IB/mlx5: Modify MAD reading counters method to use counter registers

Modify mlx5_ib_process_mad to use PPCNT and query_vport commands
instead of MAD_IFC, as MAD_IFC is deprecated on new firmware
versions (and doesn't support RoCE anyway).

Traffic counters exist in both 32-bit and 64-bit forms.
Declaring support of extended coutners results in traffic counters
to be read in their 64-bit form only via the query_vport command.
Error counters exist only in 32-bit form and read via PPCNT command.

This commit also adds counters support in RoCE.

Signed-off-by: Meny Yossefi <menyy@mellanox.com>
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Reviewed-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c
index b84d13a487cc04dcfbb501cfbd460c335bf91f8d..41d8a0036465eb8ccc12290a4054cab18ad64c29 100644 (file)
--- a/drivers/infiniband/hw/mlx5/mad.c
+++ b/drivers/infiniband/hw/mlx5/mad.c
@@ -31,8 +31,10 @@
  */
 
 #include <linux/mlx5/cmd.h>
+#include <linux/mlx5/vport.h>
 #include <rdma/ib_mad.h>
 #include <rdma/ib_smi.h>
+#include <rdma/ib_pma.h>
 #include "mlx5_ib.h"
 
 enum {
@@ -57,20 +59,12 @@ int mlx5_MAD_IFC(struct mlx5_ib_dev *dev, int ignore_mkey, int ignore_bkey,
        return mlx5_core_mad_ifc(dev->mdev, in_mad, response_mad, op_modifier, port);
 }
 
-int mlx5_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
-                       const struct ib_wc *in_wc, const struct ib_grh *in_grh,
-                       const struct ib_mad_hdr *in, size_t in_mad_size,
-                       struct ib_mad_hdr *out, size_t *out_mad_size,
-                       u16 *out_mad_pkey_index)
+static int process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
+                      const struct ib_wc *in_wc, const struct ib_grh *in_grh,
+                      const struct ib_mad *in_mad, struct ib_mad *out_mad)
 {
        u16 slid;
        int err;
-       const struct ib_mad *in_mad = (const struct ib_mad *)in;
-       struct ib_mad *out_mad = (struct ib_mad *)out;
-
-       if (WARN_ON_ONCE(in_mad_size != sizeof(*in_mad) ||
-                        *out_mad_size != sizeof(*out_mad)))
-               return IB_MAD_RESULT_FAILURE;
 
        slid = in_wc ? in_wc->slid : be16_to_cpu(IB_LID_PERMISSIVE);
 
@@ -117,6 +111,156 @@ int mlx5_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
        return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
 }
 
+static void pma_cnt_ext_assign(struct ib_pma_portcounters_ext *pma_cnt_ext,
+                              void *out)
+{
+#define MLX5_SUM_CNT(p, cntr1, cntr2)  \
+       (MLX5_GET64(query_vport_counter_out, p, cntr1) + \
+       MLX5_GET64(query_vport_counter_out, p, cntr2))
+
+       pma_cnt_ext->port_xmit_data =
+               cpu_to_be64(MLX5_SUM_CNT(out, transmitted_ib_unicast.octets,
+                                        transmitted_ib_multicast.octets) >> 2);
+       pma_cnt_ext->port_xmit_data =
+               cpu_to_be64(MLX5_SUM_CNT(out, received_ib_unicast.octets,
+                                        received_ib_multicast.octets) >> 2);
+       pma_cnt_ext->port_xmit_packets =
+               cpu_to_be64(MLX5_SUM_CNT(out, transmitted_ib_unicast.packets,
+                                        transmitted_ib_multicast.packets));
+       pma_cnt_ext->port_rcv_packets =
+               cpu_to_be64(MLX5_SUM_CNT(out, received_ib_unicast.packets,
+                                        received_ib_multicast.packets));
+       pma_cnt_ext->port_unicast_xmit_packets =
+               MLX5_GET64_BE(query_vport_counter_out,
+                             out, transmitted_ib_unicast.packets);
+       pma_cnt_ext->port_unicast_rcv_packets =
+               MLX5_GET64_BE(query_vport_counter_out,
+                             out, received_ib_unicast.packets);
+       pma_cnt_ext->port_multicast_xmit_packets =
+               MLX5_GET64_BE(query_vport_counter_out,
+                             out, transmitted_ib_multicast.packets);
+       pma_cnt_ext->port_multicast_rcv_packets =
+               MLX5_GET64_BE(query_vport_counter_out,
+                             out, received_ib_multicast.packets);
+}
+
+static void pma_cnt_assign(struct ib_pma_portcounters *pma_cnt,
+                          void *out)
+{
+       /* Traffic counters will be reported in
+        * their 64bit form via ib_pma_portcounters_ext by default.
+        */
+       void *out_pma = MLX5_ADDR_OF(ppcnt_reg, out,
+                                    counter_set);
+
+#define MLX5_ASSIGN_PMA_CNTR(counter_var, counter_name)        {               \
+       counter_var = MLX5_GET_BE(typeof(counter_var),                  \
+                                 ib_port_cntrs_grp_data_layout,        \
+                                 out_pma, counter_name);               \
+       }
+
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->symbol_error_counter,
+                            symbol_error_counter);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->link_error_recovery_counter,
+                            link_error_recovery_counter);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->link_downed_counter,
+                            link_downed_counter);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_rcv_errors,
+                            port_rcv_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_rcv_remphys_errors,
+                            port_rcv_remote_physical_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_rcv_switch_relay_errors,
+                            port_rcv_switch_relay_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_xmit_discards,
+                            port_xmit_discards);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_xmit_constraint_errors,
+                            port_xmit_constraint_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->port_rcv_constraint_errors,
+                            port_rcv_constraint_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->link_overrun_errors,
+                            link_overrun_errors);
+       MLX5_ASSIGN_PMA_CNTR(pma_cnt->vl15_dropped,
+                            vl_15_dropped);
+}
+
+static int process_pma_cmd(struct ib_device *ibdev, u8 port_num,
+                          const struct ib_mad *in_mad, struct ib_mad *out_mad)
+{
+       struct mlx5_ib_dev *dev = to_mdev(ibdev);
+       int err;
+       void *out_cnt;
+
+       /* Decalring support of extended counters */
+       if (in_mad->mad_hdr.attr_id == IB_PMA_CLASS_PORT_INFO) {
+               struct ib_class_port_info cpi = {};
+
+               cpi.capability_mask = IB_PMA_CLASS_CAP_EXT_WIDTH;
+               memcpy((out_mad->data + 40), &cpi, sizeof(cpi));
+               return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
+       }
+
+       if (in_mad->mad_hdr.attr_id == IB_PMA_PORT_COUNTERS_EXT) {
+               struct ib_pma_portcounters_ext *pma_cnt_ext =
+                       (struct ib_pma_portcounters_ext *)(out_mad->data + 40);
+               int sz = MLX5_ST_SZ_BYTES(query_vport_counter_out);
+
+               out_cnt = mlx5_vzalloc(sz);
+               if (!out_cnt)
+                       return IB_MAD_RESULT_FAILURE;
+
+               err = mlx5_core_query_vport_counter(dev->mdev, 0,
+                                                   port_num, out_cnt, sz);
+               if (!err)
+                       pma_cnt_ext_assign(pma_cnt_ext, out_cnt);
+       } else {
+               struct ib_pma_portcounters *pma_cnt =
+                       (struct ib_pma_portcounters *)(out_mad->data + 40);
+               int sz = MLX5_ST_SZ_BYTES(ppcnt_reg);
+
+               out_cnt = mlx5_vzalloc(sz);
+               if (!out_cnt)
+                       return IB_MAD_RESULT_FAILURE;
+
+               err = mlx5_core_query_ib_ppcnt(dev->mdev, port_num,
+                                              out_cnt, sz);
+               if (!err)
+                       pma_cnt_assign(pma_cnt, out_cnt);
+               }
+
+       kvfree(out_cnt);
+       if (err)
+               return IB_MAD_RESULT_FAILURE;
+
+       return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
+}
+
+int mlx5_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
+                       const struct ib_wc *in_wc, const struct ib_grh *in_grh,
+                       const struct ib_mad_hdr *in, size_t in_mad_size,
+                       struct ib_mad_hdr *out, size_t *out_mad_size,
+                       u16 *out_mad_pkey_index)
+{
+       struct mlx5_ib_dev *dev = to_mdev(ibdev);
+       struct mlx5_core_dev *mdev = dev->mdev;
+       const struct ib_mad *in_mad = (const struct ib_mad *)in;
+       struct ib_mad *out_mad = (struct ib_mad *)out;
+
+       if (WARN_ON_ONCE(in_mad_size != sizeof(*in_mad) ||
+                        *out_mad_size != sizeof(*out_mad)))
+               return IB_MAD_RESULT_FAILURE;
+
+       memset(out_mad->data, 0, sizeof(out_mad->data));
+
+       if (MLX5_CAP_GEN(mdev, vport_counters) &&
+           in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT &&
+           in_mad->mad_hdr.method == IB_MGMT_METHOD_GET) {
+               return process_pma_cmd(ibdev, port_num, in_mad, out_mad);
+       } else {
+               return process_mad(ibdev, mad_flags, port_num, in_wc, in_grh,
+                                  in_mad, out_mad);
+       }
+}
+
 int mlx5_query_ext_port_caps(struct mlx5_ib_dev *dev, u8 port)
 {
        struct ib_smp *in_mad  = NULL;

diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 99e2edc93d276f6c3891476cd918c77a67e5cb94..12079fd204131253a911b4ba4d092284fee89f81 100644 (file)
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -105,6 +105,29 @@ __mlx5_mask(typ, fld))
        ___t; \
 })
 
+/* Big endian getters */
+#define MLX5_GET64_BE(typ, p, fld) (*((__be64 *)(p) +\
+       __mlx5_64_off(typ, fld)))
+
+#define MLX5_GET_BE(type_t, typ, p, fld) ({                              \
+               type_t tmp;                                               \
+               switch (sizeof(tmp)) {                                    \
+               case sizeof(u8):                                          \
+                       tmp = (__force type_t)MLX5_GET(typ, p, fld);      \
+                       break;                                            \
+               case sizeof(u16):                                         \
+                       tmp = (__force type_t)cpu_to_be16(MLX5_GET(typ, p, fld)); \
+                       break;                                            \
+               case sizeof(u32):                                         \
+                       tmp = (__force type_t)cpu_to_be32(MLX5_GET(typ, p, fld)); \
+                       break;                                            \
+               case sizeof(u64):                                         \
+                       tmp = (__force type_t)MLX5_GET64_BE(typ, p, fld); \
+                       break;                                            \
+                       }                                                 \
+               tmp;                                                      \
+               })
+
 enum {
        MLX5_MAX_COMMANDS               = 32,
        MLX5_CMD_DATA_BLOCK_SIZE        = 512,