summaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/hw')
-rw-r--r--drivers/infiniband/hw/hfi1/init.c1
-rw-r--r--drivers/infiniband/hw/hfi1/pcie.c4
-rw-r--r--drivers/infiniband/hw/hfi1/rc.c16
-rw-r--r--drivers/infiniband/hw/hfi1/sdma.c4
-rw-r--r--drivers/infiniband/hw/hfi1/tid_rdma.c57
-rw-r--r--drivers/infiniband/hw/hfi1/tid_rdma.h3
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hem.h2
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_srq.c2
-rw-r--r--drivers/infiniband/hw/mlx5/ib_rep.c2
-rw-r--r--drivers/infiniband/hw/mlx5/ib_rep.h2
-rw-r--r--drivers/infiniband/hw/mlx5/main.c41
11 files changed, 71 insertions, 63 deletions
diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c
index 71cb9525c074..26b792bb1027 100644
--- a/drivers/infiniband/hw/hfi1/init.c
+++ b/drivers/infiniband/hw/hfi1/init.c
@@ -1489,7 +1489,6 @@ static int __init hfi1_mod_init(void)
goto bail_dev;
}
- hfi1_compute_tid_rdma_flow_wt();
/*
* These must be called before the driver is registered with
* the PCI subsystem.
diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c
index 61aa5504d7c3..61362bd6d3ce 100644
--- a/drivers/infiniband/hw/hfi1/pcie.c
+++ b/drivers/infiniband/hw/hfi1/pcie.c
@@ -319,7 +319,9 @@ int pcie_speeds(struct hfi1_devdata *dd)
/*
* bus->max_bus_speed is set from the bridge's linkcap Max Link Speed
*/
- if (parent && dd->pcidev->bus->max_bus_speed != PCIE_SPEED_8_0GT) {
+ if (parent &&
+ (dd->pcidev->bus->max_bus_speed == PCIE_SPEED_2_5GT ||
+ dd->pcidev->bus->max_bus_speed == PCIE_SPEED_5_0GT)) {
dd_dev_info(dd, "Parent PCIe bridge does not support Gen3\n");
dd->link_gen3_capable = 0;
}
diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c
index 513a8aac9ccd..1a3c647675a7 100644
--- a/drivers/infiniband/hw/hfi1/rc.c
+++ b/drivers/infiniband/hw/hfi1/rc.c
@@ -2209,15 +2209,15 @@ int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode,
if (qp->s_flags & RVT_S_WAIT_RNR)
goto bail_stop;
rdi = ib_to_rvt(qp->ibqp.device);
- if (qp->s_rnr_retry == 0 &&
- !((rdi->post_parms[wqe->wr.opcode].flags &
- RVT_OPERATION_IGN_RNR_CNT) &&
- qp->s_rnr_retry_cnt == 0)) {
- status = IB_WC_RNR_RETRY_EXC_ERR;
- goto class_b;
+ if (!(rdi->post_parms[wqe->wr.opcode].flags &
+ RVT_OPERATION_IGN_RNR_CNT)) {
+ if (qp->s_rnr_retry == 0) {
+ status = IB_WC_RNR_RETRY_EXC_ERR;
+ goto class_b;
+ }
+ if (qp->s_rnr_retry_cnt < 7 && qp->s_rnr_retry_cnt > 0)
+ qp->s_rnr_retry--;
}
- if (qp->s_rnr_retry_cnt < 7 && qp->s_rnr_retry_cnt > 0)
- qp->s_rnr_retry--;
/*
* The last valid PSN is the previous PSN. For TID RDMA WRITE
diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c
index c61b6022575e..5774dfc22e18 100644
--- a/drivers/infiniband/hw/hfi1/sdma.c
+++ b/drivers/infiniband/hw/hfi1/sdma.c
@@ -881,8 +881,8 @@ struct sdma_engine *sdma_select_user_engine(struct hfi1_devdata *dd,
cpu_id = smp_processor_id();
rcu_read_lock();
- rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu_id,
- sdma_rht_params);
+ rht_node = rhashtable_lookup(dd->sdma_rht, &cpu_id,
+ sdma_rht_params);
if (rht_node && rht_node->map[vl]) {
struct sdma_rht_map_elem *map = rht_node->map[vl];
diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c
index f21fca3617d5..e53f542b60af 100644
--- a/drivers/infiniband/hw/hfi1/tid_rdma.c
+++ b/drivers/infiniband/hw/hfi1/tid_rdma.c
@@ -107,8 +107,6 @@ static u32 mask_generation(u32 a)
* C - Capcode
*/
-static u32 tid_rdma_flow_wt;
-
static void tid_rdma_trigger_resume(struct work_struct *work);
static void hfi1_kern_exp_rcv_free_flows(struct tid_rdma_request *req);
static int hfi1_kern_exp_rcv_alloc_flows(struct tid_rdma_request *req,
@@ -136,6 +134,26 @@ static void update_r_next_psn_fecn(struct hfi1_packet *packet,
struct tid_rdma_flow *flow,
bool fecn);
+static void validate_r_tid_ack(struct hfi1_qp_priv *priv)
+{
+ if (priv->r_tid_ack == HFI1_QP_WQE_INVALID)
+ priv->r_tid_ack = priv->r_tid_tail;
+}
+
+static void tid_rdma_schedule_ack(struct rvt_qp *qp)
+{
+ struct hfi1_qp_priv *priv = qp->priv;
+
+ priv->s_flags |= RVT_S_ACK_PENDING;
+ hfi1_schedule_tid_send(qp);
+}
+
+static void tid_rdma_trigger_ack(struct rvt_qp *qp)
+{
+ validate_r_tid_ack(qp->priv);
+ tid_rdma_schedule_ack(qp);
+}
+
static u64 tid_rdma_opfn_encode(struct tid_rdma_params *p)
{
return
@@ -3005,10 +3023,7 @@ nak_psn:
qpriv->s_nak_state = IB_NAK_PSN_ERROR;
/* We are NAK'ing the next expected PSN */
qpriv->s_nak_psn = mask_psn(flow->flow_state.r_next_psn);
- qpriv->s_flags |= RVT_S_ACK_PENDING;
- if (qpriv->r_tid_ack == HFI1_QP_WQE_INVALID)
- qpriv->r_tid_ack = qpriv->r_tid_tail;
- hfi1_schedule_tid_send(qp);
+ tid_rdma_trigger_ack(qp);
}
goto unlock;
}
@@ -3371,18 +3386,17 @@ u32 hfi1_build_tid_rdma_write_req(struct rvt_qp *qp, struct rvt_swqe *wqe,
return sizeof(ohdr->u.tid_rdma.w_req) / sizeof(u32);
}
-void hfi1_compute_tid_rdma_flow_wt(void)
+static u32 hfi1_compute_tid_rdma_flow_wt(struct rvt_qp *qp)
{
/*
* Heuristic for computing the RNR timeout when waiting on the flow
* queue. Rather than a computationaly expensive exact estimate of when
* a flow will be available, we assume that if a QP is at position N in
* the flow queue it has to wait approximately (N + 1) * (number of
- * segments between two sync points), assuming PMTU of 4K. The rationale
- * for this is that flows are released and recycled at each sync point.
+ * segments between two sync points). The rationale for this is that
+ * flows are released and recycled at each sync point.
*/
- tid_rdma_flow_wt = MAX_TID_FLOW_PSN * enum_to_mtu(OPA_MTU_4096) /
- TID_RDMA_MAX_SEGMENT_SIZE;
+ return (MAX_TID_FLOW_PSN * qp->pmtu) >> TID_RDMA_SEGMENT_SHIFT;
}
static u32 position_in_queue(struct hfi1_qp_priv *qpriv,
@@ -3505,7 +3519,7 @@ static void hfi1_tid_write_alloc_resources(struct rvt_qp *qp, bool intr_ctx)
if (qpriv->flow_state.index >= RXE_NUM_TID_FLOWS) {
ret = hfi1_kern_setup_hw_flow(qpriv->rcd, qp);
if (ret) {
- to_seg = tid_rdma_flow_wt *
+ to_seg = hfi1_compute_tid_rdma_flow_wt(qp) *
position_in_queue(qpriv,
&rcd->flow_queue);
break;
@@ -3526,7 +3540,7 @@ static void hfi1_tid_write_alloc_resources(struct rvt_qp *qp, bool intr_ctx)
/*
* If overtaking req->acked_tail, send an RNR NAK. Because the
* QP is not queued in this case, and the issue can only be
- * caused due a delay in scheduling the second leg which we
+ * caused by a delay in scheduling the second leg which we
* cannot estimate, we use a rather arbitrary RNR timeout of
* (MAX_FLOWS / 2) segments
*/
@@ -3534,8 +3548,7 @@ static void hfi1_tid_write_alloc_resources(struct rvt_qp *qp, bool intr_ctx)
MAX_FLOWS)) {
ret = -EAGAIN;
to_seg = MAX_FLOWS >> 1;
- qpriv->s_flags |= RVT_S_ACK_PENDING;
- hfi1_schedule_tid_send(qp);
+ tid_rdma_trigger_ack(qp);
break;
}
@@ -4335,8 +4348,7 @@ void hfi1_rc_rcv_tid_rdma_write_data(struct hfi1_packet *packet)
trace_hfi1_tid_req_rcv_write_data(qp, 0, e->opcode, e->psn, e->lpsn,
req);
trace_hfi1_tid_write_rsp_rcv_data(qp);
- if (priv->r_tid_ack == HFI1_QP_WQE_INVALID)
- priv->r_tid_ack = priv->r_tid_tail;
+ validate_r_tid_ack(priv);
if (opcode == TID_OP(WRITE_DATA_LAST)) {
release_rdma_sge_mr(e);
@@ -4375,8 +4387,7 @@ void hfi1_rc_rcv_tid_rdma_write_data(struct hfi1_packet *packet)
}
done:
- priv->s_flags |= RVT_S_ACK_PENDING;
- hfi1_schedule_tid_send(qp);
+ tid_rdma_schedule_ack(qp);
exit:
priv->r_next_psn_kdeth = flow->flow_state.r_next_psn;
if (fecn)
@@ -4388,10 +4399,7 @@ send_nak:
if (!priv->s_nak_state) {
priv->s_nak_state = IB_NAK_PSN_ERROR;
priv->s_nak_psn = flow->flow_state.r_next_psn;
- priv->s_flags |= RVT_S_ACK_PENDING;
- if (priv->r_tid_ack == HFI1_QP_WQE_INVALID)
- priv->r_tid_ack = priv->r_tid_tail;
- hfi1_schedule_tid_send(qp);
+ tid_rdma_trigger_ack(qp);
}
goto done;
}
@@ -4939,8 +4947,7 @@ void hfi1_rc_rcv_tid_rdma_resync(struct hfi1_packet *packet)
qpriv->resync = true;
/* RESYNC request always gets a TID RDMA ACK. */
qpriv->s_nak_state = 0;
- qpriv->s_flags |= RVT_S_ACK_PENDING;
- hfi1_schedule_tid_send(qp);
+ tid_rdma_trigger_ack(qp);
bail:
if (fecn)
qp->s_flags |= RVT_S_ECN;
diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.h b/drivers/infiniband/hw/hfi1/tid_rdma.h
index 1c536185261e..6e82df2190b7 100644
--- a/drivers/infiniband/hw/hfi1/tid_rdma.h
+++ b/drivers/infiniband/hw/hfi1/tid_rdma.h
@@ -17,6 +17,7 @@
#define TID_RDMA_MIN_SEGMENT_SIZE BIT(18) /* 256 KiB (for now) */
#define TID_RDMA_MAX_SEGMENT_SIZE BIT(18) /* 256 KiB (for now) */
#define TID_RDMA_MAX_PAGES (BIT(18) >> PAGE_SHIFT)
+#define TID_RDMA_SEGMENT_SHIFT 18
/*
* Bit definitions for priv->s_flags.
@@ -274,8 +275,6 @@ u32 hfi1_build_tid_rdma_write_req(struct rvt_qp *qp, struct rvt_swqe *wqe,
struct ib_other_headers *ohdr,
u32 *bth1, u32 *bth2, u32 *len);
-void hfi1_compute_tid_rdma_flow_wt(void);
-
void hfi1_rc_rcv_tid_rdma_write_req(struct hfi1_packet *packet);
u32 hfi1_build_tid_rdma_write_resp(struct rvt_qp *qp, struct rvt_ack_entry *e,
diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.h b/drivers/infiniband/hw/hns/hns_roce_hem.h
index 86783276fb1f..3bb8f78fb7b0 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hem.h
+++ b/drivers/infiniband/hw/hns/hns_roce_hem.h
@@ -59,7 +59,7 @@ enum {
#define HNS_ROCE_HEM_CHUNK_LEN \
((256 - sizeof(struct list_head) - 2 * sizeof(int)) / \
- (sizeof(struct scatterlist)))
+ (sizeof(struct scatterlist) + sizeof(void *)))
#define check_whether_bt_num_3(type, hop_num) \
(type < HEM_TYPE_MTT && hop_num == 2)
diff --git a/drivers/infiniband/hw/hns/hns_roce_srq.c b/drivers/infiniband/hw/hns/hns_roce_srq.c
index 9591457eb768..43ea2c13b212 100644
--- a/drivers/infiniband/hw/hns/hns_roce_srq.c
+++ b/drivers/infiniband/hw/hns/hns_roce_srq.c
@@ -376,7 +376,7 @@ int hns_roce_create_srq(struct ib_srq *ib_srq,
srq->max = roundup_pow_of_two(srq_init_attr->attr.max_wr + 1);
srq->max_gs = srq_init_attr->attr.max_sge;
- srq_desc_size = max(16, 16 * srq->max_gs);
+ srq_desc_size = roundup_pow_of_two(max(16, 16 * srq->max_gs));
srq->wqe_shift = ilog2(srq_desc_size);
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c
index 74ce9249e75a..5c3d052ac30b 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.c
+++ b/drivers/infiniband/hw/mlx5/ib_rep.c
@@ -35,7 +35,7 @@ mlx5_ib_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep)
int vport_index;
if (rep->vport == MLX5_VPORT_UPLINK)
- profile = &uplink_rep_profile;
+ profile = &raw_eth_profile;
else
return mlx5_ib_set_vport_rep(dev, rep);
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.h b/drivers/infiniband/hw/mlx5/ib_rep.h
index de43b423bafc..3b6750cba796 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.h
+++ b/drivers/infiniband/hw/mlx5/ib_rep.h
@@ -10,7 +10,7 @@
#include "mlx5_ib.h"
#ifdef CONFIG_MLX5_ESWITCH
-extern const struct mlx5_ib_profile uplink_rep_profile;
+extern const struct mlx5_ib_profile raw_eth_profile;
u8 mlx5_ib_eswitch_mode(struct mlx5_eswitch *esw);
struct mlx5_ib_dev *mlx5_ib_get_rep_ibdev(struct mlx5_eswitch *esw,
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 831539419c30..46ea4f0b9b51 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -1031,7 +1031,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
if (MLX5_CAP_GEN(mdev, cd))
props->device_cap_flags |= IB_DEVICE_CROSS_CHANNEL;
- if (!mlx5_core_is_pf(mdev))
+ if (mlx5_core_is_vf(mdev))
props->device_cap_flags |= IB_DEVICE_VIRTUAL_FUNCTION;
if (mlx5_ib_port_link_layer(ibdev, 1) ==
@@ -5145,8 +5145,7 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num,
immutable->pkey_tbl_len = attr.pkey_tbl_len;
immutable->gid_tbl_len = attr.gid_tbl_len;
immutable->core_cap_flags = get_core_cap_flags(ibdev, &rep);
- if ((ll == IB_LINK_LAYER_INFINIBAND) || MLX5_CAP_GEN(dev->mdev, roce))
- immutable->max_mad_size = IB_MGMT_MAD_SIZE;
+ immutable->max_mad_size = IB_MGMT_MAD_SIZE;
return 0;
}
@@ -5249,11 +5248,9 @@ static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
{
int err;
- if (MLX5_CAP_GEN(dev->mdev, roce)) {
- err = mlx5_nic_vport_enable_roce(dev->mdev);
- if (err)
- return err;
- }
+ err = mlx5_nic_vport_enable_roce(dev->mdev);
+ if (err)
+ return err;
err = mlx5_eth_lag_init(dev);
if (err)
@@ -5262,8 +5259,7 @@ static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
return 0;
err_disable_roce:
- if (MLX5_CAP_GEN(dev->mdev, roce))
- mlx5_nic_vport_disable_roce(dev->mdev);
+ mlx5_nic_vport_disable_roce(dev->mdev);
return err;
}
@@ -5271,8 +5267,7 @@ err_disable_roce:
static void mlx5_disable_eth(struct mlx5_ib_dev *dev)
{
mlx5_eth_lag_cleanup(dev);
- if (MLX5_CAP_GEN(dev->mdev, roce))
- mlx5_nic_vport_disable_roce(dev->mdev);
+ mlx5_nic_vport_disable_roce(dev->mdev);
}
struct mlx5_ib_counter {
@@ -6444,7 +6439,7 @@ static const struct ib_device_ops mlx5_ib_dev_port_rep_ops = {
.query_port = mlx5_ib_rep_query_port,
};
-static int mlx5_ib_stage_rep_non_default_cb(struct mlx5_ib_dev *dev)
+static int mlx5_ib_stage_raw_eth_non_default_cb(struct mlx5_ib_dev *dev)
{
ib_set_device_ops(&dev->ib_dev, &mlx5_ib_dev_port_rep_ops);
return 0;
@@ -6484,7 +6479,7 @@ static void mlx5_ib_stage_common_roce_cleanup(struct mlx5_ib_dev *dev)
mlx5_remove_netdev_notifier(dev, port_num);
}
-static int mlx5_ib_stage_rep_roce_init(struct mlx5_ib_dev *dev)
+static int mlx5_ib_stage_raw_eth_roce_init(struct mlx5_ib_dev *dev)
{
struct mlx5_core_dev *mdev = dev->mdev;
enum rdma_link_layer ll;
@@ -6500,7 +6495,7 @@ static int mlx5_ib_stage_rep_roce_init(struct mlx5_ib_dev *dev)
return err;
}
-static void mlx5_ib_stage_rep_roce_cleanup(struct mlx5_ib_dev *dev)
+static void mlx5_ib_stage_raw_eth_roce_cleanup(struct mlx5_ib_dev *dev)
{
mlx5_ib_stage_common_roce_cleanup(dev);
}
@@ -6807,7 +6802,7 @@ static const struct mlx5_ib_profile pf_profile = {
mlx5_ib_stage_delay_drop_cleanup),
};
-const struct mlx5_ib_profile uplink_rep_profile = {
+const struct mlx5_ib_profile raw_eth_profile = {
STAGE_CREATE(MLX5_IB_STAGE_INIT,
mlx5_ib_stage_init_init,
mlx5_ib_stage_init_cleanup),
@@ -6818,11 +6813,11 @@ const struct mlx5_ib_profile uplink_rep_profile = {
mlx5_ib_stage_caps_init,
NULL),
STAGE_CREATE(MLX5_IB_STAGE_NON_DEFAULT_CB,
- mlx5_ib_stage_rep_non_default_cb,
+ mlx5_ib_stage_raw_eth_non_default_cb,
NULL),
STAGE_CREATE(MLX5_IB_STAGE_ROCE,
- mlx5_ib_stage_rep_roce_init,
- mlx5_ib_stage_rep_roce_cleanup),
+ mlx5_ib_stage_raw_eth_roce_init,
+ mlx5_ib_stage_raw_eth_roce_cleanup),
STAGE_CREATE(MLX5_IB_STAGE_SRQ,
mlx5_init_srq_table,
mlx5_cleanup_srq_table),
@@ -6898,6 +6893,7 @@ static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev)
static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
{
+ const struct mlx5_ib_profile *profile;
enum rdma_link_layer ll;
struct mlx5_ib_dev *dev;
int port_type_cap;
@@ -6933,7 +6929,12 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
dev->mdev = mdev;
dev->num_ports = num_ports;
- return __mlx5_ib_add(dev, &pf_profile);
+ if (ll == IB_LINK_LAYER_ETHERNET && !mlx5_is_roce_enabled(mdev))
+ profile = &raw_eth_profile;
+ else
+ profile = &pf_profile;
+
+ return __mlx5_ib_add(dev, profile);
}
static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context)
generated by cgit (git 2.34.1) at 2025-04-05 04:29:55 +0000