diff options
Diffstat (limited to 'drivers/net/ethernet/chelsio')
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h | 11 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c | 57 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 32 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c | 6 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c | 515 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c | 186 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 152 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 8 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/sched.c | 3 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/sge.c | 376 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 51 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h | 1 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/t4_regs.h | 29 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h | 12 | ||||
| -rw-r--r-- | drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c | 4 |
15 files changed, 1220 insertions, 223 deletions
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h index 3c5057868ab3..36d25883d123 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_entity.h @@ -120,6 +120,8 @@ struct cudbg_mem_desc { u32 idx; }; +#define CUDBG_MEMINFO_REV 1 + struct cudbg_meminfo { struct cudbg_mem_desc avail[4]; struct cudbg_mem_desc mem[ARRAY_SIZE(cudbg_region) + 3]; @@ -137,6 +139,9 @@ struct cudbg_meminfo { u32 port_alloc[4]; u32 loopback_used[NCHAN]; u32 loopback_alloc[NCHAN]; + u32 p_structs_free_cnt; + u32 free_rx_cnt; + u32 free_tx_cnt; }; struct cudbg_cim_pif_la { @@ -281,12 +286,18 @@ struct cudbg_tid_data { #define CUDBG_NUM_ULPTX 11 #define CUDBG_NUM_ULPTX_READ 512 +#define CUDBG_NUM_ULPTX_ASIC 6 +#define CUDBG_NUM_ULPTX_ASIC_READ 128 + +#define CUDBG_ULPTX_LA_REV 1 struct cudbg_ulptx_la { u32 rdptr[CUDBG_NUM_ULPTX]; u32 wrptr[CUDBG_NUM_ULPTX]; u32 rddata[CUDBG_NUM_ULPTX]; u32 rd_data[CUDBG_NUM_ULPTX][CUDBG_NUM_ULPTX_READ]; + u32 rdptr_asic[CUDBG_NUM_ULPTX_ASIC_READ]; + u32 rddata_asic[CUDBG_NUM_ULPTX_ASIC_READ][CUDBG_NUM_ULPTX_ASIC]; }; #define CUDBG_CHAC_PBT_ADDR 0x2800 diff --git a/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c index 0afcfe99bff3..d97e0d7e541a 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cudbg_lib.c @@ -349,6 +349,11 @@ int cudbg_fill_meminfo(struct adapter *padap, meminfo_buff->up_extmem2_hi = hi; lo = t4_read_reg(padap, TP_PMM_RX_MAX_PAGE_A); + for (i = 0, meminfo_buff->free_rx_cnt = 0; i < 2; i++) + meminfo_buff->free_rx_cnt += + FREERXPAGECOUNT_G(t4_read_reg(padap, + TP_FLM_FREE_RX_CNT_A)); + meminfo_buff->rx_pages_data[0] = PMRXMAXPAGE_G(lo); meminfo_buff->rx_pages_data[1] = t4_read_reg(padap, TP_PMM_RX_PAGE_SIZE_A) >> 10; @@ -356,6 +361,11 @@ int cudbg_fill_meminfo(struct adapter *padap, lo = t4_read_reg(padap, TP_PMM_TX_MAX_PAGE_A); hi = t4_read_reg(padap, TP_PMM_TX_PAGE_SIZE_A); + for (i = 0, meminfo_buff->free_tx_cnt = 0; i < 4; i++) + meminfo_buff->free_tx_cnt += + FREETXPAGECOUNT_G(t4_read_reg(padap, + TP_FLM_FREE_TX_CNT_A)); + meminfo_buff->tx_pages_data[0] = PMTXMAXPAGE_G(lo); meminfo_buff->tx_pages_data[1] = hi >= (1 << 20) ? (hi >> 20) : (hi >> 10); @@ -364,6 +374,8 @@ int cudbg_fill_meminfo(struct adapter *padap, meminfo_buff->tx_pages_data[3] = 1 << PMTXNUMCHN_G(lo); meminfo_buff->p_structs = t4_read_reg(padap, TP_CMM_MM_MAX_PSTRUCT_A); + meminfo_buff->p_structs_free_cnt = + FREEPSTRUCTCOUNT_G(t4_read_reg(padap, TP_FLM_FREE_PS_CNT_A)); for (i = 0; i < 4; i++) { if (CHELSIO_CHIP_VERSION(padap->params.chip) > CHELSIO_T5) @@ -1465,14 +1477,23 @@ int cudbg_collect_meminfo(struct cudbg_init *pdbg_init, struct adapter *padap = pdbg_init->adap; struct cudbg_buffer temp_buff = { 0 }; struct cudbg_meminfo *meminfo_buff; + struct cudbg_ver_hdr *ver_hdr; int rc; - rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_meminfo), + rc = cudbg_get_buff(pdbg_init, dbg_buff, + sizeof(struct cudbg_ver_hdr) + + sizeof(struct cudbg_meminfo), &temp_buff); if (rc) return rc; - meminfo_buff = (struct cudbg_meminfo *)temp_buff.data; + ver_hdr = (struct cudbg_ver_hdr *)temp_buff.data; + ver_hdr->signature = CUDBG_ENTITY_SIGNATURE; + ver_hdr->revision = CUDBG_MEMINFO_REV; + ver_hdr->size = sizeof(struct cudbg_meminfo); + + meminfo_buff = (struct cudbg_meminfo *)(temp_buff.data + + sizeof(*ver_hdr)); rc = cudbg_fill_meminfo(padap, meminfo_buff); if (rc) { cudbg_err->sys_err = rc; @@ -2586,15 +2607,24 @@ int cudbg_collect_ulptx_la(struct cudbg_init *pdbg_init, struct adapter *padap = pdbg_init->adap; struct cudbg_buffer temp_buff = { 0 }; struct cudbg_ulptx_la *ulptx_la_buff; + struct cudbg_ver_hdr *ver_hdr; u32 i, j; int rc; - rc = cudbg_get_buff(pdbg_init, dbg_buff, sizeof(struct cudbg_ulptx_la), + rc = cudbg_get_buff(pdbg_init, dbg_buff, + sizeof(struct cudbg_ver_hdr) + + sizeof(struct cudbg_ulptx_la), &temp_buff); if (rc) return rc; - ulptx_la_buff = (struct cudbg_ulptx_la *)temp_buff.data; + ver_hdr = (struct cudbg_ver_hdr *)temp_buff.data; + ver_hdr->signature = CUDBG_ENTITY_SIGNATURE; + ver_hdr->revision = CUDBG_ULPTX_LA_REV; + ver_hdr->size = sizeof(struct cudbg_ulptx_la); + + ulptx_la_buff = (struct cudbg_ulptx_la *)(temp_buff.data + + sizeof(*ver_hdr)); for (i = 0; i < CUDBG_NUM_ULPTX; i++) { ulptx_la_buff->rdptr[i] = t4_read_reg(padap, ULP_TX_LA_RDPTR_0_A + @@ -2610,6 +2640,25 @@ int cudbg_collect_ulptx_la(struct cudbg_init *pdbg_init, t4_read_reg(padap, ULP_TX_LA_RDDATA_0_A + 0x10 * i); } + + for (i = 0; i < CUDBG_NUM_ULPTX_ASIC_READ; i++) { + t4_write_reg(padap, ULP_TX_ASIC_DEBUG_CTRL_A, 0x1); + ulptx_la_buff->rdptr_asic[i] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_CTRL_A); + ulptx_la_buff->rddata_asic[i][0] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_0_A); + ulptx_la_buff->rddata_asic[i][1] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_1_A); + ulptx_la_buff->rddata_asic[i][2] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_2_A); + ulptx_la_buff->rddata_asic[i][3] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_3_A); + ulptx_la_buff->rddata_asic[i][4] = + t4_read_reg(padap, ULP_TX_ASIC_DEBUG_4_A); + ulptx_la_buff->rddata_asic[i][5] = + t4_read_reg(padap, PM_RX_BASE_ADDR); + } + return cudbg_write_and_release_buff(pdbg_init, &temp_buff, dbg_buff); } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 0dbe2d9e22d6..3da9299cd786 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -46,6 +46,7 @@ #include <linux/spinlock.h> #include <linux/timer.h> #include <linux/vmalloc.h> +#include <linux/rhashtable.h> #include <linux/etherdevice.h> #include <linux/net_tstamp.h> #include <linux/ptp_clock_kernel.h> @@ -319,6 +320,21 @@ struct vpd_params { u8 na[MACADDR_LEN + 1]; }; +/* Maximum resources provisioned for a PCI PF. + */ +struct pf_resources { + unsigned int nvi; /* N virtual interfaces */ + unsigned int neq; /* N egress Qs */ + unsigned int nethctrl; /* N egress ETH or CTRL Qs */ + unsigned int niqflint; /* N ingress Qs/w free list(s) & intr */ + unsigned int niq; /* N ingress Qs */ + unsigned int tc; /* PCI-E traffic class */ + unsigned int pmask; /* port access rights mask */ + unsigned int nexactf; /* N exact MPS filters */ + unsigned int r_caps; /* read capabilities */ + unsigned int wx_caps; /* write/execute capabilities */ +}; + struct pci_params { unsigned int vpd_cap_addr; unsigned char speed; @@ -346,6 +362,7 @@ struct adapter_params { struct sge_params sge; struct tp_params tp; struct vpd_params vpd; + struct pf_resources pfres; struct pci_params pci; struct devlog_params devlog; enum pcie_memwin drv_memwin; @@ -521,6 +538,15 @@ enum { MAX_INGQ = MAX_ETH_QSETS + INGQ_EXTRAS, }; +enum { + PRIV_FLAG_PORT_TX_VM_BIT, +}; + +#define PRIV_FLAG_PORT_TX_VM BIT(PRIV_FLAG_PORT_TX_VM_BIT) + +#define PRIV_FLAGS_ADAP 0 +#define PRIV_FLAGS_PORT PRIV_FLAG_PORT_TX_VM + struct adapter; struct sge_rspq; @@ -557,6 +583,7 @@ struct port_info { struct hwtstamp_config tstamp_config; bool ptp_enable; struct sched_table *sched_tbl; + u32 eth_flags; }; struct dentry; @@ -867,6 +894,7 @@ struct adapter { unsigned int flags; unsigned int adap_idx; enum chip_type chip; + u32 eth_flags; int msg_enable; __be16 vxlan_port; @@ -956,6 +984,7 @@ struct adapter { struct chcr_stats_debug chcr_stats; /* TC flower offload */ + bool tc_flower_initialized; struct rhashtable flower_tbl; struct rhashtable_params flower_ht_params; struct timer_list flower_stats_timer; @@ -1333,7 +1362,7 @@ void t4_os_link_changed(struct adapter *adap, int port_id, int link_stat); void t4_free_sge_resources(struct adapter *adap); void t4_free_ofld_rxqs(struct adapter *adap, int n, struct sge_ofld_rxq *q); irq_handler_t t4_intr_handler(struct adapter *adap); -netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev); +netdev_tx_t t4_start_xmit(struct sk_buff *skb, struct net_device *dev); int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp, const struct pkt_gl *gl); int t4_mgmt_tx(struct adapter *adap, struct sk_buff *skb); @@ -1555,6 +1584,7 @@ int t4_eeprom_ptov(unsigned int phys_addr, unsigned int fn, unsigned int sz); int t4_seeprom_wp(struct adapter *adapter, bool enable); int t4_get_raw_vpd_params(struct adapter *adapter, struct vpd_params *p); int t4_get_vpd_params(struct adapter *adapter, struct vpd_params *p); +int t4_get_pfres(struct adapter *adapter); int t4_read_flash(struct adapter *adapter, unsigned int addr, unsigned int nwords, u32 *data, int byte_oriented); int t4_load_fw(struct adapter *adapter, const u8 *fw_data, unsigned int size); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c index 8d751efcb90e..5f01c0a7fd98 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c @@ -224,7 +224,8 @@ static u32 cxgb4_get_entity_length(struct adapter *adap, u32 entity) len = sizeof(struct cudbg_tp_la) + TPLA_SIZE * sizeof(u64); break; case CUDBG_MEMINFO: - len = sizeof(struct cudbg_meminfo); + len = sizeof(struct cudbg_ver_hdr) + + sizeof(struct cudbg_meminfo); break; case CUDBG_CIM_PIF_LA: len = sizeof(struct cudbg_cim_pif_la); @@ -273,7 +274,8 @@ static u32 cxgb4_get_entity_length(struct adapter *adap, u32 entity) } break; case CUDBG_ULPTX_LA: - len = sizeof(struct cudbg_ulptx_la); + len = sizeof(struct cudbg_ver_hdr) + + sizeof(struct cudbg_ulptx_la); break; case CUDBG_UP_CIM_INDIRECT: n = 0; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index c301aaf79d64..6f312e03432f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -2414,6 +2414,44 @@ static const struct file_operations rss_vf_config_debugfs_fops = { .release = seq_release_private }; +static int resources_show(struct seq_file *seq, void *v) +{ + struct adapter *adapter = seq->private; + struct pf_resources *pfres = &adapter->params.pfres; + + #define S(desc, fmt, var) \ + seq_printf(seq, "%-60s " fmt "\n", \ + desc " (" #var "):", pfres->var) + + S("Virtual Interfaces", "%d", nvi); + S("Egress Queues", "%d", neq); + S("Ethernet Control", "%d", nethctrl); + S("Ingress Queues/w Free Lists/Interrupts", "%d", niqflint); + S("Ingress Queues", "%d", niq); + S("Traffic Class", "%d", tc); + S("Port Access Rights Mask", "%#x", pmask); + S("MAC Address Filters", "%d", nexactf); + S("Firmware Command Read Capabilities", "%#x", r_caps); + S("Firmware Command Write/Execute Capabilities", "%#x", wx_caps); + + #undef S + + return 0; +} + +static int resources_open(struct inode *inode, struct file *file) +{ + return single_open(file, resources_show, inode->i_private); +} + +static const struct file_operations resources_debugfs_fops = { + .owner = THIS_MODULE, + .open = resources_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + /** * ethqset2pinfo - return port_info of an Ethernet Queue Set * @adap: the adapter @@ -2436,16 +2474,64 @@ static inline struct port_info *ethqset2pinfo(struct adapter *adap, int qset) return NULL; } +static int sge_qinfo_uld_txq_entries(const struct adapter *adap, int uld) +{ + const struct sge_uld_txq_info *utxq_info = adap->sge.uld_txq_info[uld]; + + if (!utxq_info) + return 0; + + return DIV_ROUND_UP(utxq_info->ntxq, 4); +} + +static int sge_qinfo_uld_rspq_entries(const struct adapter *adap, int uld, + bool ciq) +{ + const struct sge_uld_rxq_info *urxq_info = adap->sge.uld_rxq_info[uld]; + + if (!urxq_info) + return 0; + + return ciq ? DIV_ROUND_UP(urxq_info->nciq, 4) : + DIV_ROUND_UP(urxq_info->nrxq, 4); +} + +static int sge_qinfo_uld_rxq_entries(const struct adapter *adap, int uld) +{ + return sge_qinfo_uld_rspq_entries(adap, uld, false); +} + +static int sge_qinfo_uld_ciq_entries(const struct adapter *adap, int uld) +{ + return sge_qinfo_uld_rspq_entries(adap, uld, true); +} + static int sge_qinfo_show(struct seq_file *seq, void *v) { + int uld_rxq_entries[CXGB4_ULD_MAX] = { 0 }; + int uld_ciq_entries[CXGB4_ULD_MAX] = { 0 }; + int uld_txq_entries[CXGB4_TX_MAX] = { 0 }; + const struct sge_uld_txq_info *utxq_info; + const struct sge_uld_rxq_info *urxq_info; struct adapter *adap = seq->private; - int eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4); - int ofld_entries = DIV_ROUND_UP(adap->sge.ofldqsets, 4); - int ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4); - int i, r = (uintptr_t)v - 1; - int ofld_idx = r - eth_entries; - int ctrl_idx = ofld_idx - ofld_entries; - int fq_idx = ctrl_idx - ctrl_entries; + int i, n, r = (uintptr_t)v - 1; + int eth_entries, ctrl_entries; + struct sge *s = &adap->sge; + + eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4); + ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4); + + mutex_lock(&uld_mutex); + if (s->uld_txq_info) + for (i = 0; i < ARRAY_SIZE(uld_txq_entries); i++) + uld_txq_entries[i] = sge_qinfo_uld_txq_entries(adap, i); + + if (s->uld_rxq_info) { + for (i = 0; i < ARRAY_SIZE(uld_rxq_entries); i++) { + uld_rxq_entries[i] = sge_qinfo_uld_rxq_entries(adap, i); + uld_ciq_entries[i] = sge_qinfo_uld_ciq_entries(adap, i); + } + } if (r) seq_putc(seq, '\n'); @@ -2467,9 +2553,10 @@ do { \ if (r < eth_entries) { int base_qset = r * 4; - const struct sge_eth_rxq *rx = &adap->sge.ethrxq[base_qset]; - const struct sge_eth_txq *tx = &adap->sge.ethtxq[base_qset]; - int n = min(4, adap->sge.ethqsets - 4 * r); + const struct sge_eth_rxq *rx = &s->ethrxq[base_qset]; + const struct sge_eth_txq *tx = &s->ethtxq[base_qset]; + + n = min(4, s->ethqsets - 4 * r); S("QType:", "Ethernet"); S("Interface:", @@ -2494,8 +2581,7 @@ do { \ R("RspQ CIDX:", rspq.cidx); R("RspQ Gen:", rspq.gen); S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); - S3("u", "Intr pktcnt:", - adap->sge.counter_val[rx[i].rspq.pktcnt_idx]); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); R("FL ID:", fl.cntxt_id); R("FL size:", fl.size - 8); R("FL pend:", fl.pend_cred); @@ -2520,9 +2606,196 @@ do { \ RL("FLLow:", fl.low); RL("FLStarving:", fl.starving); - } else if (ctrl_idx < ctrl_entries) { - const struct sge_ctrl_txq *tx = &adap->sge.ctrlq[ctrl_idx * 4]; - int n = min(4, adap->params.nports - 4 * ctrl_idx); + goto unlock; + } + + r -= eth_entries; + if (r < uld_txq_entries[CXGB4_TX_OFLD]) { + const struct sge_uld_txq *tx; + + utxq_info = s->uld_txq_info[CXGB4_TX_OFLD]; + tx = &utxq_info->uldtxq[r * 4]; + n = min(4, utxq_info->ntxq - 4 * r); + + S("QType:", "OFLD-TXQ"); + T("TxQ ID:", q.cntxt_id); + T("TxQ size:", q.size); + T("TxQ inuse:", q.in_use); + T("TxQ CIDX:", q.cidx); + T("TxQ PIDX:", q.pidx); + + goto unlock; + } + + r -= uld_txq_entries[CXGB4_TX_OFLD]; + if (r < uld_rxq_entries[CXGB4_ULD_RDMA]) { + const struct sge_ofld_rxq *rx; + + urxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA]; + rx = &urxq_info->uldrxq[r * 4]; + n = min(4, urxq_info->nrxq - 4 * r); + + S("QType:", "RDMA-CPL"); + S("Interface:", + rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A"); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + R("FL ID:", fl.cntxt_id); + R("FL size:", fl.size - 8); + R("FL pend:", fl.pend_cred); + R("FL avail:", fl.avail); + R("FL PIDX:", fl.pidx); + R("FL CIDX:", fl.cidx); + + goto unlock; + } + + r -= uld_rxq_entries[CXGB4_ULD_RDMA]; + if (r < uld_ciq_entries[CXGB4_ULD_RDMA]) { + const struct sge_ofld_rxq *rx; + int ciq_idx = 0; + + urxq_info = s->uld_rxq_info[CXGB4_ULD_RDMA]; + ciq_idx = urxq_info->nrxq + (r * 4); + rx = &urxq_info->uldrxq[ciq_idx]; + n = min(4, urxq_info->nciq - 4 * r); + + S("QType:", "RDMA-CIQ"); + S("Interface:", + rx[i].rspq.netdev ? rx[i].rspq.netdev->name : "N/A"); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + + goto unlock; + } + + r -= uld_ciq_entries[CXGB4_ULD_RDMA]; + if (r < uld_rxq_entries[CXGB4_ULD_ISCSI]) { + const struct sge_ofld_rxq *rx; + + urxq_info = s->uld_rxq_info[CXGB4_ULD_ISCSI]; + rx = &urxq_info->uldrxq[r * 4]; + n = min(4, urxq_info->nrxq - 4 * r); + + S("QType:", "iSCSI"); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + R("FL ID:", fl.cntxt_id); + R("FL size:", fl.size - 8); + R("FL pend:", fl.pend_cred); + R("FL avail:", fl.avail); + R("FL PIDX:", fl.pidx); + R("FL CIDX:", fl.cidx); + + goto unlock; + } + + r -= uld_rxq_entries[CXGB4_ULD_ISCSI]; + if (r < uld_rxq_entries[CXGB4_ULD_ISCSIT]) { + const struct sge_ofld_rxq *rx; + + urxq_info = s->uld_rxq_info[CXGB4_ULD_ISCSIT]; + rx = &urxq_info->uldrxq[r * 4]; + n = min(4, urxq_info->nrxq - 4 * r); + + S("QType:", "iSCSIT"); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + R("FL ID:", fl.cntxt_id); + R("FL size:", fl.size - 8); + R("FL pend:", fl.pend_cred); + R("FL avail:", fl.avail); + R("FL PIDX:", fl.pidx); + R("FL CIDX:", fl.cidx); + + goto unlock; + } + + r -= uld_rxq_entries[CXGB4_ULD_ISCSIT]; + if (r < uld_rxq_entries[CXGB4_ULD_TLS]) { + const struct sge_ofld_rxq *rx; + + urxq_info = s->uld_rxq_info[CXGB4_ULD_TLS]; + rx = &urxq_info->uldrxq[r * 4]; + n = min(4, urxq_info->nrxq - 4 * r); + + S("QType:", "TLS"); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + R("FL ID:", fl.cntxt_id); + R("FL size:", fl.size - 8); + R("FL pend:", fl.pend_cred); + R("FL avail:", fl.avail); + R("FL PIDX:", fl.pidx); + R("FL CIDX:", fl.cidx); + + goto unlock; + } + + r -= uld_rxq_entries[CXGB4_ULD_TLS]; + if (r < uld_txq_entries[CXGB4_TX_CRYPTO]) { + const struct sge_ofld_rxq *rx; + const struct sge_uld_txq *tx; + + utxq_info = s->uld_txq_info[CXGB4_TX_CRYPTO]; + urxq_info = s->uld_rxq_info[CXGB4_ULD_CRYPTO]; + tx = &utxq_info->uldtxq[r * 4]; + rx = &urxq_info->uldrxq[r * 4]; + n = min(4, utxq_info->ntxq - 4 * r); + + S("QType:", "Crypto"); + T("TxQ ID:", q.cntxt_id); + T("TxQ size:", q.size); + T("TxQ inuse:", q.in_use); + T("TxQ CIDX:", q.cidx); + T("TxQ PIDX:", q.pidx); + R("RspQ ID:", rspq.abs_id); + R("RspQ size:", rspq.size); + R("RspQE size:", rspq.iqe_len); + R("RspQ CIDX:", rspq.cidx); + R("RspQ Gen:", rspq.gen); + S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq)); + S3("u", "Intr pktcnt:", s->counter_val[rx[i].rspq.pktcnt_idx]); + R("FL ID:", fl.cntxt_id); + R("FL size:", fl.size - 8); + R("FL pend:", fl.pend_cred); + R("FL avail:", fl.avail); + R("FL PIDX:", fl.pidx); + R("FL CIDX:", fl.cidx); + + goto unlock; + } + + r -= uld_txq_entries[CXGB4_TX_CRYPTO]; + if (r < ctrl_entries) { + const struct sge_ctrl_txq *tx = &s->ctrlq[r * 4]; + + n = min(4, adap->params.nports - 4 * r); S("QType:", "Control"); T("TxQ ID:", q.cntxt_id); @@ -2532,8 +2805,13 @@ do { \ T("TxQ PIDX:", q.pidx); TL("TxQFull:", q.stops); TL("TxQRestarts:", q.restarts); - } else if (fq_idx == 0) { - const struct sge_rspq *evtq = &adap->sge.fw_evtq; + + goto unlock; + } + + r -= ctrl_entries; + if (r < 1) { + const struct sge_rspq *evtq = &s->fw_evtq; seq_printf(seq, "%-12s %16s\n", "QType:", "FW event queue"); seq_printf(seq, "%-12s %16u\n", "RspQ ID:", evtq->abs_id); @@ -2544,8 +2822,13 @@ do { \ seq_printf(seq, "%-12s %16u\n", "Intr delay:", qtimer_val(adap, evtq)); seq_printf(seq, "%-12s %16u\n", "Intr pktcnt:", - adap->sge.counter_val[evtq->pktcnt_idx]); + s->counter_val[evtq->pktcnt_idx]); + + goto unlock; } + +unlock: + mutex_unlock(&uld_mutex); #undef R #undef RL #undef T @@ -2559,8 +2842,21 @@ do { \ static int sge_queue_entries(const struct adapter *adap) { + int tot_uld_entries = 0; + int i; + + mutex_lock(&uld_mutex); + for (i = 0; i < CXGB4_TX_MAX; i++) + tot_uld_entries += sge_qinfo_uld_txq_entries(adap, i); + + for (i = 0; i < CXGB4_ULD_MAX; i++) { + tot_uld_entries += sge_qinfo_uld_rxq_entries(adap, i); + tot_uld_entries += sge_qinfo_uld_ciq_entries(adap, i); + } + mutex_unlock(&uld_mutex); + return DIV_ROUND_UP(adap->sge.ethqsets, 4) + - DIV_ROUND_UP(adap->sge.ofldqsets, 4) + + tot_uld_entries + DIV_ROUND_UP(MAX_CTRL_QUEUES, 4) + 1; } @@ -2851,15 +3147,17 @@ static int meminfo_show(struct seq_file *seq, void *v) mem_region_show(seq, "uP Extmem2:", meminfo.up_extmem2_lo, meminfo.up_extmem2_hi); - seq_printf(seq, "\n%u Rx pages of size %uKiB for %u channels\n", - meminfo.rx_pages_data[0], meminfo.rx_pages_data[1], - meminfo.rx_pages_data[2]); + seq_printf(seq, "\n%u Rx pages (%u free) of size %uKiB for %u channels\n", + meminfo.rx_pages_data[0], meminfo.free_rx_cnt, + meminfo.rx_pages_data[1], meminfo.rx_pages_data[2]); - seq_printf(seq, "%u Tx pages of size %u%ciB for %u channels\n", - meminfo.tx_pages_data[0], meminfo.tx_pages_data[1], - meminfo.tx_pages_data[2], meminfo.tx_pages_data[3]); + seq_printf(seq, "%u Tx pages (%u free) of size %u%ciB for %u channels\n", + meminfo.tx_pages_data[0], meminfo.free_tx_cnt, + meminfo.tx_pages_data[1], meminfo.tx_pages_data[2], + meminfo.tx_pages_data[3]); - seq_printf(seq, "%u p-structs\n\n", meminfo.p_structs); + seq_printf(seq, "%u p-structs (%u free)\n\n", + meminfo.p_structs, meminfo.p_structs_free_cnt); for (i = 0; i < 4; i++) /* For T6 these are MAC buffer groups */ @@ -2924,6 +3222,169 @@ static const struct file_operations chcr_stats_debugfs_fops = { .llseek = seq_lseek, .release = single_release, }; + +#define PRINT_ADAP_STATS(string, value) \ + seq_printf(seq, "%-25s %-20llu\n", (string), \ + (unsigned long long)(value)) + +#define PRINT_CH_STATS(string, value) \ +do { \ + seq_printf(seq, "%-25s ", (string)); \ + for (i = 0; i < adap->params.arch.nchan; i++) \ + seq_printf(seq, "%-20llu ", \ + (unsigned long long)stats.value[i]); \ + seq_printf(seq, "\n"); \ +} while (0) + +#define PRINT_CH_STATS2(string, value) \ +do { \ + seq_printf(seq, "%-25s ", (string)); \ + for (i = 0; i < adap->params.arch.nchan; i++) \ + seq_printf(seq, "%-20llu ", \ + (unsigned long long)stats[i].value); \ + seq_printf(seq, "\n"); \ +} while (0) + +static void show_tcp_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_tcp_stats v4, v6; + + spin_lock(&adap->stats_lock); + t4_tp_get_tcp_stats(adap, &v4, &v6, false); + spin_unlock(&adap->stats_lock); + + PRINT_ADAP_STATS("tcp_ipv4_out_rsts:", v4.tcp_out_rsts); + PRINT_ADAP_STATS("tcp_ipv4_in_segs:", v4.tcp_in_segs); + PRINT_ADAP_STATS("tcp_ipv4_out_segs:", v4.tcp_out_segs); + PRINT_ADAP_STATS("tcp_ipv4_retrans_segs:", v4.tcp_retrans_segs); + PRINT_ADAP_STATS("tcp_ipv6_out_rsts:", v6.tcp_out_rsts); + PRINT_ADAP_STATS("tcp_ipv6_in_segs:", v6.tcp_in_segs); + PRINT_ADAP_STATS("tcp_ipv6_out_segs:", v6.tcp_out_segs); + PRINT_ADAP_STATS("tcp_ipv6_retrans_segs:", v6.tcp_retrans_segs); +} + +static void show_ddp_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_usm_stats stats; + + spin_lock(&adap->stats_lock); + t4_get_usm_stats(adap, &stats, false); + spin_unlock(&adap->stats_lock); + + PRINT_ADAP_STATS("usm_ddp_frames:", stats.frames); + PRINT_ADAP_STATS("usm_ddp_octets:", stats.octets); + PRINT_ADAP_STATS("usm_ddp_drops:", stats.drops); +} + +static void show_rdma_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_rdma_stats stats; + + spin_lock(&adap->stats_lock); + t4_tp_get_rdma_stats(adap, &stats, false); + spin_unlock(&adap->stats_lock); + + PRINT_ADAP_STATS("rdma_no_rqe_mod_defer:", stats.rqe_dfr_mod); + PRINT_ADAP_STATS("rdma_no_rqe_pkt_defer:", stats.rqe_dfr_pkt); +} + +static void show_tp_err_adapter_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_err_stats stats; + + spin_lock(&adap->stats_lock); + t4_tp_get_err_stats(adap, &stats, false); + spin_unlock(&adap->stats_lock); + + PRINT_ADAP_STATS("tp_err_ofld_no_neigh:", stats.ofld_no_neigh); + PRINT_ADAP_STATS("tp_err_ofld_cong_defer:", stats.ofld_cong_defer); +} + +static void show_cpl_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_cpl_stats stats; + u8 i; + + spin_lock(&adap->stats_lock); + t4_tp_get_cpl_stats(adap, &stats, false); + spin_unlock(&adap->stats_lock); + + PRINT_CH_STATS("tp_cpl_requests:", req); + PRINT_CH_STATS("tp_cpl_responses:", rsp); +} + +static void show_tp_err_channel_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_err_stats stats; + u8 i; + + spin_lock(&adap->stats_lock); + t4_tp_get_err_stats(adap, &stats, false); + spin_unlock(&adap->stats_lock); + + PRINT_CH_STATS("tp_mac_in_errs:", mac_in_errs); + PRINT_CH_STATS("tp_hdr_in_errs:", hdr_in_errs); + PRINT_CH_STATS("tp_tcp_in_errs:", tcp_in_errs); + PRINT_CH_STATS("tp_tcp6_in_errs:", tcp6_in_errs); + PRINT_CH_STATS("tp_tnl_cong_drops:", tnl_cong_drops); + PRINT_CH_STATS("tp_tnl_tx_drops:", tnl_tx_drops); + PRINT_CH_STATS("tp_ofld_vlan_drops:", ofld_vlan_drops); + PRINT_CH_STATS("tp_ofld_chan_drops:", ofld_chan_drops); +} + +static void show_fcoe_stats(struct seq_file *seq) +{ + struct adapter *adap = seq->private; + struct tp_fcoe_stats stats[NCHAN]; + u8 i; + + spin_lock(&adap->stats_lock); + for (i = 0; i < adap->params.arch.nchan; i++) + t4_get_fcoe_stats(adap, i, &stats[i], false); + spin_unlock(&adap->stats_lock); + + PRINT_CH_STATS2("fcoe_octets_ddp", octets_ddp); + PRINT_CH_STATS2("fcoe_frames_ddp", frames_ddp); + PRINT_CH_STATS2("fcoe_frames_drop", frames_drop); +} + +#undef PRINT_CH_STATS2 +#undef PRINT_CH_STATS +#undef PRINT_ADAP_STATS + +static int tp_stats_show(struct seq_file *seq, void *v) +{ + struct adapter *adap = seq->private; + + seq_puts(seq, "\n--------Adapter Stats--------\n"); + show_tcp_stats(seq); + show_ddp_stats(seq); + show_rdma_stats(seq); + show_tp_err_adapter_stats(seq); + + seq_puts(seq, "\n-------- Channel Stats --------\n"); + if (adap->params.arch.nchan == NCHAN) + seq_printf(seq, "%-25s %-20s %-20s %-20s %-20s\n", + " ", "channel 0", "channel 1", + "channel 2", "channel 3"); + else + seq_printf(seq, "%-25s %-20s %-20s\n", + " ", "channel 0", "channel 1"); + show_cpl_stats(seq); + show_tp_err_channel_stats(seq); + show_fcoe_stats(seq); + + return 0; +} + +DEFINE_SIMPLE_DEBUGFS_FILE(tp_stats); + /* Add an array of Debug FS files. */ void add_debugfs_files(struct adapter *adap, @@ -2973,6 +3434,7 @@ int t4_setup_debugfs(struct adapter *adap) { "rss_key", &rss_key_debugfs_fops, 0400, 0 }, { "rss_pf_config", &rss_pf_config_debugfs_fops, 0400, 0 }, { "rss_vf_config", &rss_vf_config_debugfs_fops, 0400, 0 }, + { "resources", &resources_debugfs_fops, 0400, 0 }, { "sge_qinfo", &sge_qinfo_debugfs_fops, 0400, 0 }, { "ibq_tp0", &cim_ibq_fops, 0400, 0 }, { "ibq_tp1", &cim_ibq_fops, 0400, 1 }, @@ -2999,6 +3461,7 @@ int t4_setup_debugfs(struct adapter *adap) { "blocked_fl", &blocked_fl_fops, 0600, 0 }, { "meminfo", &meminfo_fops, 0400, 0 }, { "crypto", &chcr_stats_debugfs_fops, 0400, 0 }, + { "tp_stats", &tp_stats_debugfs_fops, 0400, 0 }, }; /* Debug FS nodes common to all T5 and later adapters. diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index f7eef93ffc87..d07230c892a5 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -115,42 +115,10 @@ static char adapter_stats_strings[][ETH_GSTRING_LEN] = { "db_drop ", "db_full ", "db_empty ", - "tcp_ipv4_out_rsts ", - "tcp_ipv4_in_segs ", - "tcp_ipv4_out_segs ", - "tcp_ipv4_retrans_segs ", - "tcp_ipv6_out_rsts ", - "tcp_ipv6_in_segs ", - "tcp_ipv6_out_segs ", - "tcp_ipv6_retrans_segs ", - "usm_ddp_frames ", - "usm_ddp_octets ", - "usm_ddp_drops ", - "rdma_no_rqe_mod_defer ", - "rdma_no_rqe_pkt_defer ", - "tp_err_ofld_no_neigh ", - "tp_err_ofld_cong_defer ", "write_coal_success ", "write_coal_fail ", }; -static char channel_stats_strings[][ETH_GSTRING_LEN] = { - "--------Channel--------- ", - "tp_cpl_requests ", - "tp_cpl_responses ", - "tp_mac_in_errs ", - "tp_hdr_in_errs ", - "tp_tcp_in_errs ", - "tp_tcp6_in_errs ", - "tp_tnl_cong_drops ", - "tp_tnl_tx_drops ", - "tp_ofld_vlan_drops ", - "tp_ofld_chan_drops ", - "fcoe_octets_ddp ", - "fcoe_frames_ddp ", - "fcoe_frames_drop ", -}; - static char loopback_stats_strings[][ETH_GSTRING_LEN] = { "-------Loopback----------- ", "octets_ok ", @@ -177,14 +145,19 @@ static char loopback_stats_strings[][ETH_GSTRING_LEN] = { "bg3_frames_trunc ", }; +static const char cxgb4_priv_flags_strings[][ETH_GSTRING_LEN] = { + [PRIV_FLAG_PORT_TX_VM_BIT] = "port_tx_vm_wr", +}; + static int get_sset_count(struct net_device *dev, int sset) { switch (sset) { case ETH_SS_STATS: return ARRAY_SIZE(stats_strings) + ARRAY_SIZE(adapter_stats_strings) + - ARRAY_SIZE(channel_stats_strings) + ARRAY_SIZE(loopback_stats_strings); + case ETH_SS_PRIV_FLAGS: + return ARRAY_SIZE(cxgb4_priv_flags_strings); default: return -EOPNOTSUPP; } @@ -235,6 +208,7 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) FW_HDR_FW_VER_MINOR_G(exprom_vers), FW_HDR_FW_VER_MICRO_G(exprom_vers), FW_HDR_FW_VER_BUILD_G(exprom_vers)); + info->n_priv_flags = ARRAY_SIZE(cxgb4_priv_flags_strings); } static void get_strings(struct net_device *dev, u32 stringset, u8 *data) @@ -245,11 +219,11 @@ static void get_strings(struct net_device *dev, u32 stringset, u8 *data) memcpy(data, adapter_stats_strings, sizeof(adapter_stats_strings)); data += sizeof(adapter_stats_strings); - memcpy(data, channel_stats_strings, - sizeof(channel_stats_strings)); - data += sizeof(channel_stats_strings); memcpy(data, loopback_stats_strings, sizeof(loopback_stats_strings)); + } else if (stringset == ETH_SS_PRIV_FLAGS) { + memcpy(data, cxgb4_priv_flags_strings, + sizeof(cxgb4_priv_flags_strings)); } } @@ -270,41 +244,10 @@ struct adapter_stats { u64 db_drop; u64 db_full; u64 db_empty; - u64 tcp_v4_out_rsts; - u64 tcp_v4_in_segs; - u64 tcp_v4_out_segs; - u64 tcp_v4_retrans_segs; - u64 tcp_v6_out_rsts; - u64 tcp_v6_in_segs; - u64 tcp_v6_out_segs; - u64 tcp_v6_retrans_segs; - u64 frames; - u64 octets; - u64 drops; - u64 rqe_dfr_mod; - u64 rqe_dfr_pkt; - u64 ofld_no_neigh; - u64 ofld_cong_defer; u64 wc_success; u64 wc_fail; }; -struct channel_stats { - u64 cpl_req; - u64 cpl_rsp; - u64 mac_in_errs; - u64 hdr_in_errs; - u64 tcp_in_errs; - u64 tcp6_in_errs; - u64 tnl_cong_drops; - u64 tnl_tx_drops; - u64 ofld_vlan_drops; - u64 ofld_chan_drops; - u64 octets_ddp; - u64 frames_ddp; - u64 frames_drop; -}; - static void collect_sge_port_stats(const struct adapter *adap, const struct port_info *p, struct queue_port_stats *s) @@ -327,45 +270,14 @@ static void collect_sge_port_stats(const struct adapter *adap, static void collect_adapter_stats(struct adapter *adap, struct adapter_stats *s) { - struct tp_tcp_stats v4, v6; - struct tp_rdma_stats rdma_stats; - struct tp_err_stats err_stats; - struct tp_usm_stats usm_stats; u64 val1, val2; memset(s, 0, sizeof(*s)); - spin_lock(&adap->stats_lock); - t4_tp_get_tcp_stats(adap, &v4, &v6, false); - t4_tp_get_rdma_stats(adap, &rdma_stats, false); - t4_get_usm_stats(adap, &usm_stats, false); - t4_tp_get_err_stats(adap, &err_stats, false); - spin_unlock(&adap->stats_lock); - s->db_drop = adap->db_stats.db_drop; s->db_full = adap->db_stats.db_full; s->db_empty = adap->db_stats.db_empty; - s->tcp_v4_out_rsts = v4.tcp_out_rsts; - s->tcp_v4_in_segs = v4.tcp_in_segs; - s->tcp_v4_out_segs = v4.tcp_out_segs; - s->tcp_v4_retrans_segs = v4.tcp_retrans_segs; - s->tcp_v6_out_rsts = v6.tcp_out_rsts; - s->tcp_v6_in_segs = v6.tcp_in_segs; - s->tcp_v6_out_segs = v6.tcp_out_segs; - s->tcp_v6_retrans_segs = v6.tcp_retrans_segs; - - if (is_offload(adap)) { - s->frames = usm_stats.frames; - s->octets = usm_stats.octets; - s->drops = usm_stats.drops; - s->rqe_dfr_mod = rdma_stats.rqe_dfr_mod; - s->rqe_dfr_pkt = rdma_stats.rqe_dfr_pkt; - } - - s->ofld_no_neigh = err_stats.ofld_no_neigh; - s->ofld_cong_defer = err_stats.ofld_cong_defer; - if (!is_t4(adap->params.chip)) { int v; @@ -379,36 +291,6 @@ static void collect_adapter_stats(struct adapter *adap, struct adapter_stats *s) } } -static void collect_channel_stats(struct adapter *adap, struct channel_stats *s, - u8 i) -{ - struct tp_cpl_stats cpl_stats; - struct tp_err_stats err_stats; - struct tp_fcoe_stats fcoe_stats; - - memset(s, 0, sizeof(*s)); - - spin_lock(&adap->stats_lock); - t4_tp_get_cpl_stats(adap, &cpl_stats, false); - t4_tp_get_err_stats(adap, &err_stats, false); - t4_get_fcoe_stats(adap, i, &fcoe_stats, false); - spin_unlock(&adap->stats_lock); - - s->cpl_req = cpl_stats.req[i]; - s->cpl_rsp = cpl_stats.rsp[i]; - s->mac_in_errs = err_stats.mac_in_errs[i]; - s->hdr_in_errs = err_stats.hdr_in_errs[i]; - s->tcp_in_errs = err_stats.tcp_in_errs[i]; - s->tcp6_in_errs = err_stats.tcp6_in_errs[i]; - s->tnl_cong_drops = err_stats.tnl_cong_drops[i]; - s->tnl_tx_drops = err_stats.tnl_tx_drops[i]; - s->ofld_vlan_drops = err_stats.ofld_vlan_drops[i]; - s->ofld_chan_drops = err_stats.ofld_chan_drops[i]; - s->octets_ddp = fcoe_stats.octets_ddp; - s->frames_ddp = fcoe_stats.frames_ddp; - s->frames_drop = fcoe_stats.frames_drop; -} - static void get_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data) { @@ -429,11 +311,6 @@ static void get_stats(struct net_device *dev, struct ethtool_stats *stats, data += sizeof(struct adapter_stats) / sizeof(u64); *data++ = (u64)pi->port_id; - collect_channel_stats(adapter, (struct channel_stats *)data, - pi->port_id); - data += sizeof(struct channel_stats) / sizeof(u64); - - *data++ = (u64)pi->port_id; memset(&s, 0, sizeof(s)); t4_get_lb_stats(adapter, pi->port_id, &s); @@ -751,13 +628,10 @@ static int get_link_ksettings(struct net_device *dev, fw_caps_to_lmm(pi->port_type, pi->link_cfg.lpacaps, link_ksettings->link_modes.lp_advertising); - if (netif_carrier_ok(dev)) { - base->speed = pi->link_cfg.speed; - base->duplex = DUPLEX_FULL; - } else { - base->speed = SPEED_UNKNOWN; - base->duplex = DUPLEX_UNKNOWN; - } + base->speed = (netif_carrier_ok(dev) + ? pi->link_cfg.speed + : SPEED_UNKNOWN); + base->duplex = DUPLEX_FULL; if (pi->link_cfg.fc & PAUSE_RX) { if (pi->link_cfg.fc & PAUSE_TX) { @@ -1499,6 +1373,36 @@ static int cxgb4_get_module_eeprom(struct net_device *dev, offset, len, &data[eprom->len - len]); } +static u32 cxgb4_get_priv_flags(struct net_device *netdev) +{ + struct port_info *pi = netdev_priv(netdev); + struct adapter *adapter = pi->adapter; + + return (adapter->eth_flags | pi->eth_flags); +} + +/** + * set_flags - set/unset specified flags if passed in new_flags + * @cur_flags: pointer to current flags + * @new_flags: new incoming flags + * @flags: set of flags to set/unset + */ +static inline void set_flags(u32 *cur_flags, u32 new_flags, u32 flags) +{ + *cur_flags = (*cur_flags & ~flags) | (new_flags & flags); +} + +static int cxgb4_set_priv_flags(struct net_device *netdev, u32 flags) +{ + struct port_info *pi = netdev_priv(netdev); + struct adapter *adapter = pi->adapter; + + set_flags(&adapter->eth_flags, flags, PRIV_FLAGS_ADAP); + set_flags(&pi->eth_flags, flags, PRIV_FLAGS_PORT); + + return 0; +} + static const struct ethtool_ops cxgb_ethtool_ops = { .get_link_ksettings = get_link_ksettings, .set_link_ksettings = set_link_ksettings, @@ -1535,6 +1439,8 @@ static const struct ethtool_ops cxgb_ethtool_ops = { .get_dump_data = get_dump_data, .get_module_info = cxgb4_get_module_info, .get_module_eeprom = cxgb4_get_module_eeprom, + .get_priv_flags = cxgb4_get_priv_flags, + .set_priv_flags = cxgb4_set_priv_flags, }; void cxgb4_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index a8926e97935e..0f7ce71205e6 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -554,10 +554,9 @@ static int fwevtq_handler(struct sge_rspq *q, const __be64 *rsp, dev = q->adap->port[q->adap->chan_map[port]]; dcbxdis = (action == FW_PORT_ACTION_GET_PORT_INFO - ? !!(pcmd->u.info.dcbxdis_pkd & - FW_PORT_CMD_DCBXDIS_F) - : !!(pcmd->u.info32.lstatus32_to_cbllen32 & - FW_PORT_CMD_DCBXDIS32_F)); + ? !!(pcmd->u.info.dcbxdis_pkd & FW_PORT_CMD_DCBXDIS_F) + : !!(be32_to_cpu(pcmd->u.info32.lstatus32_to_cbllen32) + & FW_PORT_CMD_DCBXDIS32_F)); state_input = (dcbxdis ? CXGB4_DCB_INPUT_FW_DISABLED : CXGB4_DCB_INPUT_FW_ENABLED); @@ -924,12 +923,14 @@ static int setup_sge_queues(struct adapter *adap) QUEUENUMBER_V(s->ethrxq[0].rspq.abs_id)); return 0; freeout: + dev_err(adap->pdev_dev, "Can't allocate queues, err=%d\n", -err); t4_free_sge_resources(adap); return err; } static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv, select_queue_fallback_t fallback) + struct net_device *sb_dev, + select_queue_fallback_t fallback) { int txq; @@ -971,7 +972,7 @@ static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb, return txq; } - return fallback(dev, skb) % dev->real_num_tx_queues; + return fallback(dev, skb, NULL) % dev->real_num_tx_queues; } static int closest_timer(const struct sge *s, int time) @@ -3016,7 +3017,7 @@ static int cxgb_setup_tc_block(struct net_device *dev, switch (f->command) { case TC_BLOCK_BIND: return tcf_block_cb_register(f->block, cxgb_setup_tc_block_cb, - pi, dev); + pi, dev, f->extack); case TC_BLOCK_UNBIND: tcf_block_cb_unregister(f->block, cxgb_setup_tc_block_cb, pi); return 0; @@ -3219,7 +3220,7 @@ static netdev_features_t cxgb_fix_features(struct net_device *dev, static const struct net_device_ops cxgb4_netdev_ops = { .ndo_open = cxgb_open, .ndo_stop = cxgb_close, - .ndo_start_xmit = t4_eth_xmit, + .ndo_start_xmit = t4_start_xmit, .ndo_select_queue = cxgb_select_queue, .ndo_get_stats64 = cxgb_get_stats, .ndo_set_rx_mode = cxgb_set_rxmode, @@ -3538,6 +3539,16 @@ static int adap_init1(struct adapter *adap, struct fw_caps_config_cmd *c) u32 v; int ret; + /* Now that we've successfully configured and initialized the adapter + * can ask the Firmware what resources it has provisioned for us. + */ + ret = t4_get_pfres(adap); + if (ret) { + dev_err(adap->pdev_dev, + "Unable to retrieve resource provisioning information\n"); + return ret; + } + /* get device capabilities */ memset(c, 0, sizeof(*c)); c->op_to_write = htonl(FW_CMD_OP_V(FW_CAPS_CONFIG_CMD) | @@ -4172,32 +4183,6 @@ static int adap_init0(struct adapter *adap) goto bye; } - /* - * Grab VPD parameters. This should be done after we establish a - * connection to the firmware since some of the VPD parameters - * (notably the Core Clock frequency) are retrieved via requests to - * the firmware. On the other hand, we need these fairly early on - * so we do this right after getting ahold of the firmware. - */ - ret = t4_get_vpd_params(adap, &adap->params.vpd); - if (ret < 0) - goto bye; - - /* - * Find out what ports are available to us. Note that we need to do - * this before calling adap_init0_no_config() since it needs nports - * and portvec ... - */ - v = - FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | - FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC); - ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &v, &port_vec); - if (ret < 0) - goto bye; - - adap->params.nports = hweight32(port_vec); - adap->params.portvec = port_vec; - /* If the firmware is initialized already, emit a simply note to that * effect. Otherwise, it's time to try initializing the adapter. */ @@ -4248,6 +4233,45 @@ static int adap_init0(struct adapter *adap) } } + /* Now that we've successfully configured and initialized the adapter + * (or found it already initialized), we can ask the Firmware what + * resources it has provisioned for us. + */ + ret = t4_get_pfres(adap); + if (ret) { + dev_err(adap->pdev_dev, + "Unable to retrieve resource provisioning information\n"); + goto bye; + } + + /* Grab VPD parameters. This should be done after we establish a + * connection to the firmware since some of the VPD parameters + * (notably the Core Clock frequency) are retrieved via requests to + * the firmware. On the other hand, we need these fairly early on + * so we do this right after getting ahold of the firmware. + * + * We need to do this after initializing the adapter because someone + * could have FLASHed a new VPD which won't be read by the firmware + * until we do the RESET ... + */ + ret = t4_get_vpd_params(adap, &adap->params.vpd); + if (ret < 0) + goto bye; + + /* Find out what ports are available to us. Note that we need to do + * this before calling adap_init0_no_config() since it needs nports + * and portvec ... + */ + v = + FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_PORTVEC); + ret = t4_query_params(adap, adap->mbox, adap->pf, 0, 1, &v, &port_vec); + if (ret < 0) + goto bye; + + adap->params.nports = hweight32(port_vec); + adap->params.portvec = port_vec; + /* Give the SGE code a chance to pull in anything that it needs ... * Note that this must be called after we retrieve our VPD parameters * in order to know how to convert core ticks to seconds, etc. @@ -4799,10 +4823,12 @@ static inline bool is_x_10g_port(const struct link_config *lc) * of ports we found and the number of available CPUs. Most settings can be * modified by the admin prior to actual use. */ -static void cfg_queues(struct adapter *adap) +static int cfg_queues(struct adapter *adap) { struct sge *s = &adap->sge; - int i = 0, n10g = 0, qidx = 0; + int i, n10g = 0, qidx = 0; + int niqflint, neq, avail_eth_qsets; + int max_eth_qsets = 32; #ifndef CONFIG_CHELSIO_T4_DCB int q10g = 0; #endif @@ -4814,16 +4840,46 @@ static void cfg_queues(struct adapter *adap) adap->params.crypto = 0; } - n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg); + /* Calculate the number of Ethernet Queue Sets available based on + * resources provisioned for us. We always have an Asynchronous + * Firmware Event Ingress Queue. If we're operating in MSI or Legacy + * IRQ Pin Interrupt mode, then we'll also have a Forwarded Interrupt + * Ingress Queue. Meanwhile, we need two Egress Queues for each + * Queue Set: one for the Free List and one for the Ethernet TX Queue. + * + * Note that we should also take into account all of the various + * Offload Queues. But, in any situation where we're operating in + * a Resource Constrained Provisioning environment, doing any Offload + * at all is problematic ... + */ + niqflint = adap->params.pfres.niqflint - 1; + if (!(adap->flags & USING_MSIX)) + niqflint--; + neq = adap->params.pfres.neq / 2; + avail_eth_qsets = min(niqflint, neq); + + if (avail_eth_qsets > max_eth_qsets) + avail_eth_qsets = max_eth_qsets; + + if (avail_eth_qsets < adap->params.nports) { + dev_err(adap->pdev_dev, "avail_eth_qsets=%d < nports=%d\n", + avail_eth_qsets, adap->params.nports); + return -ENOMEM; + } + + /* Count the number of 10Gb/s or better ports */ + for_each_port(adap, i) + n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg); + #ifdef CONFIG_CHELSIO_T4_DCB /* For Data Center Bridging support we need to be able to support up * to 8 Traffic Priorities; each of which will be assigned to its * own TX Queue in order to prevent Head-Of-Line Blocking. */ - if (adap->params.nports * 8 > MAX_ETH_QSETS) { - dev_err(adap->pdev_dev, "MAX_ETH_QSETS=%d < %d!\n", - MAX_ETH_QSETS, adap->params.nports * 8); - BUG_ON(1); + if (adap->params.nports * 8 > avail_eth_qsets) { + dev_err(adap->pdev_dev, "DCB avail_eth_qsets=%d < %d!\n", + avail_eth_qsets, adap->params.nports * 8); + return -ENOMEM; } for_each_port(adap, i) { @@ -4839,7 +4895,7 @@ static void cfg_queues(struct adapter *adap) * per 10G port. */ if (n10g) - q10g = (MAX_ETH_QSETS - (adap->params.nports - n10g)) / n10g; + q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g; if (q10g > netif_get_num_default_rss_queues()) q10g = netif_get_num_default_rss_queues(); @@ -4890,6 +4946,8 @@ static void cfg_queues(struct adapter *adap) init_rspq(adap, &s->fw_evtq, 0, 1, 1024, 64); init_rspq(adap, &s->intrq, 0, 1, 512, 64); + + return 0; } /* @@ -5630,10 +5688,15 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) } } + if (!(adapter->flags & FW_OK)) + goto fw_attach_fail; + /* Configure queues and allocate tables now, they can be needed as * soon as the first register_netdev completes. */ - cfg_queues(adapter); + err = cfg_queues(adapter); + if (err) + goto out_free_dev; adapter->smt = t4_init_smt(); if (!adapter->smt) { @@ -5705,7 +5768,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (t4_read_reg(adapter, LE_DB_CONFIG_A) & HASHEN_F) { u32 hash_base, hash_reg; - if (chip <= CHELSIO_T5) { + if (chip_ver <= CHELSIO_T5) { hash_reg = LE_DB_TID_HASHBASE_A; hash_base = t4_read_reg(adapter, hash_reg); adapter->tids.hash_base = hash_base / 4; @@ -5740,6 +5803,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto out_free_dev; } +fw_attach_fail: /* * The card is now ready to go. If any errors occur during device * registration we do not fail the whole card but rather proceed only diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index 3ddd2c4acf68..623f73dd7738 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -874,6 +874,9 @@ int cxgb4_init_tc_flower(struct adapter *adap) { int ret; + if (adap->tc_flower_initialized) + return -EEXIST; + adap->flower_ht_params = cxgb4_tc_flower_ht_params; ret = rhashtable_init(&adap->flower_tbl, &adap->flower_ht_params); if (ret) @@ -882,13 +885,18 @@ int cxgb4_init_tc_flower(struct adapter *adap) INIT_WORK(&adap->flower_stats_work, ch_flower_stats_handler); timer_setup(&adap->flower_stats_timer, ch_flower_stats_cb, 0); mod_timer(&adap->flower_stats_timer, jiffies + STATS_CHECK_PERIOD); + adap->tc_flower_initialized = true; return 0; } void cxgb4_cleanup_tc_flower(struct adapter *adap) { + if (!adap->tc_flower_initialized) + return; + if (adap->flower_stats_timer.function) del_timer_sync(&adap->flower_stats_timer); cancel_work_sync(&adap->flower_stats_work); rhashtable_destroy(&adap->flower_tbl); + adap->tc_flower_initialized = false; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.c b/drivers/net/ethernet/chelsio/cxgb4/sched.c index 9148abb7994c..7fc656680299 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sched.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sched.c @@ -539,6 +539,9 @@ void t4_cleanup_sched(struct adapter *adap) struct port_info *pi = netdev2pinfo(adap->port[j]); s = pi->sched_tbl; + if (!s) + continue; + for (i = 0; i < s->sched_size; i++) { struct sched_class *e; diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 395e2a0e8d7f..6807bc3a44fb 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -1288,13 +1288,13 @@ static inline void t6_fill_tnl_lso(struct sk_buff *skb, } /** - * t4_eth_xmit - add a packet to an Ethernet Tx queue + * cxgb4_eth_xmit - add a packet to an Ethernet Tx queue * @skb: the packet * @dev: the egress net device * * Add a packet to an SGE Ethernet Tx queue. Runs with softirqs disabled. */ -netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t cxgb4_eth_xmit(struct sk_buff *skb, struct net_device *dev) { u32 wr_mid, ctrl0, op; u64 cntrl, *end, *sgl; @@ -1547,6 +1547,374 @@ out_free: dev_kfree_skb_any(skb); return NETDEV_TX_OK; } +/* Constants ... */ +enum { + /* Egress Queue sizes, producer and consumer indices are all in units + * of Egress Context Units bytes. Note that as far as the hardware is + * concerned, the free list is an Egress Queue (the host produces free + * buffers which the hardware consumes) and free list entries are + * 64-bit PCI DMA addresses. + */ + EQ_UNIT = SGE_EQ_IDXSIZE, + FL_PER_EQ_UNIT = EQ_UNIT / sizeof(__be64), + TXD_PER_EQ_UNIT = EQ_UNIT / sizeof(__be64), + + T4VF_ETHTXQ_MAX_HDR = (sizeof(struct fw_eth_tx_pkt_vm_wr) + + sizeof(struct cpl_tx_pkt_lso_core) + + sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64), +}; + +/** + * t4vf_is_eth_imm - can an Ethernet packet be sent as immediate data? + * @skb: the packet + * + * Returns whether an Ethernet packet is small enough to fit completely as + * immediate data. + */ +static inline int t4vf_is_eth_imm(const struct sk_buff *skb) +{ + /* The VF Driver uses the FW_ETH_TX_PKT_VM_WR firmware Work Request + * which does not accommodate immediate data. We could dike out all + * of the support code for immediate data but that would tie our hands + * too much if we ever want to enhace the firmware. It would also + * create more differences between the PF and VF Drivers. + */ + return false; +} + +/** + * t4vf_calc_tx_flits - calculate the number of flits for a packet TX WR + * @skb: the packet + * + * Returns the number of flits needed for a TX Work Request for the + * given Ethernet packet, including the needed WR and CPL headers. + */ +static inline unsigned int t4vf_calc_tx_flits(const struct sk_buff *skb) +{ + unsigned int flits; + + /* If the skb is small enough, we can pump it out as a work request + * with only immediate data. In that case we just have to have the + * TX Packet header plus the skb data in the Work Request. + */ + if (t4vf_is_eth_imm(skb)) + return DIV_ROUND_UP(skb->len + sizeof(struct cpl_tx_pkt), + sizeof(__be64)); + + /* Otherwise, we're going to have to construct a Scatter gather list + * of the skb body and fragments. We also include the flits necessary + * for the TX Packet Work Request and CPL. We always have a firmware + * Write Header (incorporated as part of the cpl_tx_pkt_lso and + * cpl_tx_pkt structures), followed by either a TX Packet Write CPL + * message or, if we're doing a Large Send Offload, an LSO CPL message + * with an embedded TX Packet Write CPL message. + */ + flits = sgl_len(skb_shinfo(skb)->nr_frags + 1); + if (skb_shinfo(skb)->gso_size) + flits += (sizeof(struct fw_eth_tx_pkt_vm_wr) + + sizeof(struct cpl_tx_pkt_lso_core) + + sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64); + else + flits += (sizeof(struct fw_eth_tx_pkt_vm_wr) + + sizeof(struct cpl_tx_pkt_core)) / sizeof(__be64); + return flits; +} + +/** + * cxgb4_vf_eth_xmit - add a packet to an Ethernet TX queue + * @skb: the packet + * @dev: the egress net device + * + * Add a packet to an SGE Ethernet TX queue. Runs with softirqs disabled. + */ +static netdev_tx_t cxgb4_vf_eth_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + dma_addr_t addr[MAX_SKB_FRAGS + 1]; + const struct skb_shared_info *ssi; + struct fw_eth_tx_pkt_vm_wr *wr; + int qidx, credits, max_pkt_len; + struct cpl_tx_pkt_core *cpl; + const struct port_info *pi; + unsigned int flits, ndesc; + struct sge_eth_txq *txq; + struct adapter *adapter; + u64 cntrl, *end; + u32 wr_mid; + const size_t fw_hdr_copy_len = sizeof(wr->ethmacdst) + + sizeof(wr->ethmacsrc) + + sizeof(wr->ethtype) + + sizeof(wr->vlantci); + + /* The chip minimum packet length is 10 octets but the firmware + * command that we are using requires that we copy the Ethernet header + * (including the VLAN tag) into the header so we reject anything + * smaller than that ... + */ + 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. */ + pi = netdev_priv(dev); + adapter = pi->adapter; + qidx = skb_get_queue_mapping(skb); + WARN_ON(qidx >= pi->nqsets); + txq = &adapter->sge.ethtxq[pi->first_qset + qidx]; + + /* Take this opportunity to reclaim any TX Descriptors whose DMA + * transfers have completed. + */ + cxgb4_reclaim_completed_tx(adapter, &txq->q, true); + + /* Calculate the number of flits and TX Descriptors we're going to + * need along with how many TX Descriptors will be left over after + * we inject our Work Request. + */ + flits = t4vf_calc_tx_flits(skb); + ndesc = flits_to_desc(flits); + credits = txq_avail(&txq->q) - ndesc; + + if (unlikely(credits < 0)) { + /* Not enough room for this packet's Work Request. Stop the + * TX Queue and return a "busy" condition. The queue will get + * started later on when the firmware informs us that space + * has opened up. + */ + eth_txq_stop(txq); + dev_err(adapter->pdev_dev, + "%s: TX ring %u full while queue awake!\n", + dev->name, qidx); + return NETDEV_TX_BUSY; + } + + if (!t4vf_is_eth_imm(skb) && + unlikely(cxgb4_map_skb(adapter->pdev_dev, skb, addr) < 0)) { + /* We need to map the skb into PCI DMA space (because it can't + * be in-lined directly into the Work Request) and the mapping + * operation failed. Record the error and drop the packet. + */ + txq->mapping_err++; + goto out_free; + } + + wr_mid = FW_WR_LEN16_V(DIV_ROUND_UP(flits, 2)); + if (unlikely(credits < ETHTXQ_STOP_THRES)) { + /* After we're done injecting the Work Request for this + * packet, we'll be below our "stop threshold" so stop the TX + * Queue now and schedule a request for an SGE Egress Queue + * Update message. The queue will get started later on when + * the firmware processes this Work Request and sends us an + * Egress Queue Status Update message indicating that space + * has opened up. + */ + eth_txq_stop(txq); + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + } + + /* Start filling in our Work Request. Note that we do _not_ handle + * the WR Header wrapping around the TX Descriptor Ring. If our + * maximum header size ever exceeds one TX Descriptor, we'll need to + * do something else here. + */ + WARN_ON(DIV_ROUND_UP(T4VF_ETHTXQ_MAX_HDR, TXD_PER_EQ_UNIT) > 1); + wr = (void *)&txq->q.desc[txq->q.pidx]; + wr->equiq_to_len16 = cpu_to_be32(wr_mid); + wr->r3[0] = cpu_to_be32(0); + wr->r3[1] = cpu_to_be32(0); + skb_copy_from_linear_data(skb, (void *)wr->ethmacdst, fw_hdr_copy_len); + end = (u64 *)wr + flits; + + /* If this is a Large Send Offload packet we'll put in an LSO CPL + * message with an encapsulated TX Packet CPL message. Otherwise we + * just use a TX Packet CPL message. + */ + ssi = skb_shinfo(skb); + if (ssi->gso_size) { + struct cpl_tx_pkt_lso_core *lso = (void *)(wr + 1); + bool v6 = (ssi->gso_type & SKB_GSO_TCPV6) != 0; + int l3hdr_len = skb_network_header_len(skb); + int eth_xtra_len = skb_network_offset(skb) - ETH_HLEN; + + wr->op_immdlen = + cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_PKT_VM_WR) | + FW_WR_IMMDLEN_V(sizeof(*lso) + + sizeof(*cpl))); + /* Fill in the LSO CPL message. */ + lso->lso_ctrl = + 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_V(skb->len)); + + /* Set up TX Packet CPL pointer, control word and perform + * accounting. + */ + cpl = (void *)(lso + 1); + + if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5) + cntrl = TXPKT_ETHHDR_LEN_V(eth_xtra_len); + else + cntrl = T6_TXPKT_ETHHDR_LEN_V(eth_xtra_len); + + cntrl |= TXPKT_CSUM_TYPE_V(v6 ? + TX_CSUM_TCPIP6 : TX_CSUM_TCPIP) | + TXPKT_IPHDR_LEN_V(l3hdr_len); + txq->tso++; + txq->tx_cso += ssi->gso_segs; + } else { + int len; + + len = (t4vf_is_eth_imm(skb) + ? skb->len + sizeof(*cpl) + : sizeof(*cpl)); + wr->op_immdlen = + cpu_to_be32(FW_WR_OP_V(FW_ETH_TX_PKT_VM_WR) | + FW_WR_IMMDLEN_V(len)); + + /* Set up TX Packet CPL pointer, control word and perform + * accounting. + */ + cpl = (void *)(wr + 1); + if (skb->ip_summed == CHECKSUM_PARTIAL) { + cntrl = hwcsum(adapter->params.chip, skb) | + TXPKT_IPCSUM_DIS_F; + txq->tx_cso++; + } else { + cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F; + } + } + + /* If there's a VLAN tag present, add that to the list of things to + * do in this Work Request. + */ + if (skb_vlan_tag_present(skb)) { + txq->vlan_ins++; + 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_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); + + /* Fill in the body of the TX Packet CPL message with either in-lined + * data or a Scatter/Gather List. + */ + if (t4vf_is_eth_imm(skb)) { + /* In-line the packet's data and free the skb since we don't + * need it any longer. + */ + cxgb4_inline_tx_skb(skb, &txq->q, cpl + 1); + dev_consume_skb_any(skb); + } else { + /* Write the skb's Scatter/Gather list into the TX Packet CPL + * message and retain a pointer to the skb so we can free it + * later when its DMA completes. (We store the skb pointer + * in the Software Descriptor corresponding to the last TX + * Descriptor used by the Work Request.) + * + * The retained skb will be freed when the corresponding TX + * Descriptors are reclaimed after their DMAs complete. + * However, this could take quite a while since, in general, + * the hardware is set up to be lazy about sending DMA + * completion notifications to us and we mostly perform TX + * reclaims in the transmit routine. + * + * This is good for performamce but means that we rely on new + * TX packets arriving to run the destructors of completed + * packets, which open up space in their sockets' send queues. + * Sometimes we do not get such new packets causing TX to + * stall. A single UDP transmitter is a good example of this + * situation. We have a clean up timer that periodically + * reclaims completed packets but it doesn't run often enough + * (nor do we want it to) to prevent lengthy stalls. A + * solution to this problem is to run the destructor early, + * after the packet is queued but before it's DMAd. A con is + * that we lie to socket memory accounting, but the amount of + * extra memory is reasonable (limited by the number of TX + * descriptors), the packets do actually get freed quickly by + * new packets almost always, and for protocols like TCP that + * wait for acks to really free up the data the extra memory + * is even less. On the positive side we run the destructors + * on the sending CPU rather than on a potentially different + * completing CPU, usually a good thing. + * + * Run the destructor before telling the DMA engine about the + * packet to make sure it doesn't complete and get freed + * prematurely. + */ + struct ulptx_sgl *sgl = (struct ulptx_sgl *)(cpl + 1); + struct sge_txq *tq = &txq->q; + int last_desc; + + /* If the Work Request header was an exact multiple of our TX + * Descriptor length, then it's possible that the starting SGL + * pointer lines up exactly with the end of our TX Descriptor + * ring. If that's the case, wrap around to the beginning + * here ... + */ + if (unlikely((void *)sgl == (void *)tq->stat)) { + sgl = (void *)tq->desc; + end = (void *)((void *)tq->desc + + ((void *)end - (void *)tq->stat)); + } + + cxgb4_write_sgl(skb, tq, sgl, end, 0, addr); + skb_orphan(skb); + + last_desc = tq->pidx + ndesc - 1; + if (last_desc >= tq->size) + last_desc -= tq->size; + tq->sdesc[last_desc].skb = skb; + tq->sdesc[last_desc].sgl = sgl; + } + + /* Advance our internal TX Queue state, tell the hardware about + * the new TX descriptors and return success. + */ + txq_advance(&txq->q, ndesc); + + cxgb4_ring_tx_db(adapter, &txq->q, ndesc); + return NETDEV_TX_OK; + +out_free: + /* An error of some sort happened. Free the TX skb and tell the + * OS that we've "dealt" with the packet ... + */ + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +} + +netdev_tx_t t4_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct port_info *pi = netdev_priv(dev); + + if (unlikely(pi->eth_flags & PRIV_FLAG_PORT_TX_VM)) + return cxgb4_vf_eth_xmit(skb, dev); + + return cxgb4_eth_xmit(skb, dev); +} + /** * reclaim_completed_tx_imm - reclaim completed control-queue Tx descs * @q: the SGE control Tx queue @@ -3044,7 +3412,9 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq, 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); + c.iqns_to_fl0congen = htonl(FW_IQ_CMD_IQFLINTCONGEN_F | + FW_IQ_CMD_IQTYPE_V(cong ? FW_IQ_IQTYPE_NIC + : FW_IQ_IQTYPE_OFLD)); if (fl) { enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip); diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index 3720c3e11ebb..2d9943f90a75 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -2882,6 +2882,57 @@ int t4_get_vpd_params(struct adapter *adapter, struct vpd_params *p) return 0; } +/** + * t4_get_pfres - retrieve VF resource limits + * @adapter: the adapter + * + * Retrieves configured resource limits and capabilities for a physical + * function. The results are stored in @adapter->pfres. + */ +int t4_get_pfres(struct adapter *adapter) +{ + struct pf_resources *pfres = &adapter->params.pfres; + struct fw_pfvf_cmd cmd, rpl; + int v; + u32 word; + + /* Execute PFVF Read command to get VF resource limits; bail out early + * with error on command failure. + */ + memset(&cmd, 0, sizeof(cmd)); + cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_PFVF_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_READ_F | + FW_PFVF_CMD_PFN_V(adapter->pf) | + FW_PFVF_CMD_VFN_V(0)); + cmd.retval_len16 = cpu_to_be32(FW_LEN16(cmd)); + v = t4_wr_mbox(adapter, adapter->mbox, &cmd, sizeof(cmd), &rpl); + if (v != FW_SUCCESS) + return v; + + /* Extract PF resource limits and return success. + */ + word = be32_to_cpu(rpl.niqflint_niq); + pfres->niqflint = FW_PFVF_CMD_NIQFLINT_G(word); + pfres->niq = FW_PFVF_CMD_NIQ_G(word); + + word = be32_to_cpu(rpl.type_to_neq); + pfres->neq = FW_PFVF_CMD_NEQ_G(word); + pfres->pmask = FW_PFVF_CMD_PMASK_G(word); + + word = be32_to_cpu(rpl.tc_to_nexactf); + pfres->tc = FW_PFVF_CMD_TC_G(word); + pfres->nvi = FW_PFVF_CMD_NVI_G(word); + pfres->nexactf = FW_PFVF_CMD_NEXACTF_G(word); + + word = be32_to_cpu(rpl.r_caps_to_nethctrl); + pfres->r_caps = FW_PFVF_CMD_R_CAPS_G(word); + pfres->wx_caps = FW_PFVF_CMD_WX_CAPS_G(word); + pfres->nethctrl = FW_PFVF_CMD_NETHCTRL_G(word); + + return 0; +} + /* serial flash and firmware constants */ enum { SF_ATTEMPTS = 10, /* max retries for SF operations */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h index c7f8d0441278..e3adf435913e 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h @@ -188,6 +188,7 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN CH_PCI_ID_TABLE_FENTRY(0x50ab), /* Custom T520-CR */ CH_PCI_ID_TABLE_FENTRY(0x50ac), /* Custom T540-BT */ CH_PCI_ID_TABLE_FENTRY(0x50ad), /* Custom T520-CR */ + CH_PCI_ID_TABLE_FENTRY(0x50ae), /* Custom T540-XL-SO */ /* T6 adapters: */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h index 6b55aa2eb2a5..eb222d40ddbf 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h @@ -1502,6 +1502,25 @@ #define TP_MIB_DATA_A 0x7e54 #define TP_INT_CAUSE_A 0x7e74 +#define TP_FLM_FREE_PS_CNT_A 0x7e80 +#define TP_FLM_FREE_RX_CNT_A 0x7e84 + +#define FREEPSTRUCTCOUNT_S 0 +#define FREEPSTRUCTCOUNT_M 0x1fffffU +#define FREEPSTRUCTCOUNT_G(x) (((x) >> FREEPSTRUCTCOUNT_S) & FREEPSTRUCTCOUNT_M) + +#define FREERXPAGECOUNT_S 0 +#define FREERXPAGECOUNT_M 0x1fffffU +#define FREERXPAGECOUNT_V(x) ((x) << FREERXPAGECOUNT_S) +#define FREERXPAGECOUNT_G(x) (((x) >> FREERXPAGECOUNT_S) & FREERXPAGECOUNT_M) + +#define TP_FLM_FREE_TX_CNT_A 0x7e88 + +#define FREETXPAGECOUNT_S 0 +#define FREETXPAGECOUNT_M 0x1fffffU +#define FREETXPAGECOUNT_V(x) ((x) << FREETXPAGECOUNT_S) +#define FREETXPAGECOUNT_G(x) (((x) >> FREETXPAGECOUNT_S) & FREETXPAGECOUNT_M) + #define FLMTXFLSTEMPTY_S 30 #define FLMTXFLSTEMPTY_V(x) ((x) << FLMTXFLSTEMPTY_S) #define FLMTXFLSTEMPTY_F FLMTXFLSTEMPTY_V(1U) @@ -1683,6 +1702,16 @@ #define ULP_TX_LA_RDPTR_0_A 0x8ec0 #define ULP_TX_LA_RDDATA_0_A 0x8ec4 #define ULP_TX_LA_WRPTR_0_A 0x8ec8 +#define ULP_TX_ASIC_DEBUG_CTRL_A 0x8f70 + +#define ULP_TX_ASIC_DEBUG_0_A 0x8f74 +#define ULP_TX_ASIC_DEBUG_1_A 0x8f78 +#define ULP_TX_ASIC_DEBUG_2_A 0x8f7c +#define ULP_TX_ASIC_DEBUG_3_A 0x8f80 +#define ULP_TX_ASIC_DEBUG_4_A 0x8f84 + +/* registers for module PM_RX */ +#define PM_RX_BASE_ADDR 0x8fc0 #define PMRX_E_PCMD_PAR_ERROR_S 0 #define PMRX_E_PCMD_PAR_ERROR_V(x) ((x) << PMRX_E_PCMD_PAR_ERROR_S) diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index f1967cf6d43c..5dc6c4154af8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -1472,6 +1472,12 @@ enum fw_iq_type { FW_IQ_TYPE_NO_FL_INT_CAP }; +enum fw_iq_iqtype { + FW_IQ_IQTYPE_OTHER, + FW_IQ_IQTYPE_NIC, + FW_IQ_IQTYPE_OFLD, +}; + struct fw_iq_cmd { __be32 op_to_vfn; __be32 alloc_to_len16; @@ -1586,6 +1592,12 @@ struct fw_iq_cmd { #define FW_IQ_CMD_IQFLINTISCSIC_S 26 #define FW_IQ_CMD_IQFLINTISCSIC_V(x) ((x) << FW_IQ_CMD_IQFLINTISCSIC_S) +#define FW_IQ_CMD_IQTYPE_S 24 +#define FW_IQ_CMD_IQTYPE_M 0x3 +#define FW_IQ_CMD_IQTYPE_V(x) ((x) << FW_IQ_CMD_IQTYPE_S) +#define FW_IQ_CMD_IQTYPE_G(x) \ + (((x) >> FW_IQ_CMD_IQTYPE_S) & FW_IQ_CMD_IQTYPE_M) + #define FW_IQ_CMD_FL0CNGCHMAP_S 20 #define FW_IQ_CMD_FL0CNGCHMAP_V(x) ((x) << FW_IQ_CMD_FL0CNGCHMAP_S) diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c index 0ed161642371..74849be5f004 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c @@ -412,12 +412,10 @@ int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev, ppmax * (sizeof(struct cxgbi_ppod_data)) + ppod_bmap_size * sizeof(unsigned long); - ppm = vmalloc(alloc_sz); + ppm = vzalloc(alloc_sz); if (!ppm) goto release_ppm_pool; - memset(ppm, 0, alloc_sz); - ppm->ppod_bmap = (unsigned long *)(&ppm->ppod_data[ppmax]); if ((ppod_bmap_size >> 3) > (ppmax - ppmax_pool)) { |