725,8 @@ 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", @@ -3602,13 +3738,21 @@ 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; @@ -3636,41 +3780,50 @@ 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); + +err_out: + if (trans == SWITCHDEV_TRANS_PREPARE) { + memcpy(rocker_port->ctrls, prev_ctrls, sizeof(prev_ctrls)); + rocker_port->stp_state = prev_state; + } - return rocker_port_fwding(rocker_port); + 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 * @@ -3688,6 +3841,7 @@ rocker_internal_vlan_tbl_find(struct rocker *rocker, int ifindex) } static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port, + enum switchdev_trans trans, int ifindex) { struct rocker *rocker = rocker_port->rocker; @@ -3696,7 +3850,7 @@ static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port, unsigned long lock_flags; int i; - entry = kzalloc(sizeof(*entry), GFP_KERNEL); + entry = rocker_port_kzalloc(rocker_port, trans, sizeof(*entry)); if (!entry) return 0; @@ -3706,7 +3860,7 @@ static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port, found = rocker_internal_vlan_tbl_find(rocker, ifindex); if (found) { - kfree(entry); + rocker_port_kfree(rocker_port, trans, entry); goto found; } @@ -3730,6 +3884,7 @@ found: } static void rocker_port_internal_vlan_id_put(struct rocker_port *rocker_port, + enum switchdev_trans trans, int ifindex) { struct rocker *rocker = rocker_port->rocker; @@ -3751,14 +3906,15 @@ static void rocker_port_internal_vlan_id_put(struct rocker_port *rocker_port, bit = ntohs(found->vlan_id) - ROCKER_INTERNAL_VLAN_ID_BASE; clear_bit(bit, rocker->internal_vlan_bitmap); hash_del(&found->entry); - kfree(found); + rocker_port_kfree(rocker_port, trans, found); } 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, +static int rocker_port_fib_ipv4(struct rocker_port *rocker_port, + enum switchdev_trans trans, __be32 dst, int dst_len, struct fib_info *fi, u32 tb_id, int flags) { @@ -3782,7 +3938,7 @@ 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; @@ -3793,7 +3949,7 @@ 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) @@ -3832,7 +3988,7 @@ 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; @@ -3859,7 +4015,7 @@ 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); @@ -4012,11 +4168,12 @@ static int rocker_port_vlan_rx_add_vid(struct net_device *dev, struct rocker_port *rocker_port = netdev_priv(dev); int err; - err = rocker_port_vlan(rocker_port, 0, vid); + err = rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, 0, vid); if (err) return err; - return rocker_port_router_mac(rocker_port, 0, htons(vid)); + return rocker_port_router_mac(rocker_port, SWITCHDEV_TRANS_NONE, + 0, htons(vid)); } static int rocker_port_vlan_rx_kill_vid(struct net_device *dev, @@ -4025,12 +4182,13 @@ static int rocker_port_vlan_rx_kill_vid(struct net_device *dev, struct rocker_port *rocker_port = netdev_priv(dev); int err; - err = rocker_port_router_mac(rocker_port, ROCKER_OP_FLAG_REMOVE, - htons(vid)); + err = rocker_port_router_mac(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, htons(vid)); if (err) return err; - return rocker_port_vlan(rocker_port, ROCKER_OP_FLAG_REMOVE, vid); + return rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, vid); } static int rocker_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], @@ -4045,7 +4203,8 @@ static int rocker_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], if (!rocker_port_is_bridged(rocker_port)) return -EINVAL; - return rocker_port_fdb(rocker_port, addr, vlan_id, flags); + return rocker_port_fdb(rocker_port, SWITCHDEV_TRANS_NONE, + addr, vlan_id, flags); } static int rocker_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], @@ -4059,7 +4218,8 @@ static int rocker_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], if (!rocker_port_is_bridged(rocker_port)) return -EINVAL; - return rocker_port_fdb(rocker_port, addr, vlan_id, flags); + return rocker_port_fdb(rocker_port, SWITCHDEV_TRANS_NONE, + addr, vlan_id, flags); } static int rocker_fdb_fill_info(struct sk_buff *skb, @@ -4135,58 +4295,6 @@ skip: return idx; } -static int rocker_port_bridge_setlink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) -{ - struct rocker_port *rocker_port = netdev_priv(dev); - struct nlattr *protinfo; - struct nlattr *attr; - int err; - - 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; - } - } - - return 0; -} - -static int rocker_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, - u32 filter_mask, int nlflags) -{ - struct rocker_port *rocker_port = netdev_priv(dev); - u16 mode = BRIDGE_MODE_UNDEF; - u32 mask = BR_LEARNING | BR_LEARNING_SYNC; - - return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, - rocker_port->brport_flags, mask, - nlflags); -} - static int rocker_port_get_phys_port_name(struct net_device *dev, char *buf, size_t len) { @@ -4195,9 +4303,10 @@ static int rocker_port_get_phys_port_name(struct net_device *dev, int err; err = rocker_cmd_exec(rocker_port->rocker, rocker_port, + SWITCHDEV_TRANS_NONE, rocker_cmd_get_port_settings_prep, NULL, rocker_cmd_get_port_settings_phys_name_proc, - &name, false); + &name); return err ? -EOPNOTSUPP : 0; } @@ -4212,8 +4321,9 @@ static const struct net_device_ops rocker_port_netdev_ops = { .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_bridge_getlink = switchdev_port_bridge_getlink, + .ndo_bridge_setlink = switchdev_port_bridge_setlink, + .ndo_bridge_dellink = switchdev_port_bridge_dellink, .ndo_get_phys_port_name = rocker_port_get_phys_port_name, }; @@ -4221,54 +4331,217 @@ static const struct net_device_ops rocker_port_netdev_ops = { * swdev interface ********************/ -static int rocker_port_swdev_parent_id_get(struct net_device *dev, - struct netdev_phys_item_id *psid) +static int rocker_port_attr_get(struct net_device *dev, + struct switchdev_attr *attr) { struct rocker_port *rocker_port = netdev_priv(dev); struct rocker *rocker = rocker_port->rocker; - psid->id_len = sizeof(rocker->hw.id); - memcpy(&psid->id, &rocker->hw.id, psid->id_len); + 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 0; } -static int rocker_port_swdev_port_stp_update(struct net_device *dev, u8 state) +static void rocker_port_trans_abort(struct rocker_port *rocker_port) +{ + struct list_head *mem, *tmp; + + list_for_each_safe(mem, tmp, &rocker_port->trans_mem) { + list_del(mem); + kfree(mem); + } +} + +static int rocker_port_brport_flags_set(struct rocker_port *rocker_port, + enum switchdev_trans trans, + unsigned long brport_flags) +{ + unsigned long orig_flags; + int err = 0; + + 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); + + if (trans == SWITCHDEV_TRANS_PREPARE) + rocker_port->brport_flags = orig_flags; + + return err; +} + +static int rocker_port_attr_set(struct net_device *dev, + struct switchdev_attr *attr) { struct rocker_port *rocker_port = netdev_priv(dev); + int err = 0; + + 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; + } + + 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; + } + + return err; +} + +static int rocker_port_vlan_add(struct rocker_port *rocker_port, + enum switchdev_trans trans, u16 vid, u16 flags) +{ + int err; + + /* XXX deal with flags for PVID and untagged */ + + err = rocker_port_vlan(rocker_port, trans, 0, vid); + if (err) + return err; + + return rocker_port_router_mac(rocker_port, trans, 0, htons(vid)); +} + +static int rocker_port_vlans_add(struct rocker_port *rocker_port, + enum switchdev_trans trans, + struct switchdev_obj_vlan *vlan) +{ + u16 vid; + int err; - return rocker_port_stp_update(rocker_port, state); + 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; + } + + return 0; } -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_obj_add(struct net_device *dev, + struct switchdev_obj *obj) { struct rocker_port *rocker_port = netdev_priv(dev); - int flags = 0; + 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; + } - return rocker_port_fib_ipv4(rocker_port, dst, dst_len, - fi, tb_id, flags); + 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; + default: + err = -EOPNOTSUPP; + break; + } + + 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_vlan_del(struct rocker_port *rocker_port, + u16 vid, u16 flags) +{ + int err; + + err = rocker_port_router_mac(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, htons(vid)); + if (err) + return err; + + return rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, vid); +} + +static int rocker_port_vlans_del(struct rocker_port *rocker_port, + struct switchdev_obj_vlan *vlan) +{ + u16 vid; + int err; + + 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_obj_del(struct net_device *dev, + struct switchdev_obj *obj) { struct rocker_port *rocker_port = netdev_priv(dev); - int flags = ROCKER_OP_FLAG_REMOVE; + struct switchdev_obj_ipv4_fib *fib4; + int err = 0; - return rocker_port_fib_ipv4(rocker_port, dst, dst_len, - fi, tb_id, flags); + 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; + 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, }; /******************** @@ -4399,9 +4672,10 @@ static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port, void *priv) { return rocker_cmd_exec(rocker_port->rocker, 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, @@ -4415,8 +4689,6 @@ 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) @@ -4470,8 +4742,9 @@ 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++; @@ -4583,7 +4856,8 @@ static void rocker_remove_ports(struct rocker *rocker) 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); @@ -4619,11 +4893,12 @@ 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); 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, @@ -4631,8 +4906,7 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number) rocker_carrier_init(rocker_port); dev->features |= NETIF_F_NETNS_LOCAL | - NETIF_F_HW_VLAN_CTAG_FILTER | - NETIF_F_HW_SWITCH_OFFLOAD; + NETIF_F_HW_VLAN_CTAG_FILTER; err = register_netdev(dev); if (err) { @@ -4641,11 +4915,13 @@ 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); + rocker_port_internal_vlan_id_get(rocker_port, + SWITCHDEV_TRANS_NONE, + dev->ifindex); + 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; @@ -4892,43 +5168,47 @@ static int rocker_port_bridge_join(struct rocker_port *rocker_port, { int err; - rocker_port_internal_vlan_id_put(rocker_port, + rocker_port_internal_vlan_id_put(rocker_port, SWITCHDEV_TRANS_NONE, rocker_port->dev->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); + err = rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, 0); if (err) return err; rocker_port->internal_vlan_id = rocker_port_internal_vlan_id_get(rocker_port, + SWITCHDEV_TRANS_NONE, bridge->ifindex); - return rocker_port_vlan(rocker_port, 0, 0); + return rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, 0, 0); } static int rocker_port_bridge_leave(struct rocker_port *rocker_port) { int err; - rocker_port_internal_vlan_id_put(rocker_port, + rocker_port_internal_vlan_id_put(rocker_port, SWITCHDEV_TRANS_NONE, 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(rocker_port, SWITCHDEV_TRANS_NONE, + ROCKER_OP_FLAG_REMOVE, 0); if (err) return err; rocker_port->internal_vlan_id = rocker_port_internal_vlan_id_get(rocker_port, + SWITCHDEV_TRANS_NONE, rocker_port->dev->ifindex); - err = rocker_port_vlan(rocker_port, 0, 0); + err = rocker_port_vlan(rocker_port, SWITCHDEV_TRANS_NONE, 0, 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; } @@ -4990,7 +5270,8 @@ 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/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..882117a43c3a 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); @@ -92,7 +95,28 @@ 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; } static int efx_ef10_init_datapath_caps(struct efx_nic *efx) @@ -117,6 +141,13 @@ 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, @@ -178,7 +209,7 @@ 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; @@ -209,6 +240,8 @@ 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 +263,10 @@ 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; @@ -251,10 +288,22 @@ 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 +311,7 @@ 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); @@ -279,6 +328,23 @@ fail1: return rc; } +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) +{ + return efx_ef10_probe(efx); +} +#else +static int efx_ef10_probe_vf(struct efx_nic *efx __attribute__ ((unused))) +{ + return 0; +} +#endif + static int efx_ef10_free_vis(struct efx_nic *efx) { MCDI_DECLARE_BUF_OUT_OR_ERR(outbuf, 0); @@ -687,7 +753,9 @@ 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; } @@ -1044,6 +1112,14 @@ 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 +1199,10 @@ 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; @@ -1237,6 +1317,7 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) 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; + struct efx_ef10_nic_data *nic_data = efx->nic_data; size_t inlen, outlen; dma_addr_t dma_addr; efx_qword_t *txd; @@ -1251,7 +1332,7 @@ 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; @@ -1378,19 +1459,33 @@ 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 +1497,9 @@ 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 +1516,8 @@ 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 +1531,7 @@ 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 +1559,119 @@ 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) @@ -1500,6 +1691,7 @@ static void efx_ef10_rx_init(struct efx_rx_queue *rx_queue) 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; + struct efx_ef10_nic_data *nic_data = efx->nic_data; size_t inlen, outlen; dma_addr_t dma_addr; int rc; @@ -1517,7 +1709,7 @@ 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; @@ -2286,11 +2478,12 @@ 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, @@ -3494,6 +3687,9 @@ 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 +3767,12 @@ 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) { @@ -3607,9 +3809,109 @@ 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_mcdi_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, + .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 + .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, @@ -3650,7 +3952,7 @@ 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 +3991,22 @@ 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, + .vswitching_probe = efx_ef10_vswitching_probe_pf, + .vswitching_restore = efx_ef10_vswitching_restore_pf, + .vswitching_remove = efx_ef10_vswitching_remove_pf, +#endif .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..1b93acf2d28d --- /dev/null +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -0,0 +1,430 @@ +/**************************************************************************** + * 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); + + 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, 0); + MCDI_POPULATE_DWORD_1(inbuf, VSWITCH_ALLOC_IN_FLAGS, + VSWITCH_ALLOC_IN_FLAG_AUTO_PORT, 0); + + return efx_mcdi_rpc(efx, MC_CMD_VSWITCH_ALLOC, inbuf, sizeof(inbuf), + NULL, 0, NULL); +} + +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, + 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, 0); + MCDI_POPULATE_DWORD_1(inbuf, VPORT_ALLOC_IN_FLAGS, + VPORT_ALLOC_IN_FLAG_AUTO_PORT, 0); + + 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; + } + } +} + +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->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); + + 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, + &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"); +} diff --git a/drivers/net/ethernet/sfc/ef10_sriov.h b/drivers/net/ethernet/sfc/ef10_sriov.h new file mode 100644 index 000000000000..86bac7ebb01a --- /dev/null +++ b/drivers/net/ethernet/sfc/ef10_sriov.h @@ -0,0 +1,70 @@ +/**************************************************************************** + * 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 + * @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 + */ +struct ef10_vf { + unsigned int vport_id; + unsigned int vport_assigned; + u8 mac[ETH_ALEN]; +}; + +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_mac_address_changed(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) {} + +static inline int efx_ef10_sriov_set_vf_mac(struct efx_nic *efx, int vf, + u8 *mac) +{ + return -EOPNOTSUPP; +} + +static inline int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf, + u16 vlan, u8 qos) +{ + return -EOPNOTSUPP; +} + +static inline int efx_ef10_sriov_set_vf_spoofchk(struct efx_nic *efx, int vf, + bool spoofchk) +{ + return -EOPNOTSUPP; +} + +static inline int efx_ef10_sriov_get_vf_config(struct efx_nic *efx, int vf, + struct ifla_vf_info *ivf) +{ + return -EOPNOTSUPP; +} + +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..0f127a01b5e8 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" @@ -1045,7 +1046,7 @@ static int efx_init_port(struct efx_nic *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); @@ -1200,10 +1201,12 @@ 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 +1237,8 @@ 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 +1261,7 @@ 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 +1272,8 @@ 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 +1282,23 @@ 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 +1329,19 @@ 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 +1445,15 @@ 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 +1617,6 @@ 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 +1639,9 @@ 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); @@ -1712,21 +1734,33 @@ 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 +1850,9 @@ 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); } @@ -2168,7 +2205,8 @@ static int efx_set_mac_address(struct net_device *net_dev, void *data) } ether_addr_copy(net_dev->dev_addr, new_addr); - efx->type->sriov_mac_address_changed(efx); + if (efx->type->sriov_mac_address_changed) + efx->type->sriov_mac_address_changed(efx); /* Reconfigure the MAC */ mutex_lock(&efx->mac_lock); @@ -2199,7 +2237,7 @@ 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 +2250,10 @@ 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, #endif #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = efx_netpoll, @@ -2229,29 +2267,6 @@ 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 +2279,7 @@ 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)); @@ -2292,12 +2306,9 @@ 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; @@ -2426,7 +2437,8 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok) 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 +2446,18 @@ 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 + efx_restore_filters(efx); - efx->type->sriov_reset(efx); + if (efx->type->sriov_reset) + efx->type->sriov_reset(efx); mutex_unlock(&efx->mac_lock); @@ -2655,6 +2677,8 @@ 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 +2850,9 @@ 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 +3034,8 @@ 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 +3050,12 @@ 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 +3087,26 @@ 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 +3329,9 @@ 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 +3354,11 @@ 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 +3375,10 @@ 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 +3390,9 @@ 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..9097906ecfb4 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -15,7 +15,9 @@ #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 /* TX */ int efx_probe_tx_queue(struct efx_tx_queue *tx_queue); @@ -32,6 +34,7 @@ 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); @@ -220,6 +223,13 @@ 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/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c index 4835bc0d0de8..03829b48547a 100644 --- a/drivers/net/ethernet/sfc/ethtool.c +++ b/drivers/net/ethernet/sfc/ethtool.c @@ -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..b44ee31f1a7a 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -1035,7 +1035,9 @@ 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 +1083,7 @@ 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 +1108,11 @@ 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 +1142,26 @@ 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 +1186,6 @@ 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); @@ -1385,6 +1383,9 @@ 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); @@ -1406,6 +1407,8 @@ 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 +1446,31 @@ 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 +1478,10 @@ 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) @@ -1688,6 +1696,36 @@ 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_OUT_OR_ERR(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..5df1e986e39e 100644 --- a/drivers/net/ethernet/sfc/mcdi.h +++ b/drivers/net/ethernet/sfc/mcdi.h @@ -339,6 +339,8 @@ 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..4fa6eb27cc56 100644 --- a/drivers/net/ethernet/sfc/mcdi_pcol.h +++ b/drivers/net/ethernet/sfc/mcdi_pcol.h @@ -1875,6 +1875,8 @@ #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 @@ -4087,6 +4089,27 @@ /***********************************/ +/* 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_READ_REGS * Get a dump of the MCPU registers */ diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 325dd94bca46..031a3385ad4b 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -793,7 +793,6 @@ union efx_multicast_hash { efx_oword_t oword[EFX_MCAST_HASH_ENTRIES / sizeof(efx_oword_t) / 8]; }; -struct efx_vf; struct vfdi_status; /** @@ -909,7 +908,6 @@ 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. @@ -1053,7 +1051,6 @@ 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 +1089,7 @@ 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() @@ -1226,6 +1224,8 @@ 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 +1277,8 @@ 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 +1331,23 @@ 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 (*vswitching_probe)(struct efx_nic *efx); + int (*vswitching_restore)(struct efx_nic *efx); + void (*vswitching_remove)(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..2fd30556e6c3 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 +514,25 @@ 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; -} + 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 +564,7 @@ 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/siena.c b/drivers/net/ethernet/sfc/siena.c index f12c811938d2..8b4130abad66 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, +#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..caf701ac274f 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; } @@ -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; @@ -1077,7 +1080,7 @@ 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 +1102,7 @@ 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 +1158,7 @@ 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 +1193,16 @@ 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 +1221,12 @@ 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 +1243,7 @@ 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 +1256,7 @@ 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 +1356,7 @@ 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 +1367,7 @@ 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 +1383,7 @@ 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 +1394,7 @@ 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 +1402,7 @@ 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 +1458,12 @@ 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); @@ -1481,7 +1488,7 @@ 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) { - struct efx_vf *vf; + struct siena_vf *vf; unsigned queue, qid; queue = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_SUBDATA); @@ -1500,7 +1507,7 @@ 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 +1532,7 @@ 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 +1548,10 @@ 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 +1565,7 @@ 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 +1581,6 @@ int efx_init_sriov(void) vfdi_workqueue = create_singlethread_workqueue("sfc_vfdi"); if (!vfdi_workqueue) return -ENOMEM; - return 0; } @@ -1582,14 +1589,14 @@ 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 +1606,16 @@ 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 +1626,16 @@ 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 +1650,16 @@ 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 +1673,12 @@ 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..64e3e018929e --- /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); +void 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..d4b74452a677 --- /dev/null +++ b/drivers/net/ethernet/sfc/sriov.c @@ -0,0 +1,60 @@ +/**************************************************************************** + * 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; +} diff --git a/drivers/net/ethernet/sfc/sriov.h b/drivers/net/ethernet/sfc/sriov.h new file mode 100644 index 000000000000..0b9f0f6acf3b --- /dev/null +++ b/drivers/net/ethernet/sfc/sriov.h @@ -0,0 +1,27 @@ +/**************************************************************************** + * 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); + +#endif /* CONFIG_SFC_SRIOV */ + +#endif /* EFX_SRIOV_H */ diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 43efc3a0cda5..0a28c07361cf 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); } 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: |
