/*-
* BSD LICENSE
*
- * Copyright 2012-2015 6WIND S.A.
- * Copyright 2012 Mellanox.
+ * Copyright 2012 6WIND S.A.
+ * Copyright 2012 Mellanox
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
#ifndef RTE_PMD_MLX4_H_
#define RTE_PMD_MLX4_H_
-#include <stddef.h>
#include <stdint.h>
-#include <limits.h>
/*
- * Maximum number of simultaneous MAC addresses supported.
- *
- * According to ConnectX's Programmer Reference Manual:
- * The L2 Address Match is implemented by comparing a MAC/VLAN combination
- * of 128 MAC addresses and 127 VLAN values, comprising 128x127 possible
- * L2 addresses.
+ * Runtime logging through RTE_LOG() is enabled when not in debugging mode.
+ * Intermediate LOG_*() macros add the required end-of-line characters.
*/
-#define MLX4_MAX_MAC_ADDRESSES 128
+#ifndef NDEBUG
+#define INFO(...) DEBUG(__VA_ARGS__)
+#define WARN(...) DEBUG(__VA_ARGS__)
+#define ERROR(...) DEBUG(__VA_ARGS__)
+#else
+#define LOG__(level, m, ...) \
+ RTE_LOG(level, PMD, MLX4_DRIVER_NAME ": " m "%c", __VA_ARGS__)
+#define LOG_(level, ...) LOG__(level, __VA_ARGS__, '\n')
+#define INFO(...) LOG_(INFO, __VA_ARGS__)
+#define WARN(...) LOG_(WARNING, __VA_ARGS__)
+#define ERROR(...) LOG_(ERR, __VA_ARGS__)
+#endif
-/* Maximum number of simultaneous VLAN filters supported. See above. */
-#define MLX4_MAX_VLAN_IDS 127
+/* Verbs header. */
+/* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */
+#ifdef PEDANTIC
+#pragma GCC diagnostic ignored "-Wpedantic"
+#endif
+#include <infiniband/verbs.h>
+#ifdef PEDANTIC
+#pragma GCC diagnostic error "-Wpedantic"
+#endif
/* Request send completion once in every 64 sends, might be less. */
#define MLX4_PMD_TX_PER_COMP_REQ 64
-/* Maximum number of Scatter/Gather Elements per Work Request. */
-#ifndef MLX4_PMD_SGE_WR_N
-#define MLX4_PMD_SGE_WR_N 4
-#endif
-
/* Maximum size for inline data. */
-#ifndef MLX4_PMD_MAX_INLINE
#define MLX4_PMD_MAX_INLINE 0
-#endif
/*
* Maximum number of cached Memory Pools (MPs) per TX queue. Each RTE MP
#define MLX4_PMD_TX_MP_CACHE 8
#endif
-/*
- * If defined, only use software counters. The PMD will never ask the hardware
- * for these, and many of them won't be available.
- */
-#ifndef MLX4_PMD_SOFT_COUNTERS
-#define MLX4_PMD_SOFT_COUNTERS 1
-#endif
-
/* Alarm timeout. */
#define MLX4_ALARM_TIMEOUT_US 100000
+/* Port parameter. */
+#define MLX4_PMD_PORT_KVARG "port"
+
enum {
PCI_VENDOR_ID_MELLANOX = 0x15b3,
};
PCI_DEVICE_ID_MELLANOX_CONNECTX3PRO = 0x1007,
};
-#define MLX4_DRIVER_NAME "librte_pmd_mlx4"
-
-/* Bit-field manipulation. */
-#define BITFIELD_DECLARE(bf, type, size) \
- type bf[(((size_t)(size) / (sizeof(type) * CHAR_BIT)) + \
- !!((size_t)(size) % (sizeof(type) * CHAR_BIT)))]
-#define BITFIELD_DEFINE(bf, type, size) \
- BITFIELD_DECLARE((bf), type, (size)) = { 0 }
-#define BITFIELD_SET(bf, b) \
- (assert((size_t)(b) < (sizeof(bf) * CHAR_BIT)), \
- (void)((bf)[((b) / (sizeof((bf)[0]) * CHAR_BIT))] |= \
- ((size_t)1 << ((b) % (sizeof((bf)[0]) * CHAR_BIT)))))
-#define BITFIELD_RESET(bf, b) \
- (assert((size_t)(b) < (sizeof(bf) * CHAR_BIT)), \
- (void)((bf)[((b) / (sizeof((bf)[0]) * CHAR_BIT))] &= \
- ~((size_t)1 << ((b) % (sizeof((bf)[0]) * CHAR_BIT)))))
-#define BITFIELD_ISSET(bf, b) \
- (assert((size_t)(b) < (sizeof(bf) * CHAR_BIT)), \
- !!(((bf)[((b) / (sizeof((bf)[0]) * CHAR_BIT))] & \
- ((size_t)1 << ((b) % (sizeof((bf)[0]) * CHAR_BIT))))))
+#define MLX4_DRIVER_NAME "net_mlx4"
/* Number of elements in array. */
#define elemof(a) (sizeof(a) / sizeof((a)[0]))
-/* Cast pointer p to structure member m to its parent structure of type t. */
-#define containerof(p, t, m) ((t *)((uint8_t *)(p) - offsetof(t, m)))
-
-/* Branch prediction helpers. */
-#ifndef likely
-#define likely(c) __builtin_expect(!!(c), 1)
-#endif
-#ifndef unlikely
-#define unlikely(c) __builtin_expect(!!(c), 0)
-#endif
-
/* Debugging */
#ifndef NDEBUG
#include <stdio.h>
(DEBUG__(__VA_ARGS__), 0) \
})[0])
#define DEBUG(...) DEBUG_(__VA_ARGS__, '\n')
+#ifndef MLX4_PMD_DEBUG_BROKEN_VERBS
#define claim_zero(...) assert((__VA_ARGS__) == 0)
+#else /* MLX4_PMD_DEBUG_BROKEN_VERBS */
+#define claim_zero(...) \
+ (void)(((__VA_ARGS__) == 0) || \
+ DEBUG("Assertion `(" # __VA_ARGS__ ") == 0' failed (IGNORED)."))
+#endif /* MLX4_PMD_DEBUG_BROKEN_VERBS */
#define claim_nonzero(...) assert((__VA_ARGS__) != 0)
#define claim_positive(...) assert((__VA_ARGS__) >= 0)
#else /* NDEBUG */
#define claim_positive(...) (__VA_ARGS__)
#endif /* NDEBUG */
+struct mlx4_rxq_stats {
+ unsigned int idx; /**< Mapping index. */
+ uint64_t ipackets; /**< Total of successfully received packets. */
+ uint64_t ibytes; /**< Total of successfully received bytes. */
+ uint64_t idropped; /**< Total of packets dropped when RX ring full. */
+ uint64_t rx_nombuf; /**< Total of RX mbuf allocation failures. */
+};
+
+/* RX element. */
+struct rxq_elt {
+ struct ibv_recv_wr wr; /* Work Request. */
+ struct ibv_sge sge; /* Scatter/Gather Element. */
+ /* mbuf pointer is derived from WR_ID(wr.wr_id).offset. */
+};
+
+/* RX queue descriptor. */
+struct rxq {
+ struct priv *priv; /* Back pointer to private data. */
+ struct rte_mempool *mp; /* Memory Pool for allocations. */
+ struct ibv_mr *mr; /* Memory Region (for mp). */
+ struct ibv_cq *cq; /* Completion Queue. */
+ struct ibv_qp *qp; /* Queue Pair. */
+ struct ibv_exp_qp_burst_family *if_qp; /* QP burst interface. */
+ struct ibv_exp_cq_family *if_cq; /* CQ interface. */
+ struct ibv_comp_channel *channel;
+ unsigned int port_id; /* Port ID for incoming packets. */
+ unsigned int elts_n; /* (*elts)[] length. */
+ unsigned int elts_head; /* Current index in (*elts)[]. */
+ struct rxq_elt (*elts)[]; /* Rx elements. */
+ struct mlx4_rxq_stats stats; /* RX queue counters. */
+ unsigned int socket; /* CPU socket ID for allocations. */
+ struct ibv_exp_res_domain *rd; /* Resource Domain. */
+};
+
+/* TX element. */
+struct txq_elt {
+ struct rte_mbuf *buf;
+};
+
+struct mlx4_txq_stats {
+ unsigned int idx; /**< Mapping index. */
+ uint64_t opackets; /**< Total of successfully sent packets. */
+ uint64_t obytes; /**< Total of successfully sent bytes. */
+ uint64_t odropped; /**< Total of packets not sent when TX ring full. */
+};
+
+/* TX queue descriptor. */
+struct txq {
+ struct priv *priv; /* Back pointer to private data. */
+ struct {
+ const struct rte_mempool *mp; /* Cached Memory Pool. */
+ struct ibv_mr *mr; /* Memory Region (for mp). */
+ uint32_t lkey; /* mr->lkey */
+ } mp2mr[MLX4_PMD_TX_MP_CACHE]; /* MP to MR translation table. */
+ struct ibv_cq *cq; /* Completion Queue. */
+ struct ibv_qp *qp; /* Queue Pair. */
+ struct ibv_exp_qp_burst_family *if_qp; /* QP burst interface. */
+ struct ibv_exp_cq_family *if_cq; /* CQ interface. */
+ uint32_t max_inline; /* Max inline send size <= MLX4_PMD_MAX_INLINE. */
+ unsigned int elts_n; /* (*elts)[] length. */
+ struct txq_elt (*elts)[]; /* TX elements. */
+ unsigned int elts_head; /* Current index in (*elts)[]. */
+ unsigned int elts_tail; /* First element awaiting completion. */
+ unsigned int elts_comp; /* Number of completion requests. */
+ unsigned int elts_comp_cd; /* Countdown for next completion request. */
+ unsigned int elts_comp_cd_init; /* Initial value for countdown. */
+ struct mlx4_txq_stats stats; /* TX queue counters. */
+ unsigned int socket; /* CPU socket ID for allocations. */
+ struct ibv_exp_res_domain *rd; /* Resource Domain. */
+};
+
+struct rte_flow;
+
+struct priv {
+ struct rte_eth_dev *dev; /* Ethernet device. */
+ struct ibv_context *ctx; /* Verbs context. */
+ struct ibv_device_attr device_attr; /* Device properties. */
+ struct ibv_pd *pd; /* Protection Domain. */
+ struct ether_addr mac; /* MAC address. */
+ struct ibv_flow *mac_flow; /* Flow associated with MAC address. */
+ /* Device properties. */
+ uint16_t mtu; /* Configured MTU. */
+ uint8_t port; /* Physical port number. */
+ unsigned int started:1; /* Device started, flows enabled. */
+ unsigned int vf:1; /* This is a VF device. */
+ unsigned int pending_alarm:1; /* An alarm is pending. */
+ unsigned int isolated:1; /* Toggle isolated mode. */
+ /* RX/TX queues. */
+ unsigned int rxqs_n; /* RX queues array size. */
+ unsigned int txqs_n; /* TX queues array size. */
+ struct rxq *(*rxqs)[]; /* RX queues. */
+ struct txq *(*txqs)[]; /* TX queues. */
+ struct rte_intr_handle intr_handle_dev; /* Device interrupt handler. */
+ struct rte_intr_handle intr_handle; /* Interrupt handler. */
+ struct rte_flow_drop *flow_drop_queue; /* Flow drop queue. */
+ LIST_HEAD(mlx4_flows, rte_flow) flows;
+ struct rte_intr_conf intr_conf; /* Active interrupt configuration. */
+ rte_spinlock_t lock; /* Lock for control functions. */
+};
+
+void priv_lock(struct priv *priv);
+void priv_unlock(struct priv *priv);
+
#endif /* RTE_PMD_MLX4_H_ */