aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/hw')
-rw-r--r--drivers/infiniband/hw/bnxt_re/ib_verbs.c76
-rw-r--r--drivers/infiniband/hw/bnxt_re/ib_verbs.h18
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_fp.c357
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_fp.h42
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_rcfw.c88
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_rcfw.h91
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_res.c1
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_res.h53
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_sp.c3
-rw-r--r--drivers/infiniband/hw/bnxt_re/qplib_sp.h2
-rw-r--r--drivers/infiniband/hw/bnxt_re/roce_hsi.h106
-rw-r--r--drivers/infiniband/hw/cxgb4/device.c1
-rw-r--r--drivers/infiniband/hw/efa/efa.h6
-rw-r--r--drivers/infiniband/hw/efa/efa_admin_cmds_defs.h63
-rw-r--r--drivers/infiniband/hw/efa/efa_com.c5
-rw-r--r--drivers/infiniband/hw/efa/efa_com.h3
-rw-r--r--drivers/infiniband/hw/efa/efa_com_cmd.c18
-rw-r--r--drivers/infiniband/hw/efa/efa_com_cmd.h11
-rw-r--r--drivers/infiniband/hw/efa/efa_main.c52
-rw-r--r--drivers/infiniband/hw/efa/efa_verbs.c19
-rw-r--r--drivers/infiniband/hw/hfi1/Makefile4
-rw-r--r--drivers/infiniband/hw/hfi1/affinity.c12
-rw-r--r--drivers/infiniband/hw/hfi1/affinity.h3
-rw-r--r--drivers/infiniband/hw/hfi1/chip.c303
-rw-r--r--drivers/infiniband/hw/hfi1/chip.h5
-rw-r--r--drivers/infiniband/hw/hfi1/common.h13
-rw-r--r--drivers/infiniband/hw/hfi1/driver.c231
-rw-r--r--drivers/infiniband/hw/hfi1/file_ops.c4
-rw-r--r--drivers/infiniband/hw/hfi1/hfi.h38
-rw-r--r--drivers/infiniband/hw/hfi1/init.c13
-rw-r--r--drivers/infiniband/hw/hfi1/ipoib.h171
-rw-r--r--drivers/infiniband/hw/hfi1/ipoib_main.c309
-rw-r--r--drivers/infiniband/hw/hfi1/ipoib_rx.c95
-rw-r--r--drivers/infiniband/hw/hfi1/ipoib_tx.c828
-rw-r--r--drivers/infiniband/hw/hfi1/msix.c36
-rw-r--r--drivers/infiniband/hw/hfi1/msix.h7
-rw-r--r--drivers/infiniband/hw/hfi1/netdev.h118
-rw-r--r--drivers/infiniband/hw/hfi1/netdev_rx.c481
-rw-r--r--drivers/infiniband/hw/hfi1/qp.c18
-rw-r--r--drivers/infiniband/hw/hfi1/tid_rdma.c4
-rw-r--r--drivers/infiniband/hw/hfi1/trace.c42
-rw-r--r--drivers/infiniband/hw/hfi1/trace_ctxts.h11
-rw-r--r--drivers/infiniband/hw/hfi1/verbs.c14
-rw-r--r--drivers/infiniband/hw/hfi1/vnic.h5
-rw-r--r--drivers/infiniband/hw/hfi1/vnic_main.c325
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_ah.c5
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_alloc.c148
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_common.h4
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_cq.c351
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_device.h246
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hem.c114
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hem.h11
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hw_v1.c360
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hw_v2.c1675
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_hw_v2.h15
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_main.c71
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_mr.c1644
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_qp.c509
-rw-r--r--drivers/infiniband/hw/hns/hns_roce_srq.c378
-rw-r--r--drivers/infiniband/hw/i40iw/i40iw.h9
-rw-r--r--drivers/infiniband/hw/i40iw/i40iw_verbs.c1
-rw-r--r--drivers/infiniband/hw/i40iw/i40iw_verbs.h1
-rw-r--r--drivers/infiniband/hw/mlx4/ah.c11
-rw-r--r--drivers/infiniband/hw/mlx4/main.c11
-rw-r--r--drivers/infiniband/hw/mlx4/mlx4_ib.h18
-rw-r--r--drivers/infiniband/hw/mlx4/mr.c93
-rw-r--r--drivers/infiniband/hw/mlx5/Makefile3
-rw-r--r--drivers/infiniband/hw/mlx5/ah.c35
-rw-r--r--drivers/infiniband/hw/mlx5/cmd.c114
-rw-r--r--drivers/infiniband/hw/mlx5/cmd.h4
-rw-r--r--drivers/infiniband/hw/mlx5/cong.c4
-rw-r--r--drivers/infiniband/hw/mlx5/devx.c17
-rw-r--r--drivers/infiniband/hw/mlx5/flow.c147
-rw-r--r--drivers/infiniband/hw/mlx5/gsi.c38
-rw-r--r--drivers/infiniband/hw/mlx5/ib_rep.h2
-rw-r--r--drivers/infiniband/hw/mlx5/main.c73
-rw-r--r--drivers/infiniband/hw/mlx5/mlx5_ib.h72
-rw-r--r--drivers/infiniband/hw/mlx5/odp.c10
-rw-r--r--drivers/infiniband/hw/mlx5/qos.c13
-rw-r--r--drivers/infiniband/hw/mlx5/qp.c3710
-rw-r--r--drivers/infiniband/hw/mlx5/qp.h6
-rw-r--r--drivers/infiniband/hw/mlx5/qpc.c44
-rw-r--r--drivers/infiniband/hw/mlx5/srq.c10
-rw-r--r--drivers/infiniband/hw/mlx5/srq_cmd.c111
-rw-r--r--drivers/infiniband/hw/mlx5/wr.c1504
-rw-r--r--drivers/infiniband/hw/mlx5/wr.h76
-rw-r--r--drivers/infiniband/hw/mthca/mthca_dev.h10
-rw-r--r--drivers/infiniband/hw/mthca/mthca_mr.c262
-rw-r--r--drivers/infiniband/hw/mthca/mthca_provider.c105
-rw-r--r--drivers/infiniband/hw/mthca/mthca_provider.h23
-rw-r--r--drivers/infiniband/hw/ocrdma/ocrdma.h1
-rw-r--r--drivers/infiniband/hw/ocrdma/ocrdma_ah.c3
-rw-r--r--drivers/infiniband/hw/ocrdma/ocrdma_ah.h2
-rw-r--r--drivers/infiniband/hw/ocrdma/ocrdma_hw.c1
-rw-r--r--drivers/infiniband/hw/ocrdma/ocrdma_verbs.c2
-rw-r--r--drivers/infiniband/hw/qedr/main.c1
-rw-r--r--drivers/infiniband/hw/qedr/qedr.h1
-rw-r--r--drivers/infiniband/hw/qedr/verbs.c6
-rw-r--r--drivers/infiniband/hw/qedr/verbs.h2
-rw-r--r--drivers/infiniband/hw/qib/qib_iba7322.c7
-rw-r--r--drivers/infiniband/hw/qib/qib_verbs.c1
-rw-r--r--drivers/infiniband/hw/usnic/usnic_ib_verbs.c1
-rw-r--r--drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c5
-rw-r--r--drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h2
104 files changed, 8458 insertions, 7750 deletions
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 95f6d493d1b9..8b6ad5cddfce 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -177,9 +177,6 @@ int bnxt_re_query_device(struct ib_device *ibdev,
ib_attr->max_total_mcast_qp_attach = 0;
ib_attr->max_ah = dev_attr->max_ah;
- ib_attr->max_fmr = 0;
- ib_attr->max_map_per_fmr = 0;
-
ib_attr->max_srq = dev_attr->max_srq;
ib_attr->max_srq_wr = dev_attr->max_srq_wqes;
ib_attr->max_srq_sge = dev_attr->max_srq_sges;
@@ -631,11 +628,12 @@ static u8 bnxt_re_stack_to_dev_nw_type(enum rdma_network_type ntype)
return nw_type;
}
-int bnxt_re_create_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
+int bnxt_re_create_ah(struct ib_ah *ib_ah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
struct ib_pd *ib_pd = ib_ah->pd;
struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd);
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
struct bnxt_re_dev *rdev = pd->rdev;
const struct ib_gid_attr *sgid_attr;
@@ -673,7 +671,8 @@ int bnxt_re_create_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr,
memcpy(ah->qplib_ah.dmac, ah_attr->roce.dmac, ETH_ALEN);
rc = bnxt_qplib_create_ah(&rdev->qplib_res, &ah->qplib_ah,
- !(flags & RDMA_CREATE_AH_SLEEPABLE));
+ !(init_attr->flags &
+ RDMA_CREATE_AH_SLEEPABLE));
if (rc) {
ibdev_err(&rdev->ibdev, "Failed to allocate HW AH");
return rc;
@@ -856,7 +855,7 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
if (ib_copy_from_udata(&ureq, udata, sizeof(ureq)))
return -EFAULT;
- bytes = (qplib_qp->sq.max_wqe * BNXT_QPLIB_MAX_SQE_ENTRY_SIZE);
+ bytes = (qplib_qp->sq.max_wqe * qplib_qp->sq.wqe_size);
/* Consider mapping PSN search memory only for RC QPs. */
if (qplib_qp->type == CMDQ_CREATE_QP_TYPE_RC) {
psn_sz = bnxt_qplib_is_chip_gen_p5(rdev->chip_ctx) ?
@@ -879,7 +878,7 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
qplib_qp->qp_handle = ureq.qp_handle;
if (!qp->qplib_qp.srq) {
- bytes = (qplib_qp->rq.max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
+ bytes = (qplib_qp->rq.max_wqe * qplib_qp->rq.wqe_size);
bytes = PAGE_ALIGN(bytes);
umem = ib_umem_get(&rdev->ibdev, ureq.qprva, bytes,
IB_ACCESS_LOCAL_WRITE);
@@ -976,6 +975,7 @@ static struct bnxt_re_qp *bnxt_re_create_shadow_qp
qp->qplib_qp.sig_type = true;
/* Shadow QP SQ depth should be same as QP1 RQ depth */
+ qp->qplib_qp.sq.wqe_size = bnxt_re_get_swqe_size();
qp->qplib_qp.sq.max_wqe = qp1_qp->rq.max_wqe;
qp->qplib_qp.sq.max_sge = 2;
/* Q full delta can be 1 since it is internal QP */
@@ -986,6 +986,7 @@ static struct bnxt_re_qp *bnxt_re_create_shadow_qp
qp->qplib_qp.scq = qp1_qp->scq;
qp->qplib_qp.rcq = qp1_qp->rcq;
+ qp->qplib_qp.rq.wqe_size = bnxt_re_get_rwqe_size();
qp->qplib_qp.rq.max_wqe = qp1_qp->rq.max_wqe;
qp->qplib_qp.rq.max_sge = qp1_qp->rq.max_sge;
/* Q full delta can be 1 since it is internal QP */
@@ -1021,10 +1022,12 @@ static int bnxt_re_init_rq_attr(struct bnxt_re_qp *qp,
struct bnxt_qplib_dev_attr *dev_attr;
struct bnxt_qplib_qp *qplqp;
struct bnxt_re_dev *rdev;
+ struct bnxt_qplib_q *rq;
int entries;
rdev = qp->rdev;
qplqp = &qp->qplib_qp;
+ rq = &qplqp->rq;
dev_attr = &rdev->dev_attr;
if (init_attr->srq) {
@@ -1036,23 +1039,21 @@ static int bnxt_re_init_rq_attr(struct bnxt_re_qp *qp,
return -EINVAL;
}
qplqp->srq = &srq->qplib_srq;
- qplqp->rq.max_wqe = 0;
+ rq->max_wqe = 0;
} else {
+ rq->wqe_size = bnxt_re_get_rwqe_size();
/* Allocate 1 more than what's provided so posting max doesn't
* mean empty.
*/
entries = roundup_pow_of_two(init_attr->cap.max_recv_wr + 1);
- qplqp->rq.max_wqe = min_t(u32, entries,
- dev_attr->max_qp_wqes + 1);
-
- qplqp->rq.q_full_delta = qplqp->rq.max_wqe -
- init_attr->cap.max_recv_wr;
- qplqp->rq.max_sge = init_attr->cap.max_recv_sge;
- if (qplqp->rq.max_sge > dev_attr->max_qp_sges)
- qplqp->rq.max_sge = dev_attr->max_qp_sges;
+ rq->max_wqe = min_t(u32, entries, dev_attr->max_qp_wqes + 1);
+ rq->q_full_delta = rq->max_wqe - init_attr->cap.max_recv_wr;
+ rq->max_sge = init_attr->cap.max_recv_sge;
+ if (rq->max_sge > dev_attr->max_qp_sges)
+ rq->max_sge = dev_attr->max_qp_sges;
}
- qplqp->rq.sg_info.pgsize = PAGE_SIZE;
- qplqp->rq.sg_info.pgshft = PAGE_SHIFT;
+ rq->sg_info.pgsize = PAGE_SIZE;
+ rq->sg_info.pgshft = PAGE_SHIFT;
return 0;
}
@@ -1080,15 +1081,18 @@ static void bnxt_re_init_sq_attr(struct bnxt_re_qp *qp,
struct bnxt_qplib_dev_attr *dev_attr;
struct bnxt_qplib_qp *qplqp;
struct bnxt_re_dev *rdev;
+ struct bnxt_qplib_q *sq;
int entries;
rdev = qp->rdev;
qplqp = &qp->qplib_qp;
+ sq = &qplqp->sq;
dev_attr = &rdev->dev_attr;
- qplqp->sq.max_sge = init_attr->cap.max_send_sge;
- if (qplqp->sq.max_sge > dev_attr->max_qp_sges)
- qplqp->sq.max_sge = dev_attr->max_qp_sges;
+ sq->wqe_size = bnxt_re_get_swqe_size();
+ sq->max_sge = init_attr->cap.max_send_sge;
+ if (sq->max_sge > dev_attr->max_qp_sges)
+ sq->max_sge = dev_attr->max_qp_sges;
/*
* Change the SQ depth if user has requested minimum using
* configfs. Only supported for kernel consumers
@@ -1096,9 +1100,9 @@ static void bnxt_re_init_sq_attr(struct bnxt_re_qp *qp,
entries = init_attr->cap.max_send_wr;
/* Allocate 128 + 1 more than what's provided */
entries = roundup_pow_of_two(entries + BNXT_QPLIB_RESERVED_QP_WRS + 1);
- qplqp->sq.max_wqe = min_t(u32, entries, dev_attr->max_qp_wqes +
- BNXT_QPLIB_RESERVED_QP_WRS + 1);
- qplqp->sq.q_full_delta = BNXT_QPLIB_RESERVED_QP_WRS + 1;
+ sq->max_wqe = min_t(u32, entries, dev_attr->max_qp_wqes +
+ BNXT_QPLIB_RESERVED_QP_WRS + 1);
+ sq->q_full_delta = BNXT_QPLIB_RESERVED_QP_WRS + 1;
/*
* Reserving one slot for Phantom WQE. Application can
* post one extra entry in this case. But allowing this to avoid
@@ -1511,7 +1515,7 @@ static int bnxt_re_init_user_srq(struct bnxt_re_dev *rdev,
if (ib_copy_from_udata(&ureq, udata, sizeof(ureq)))
return -EFAULT;
- bytes = (qplib_srq->max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
+ bytes = (qplib_srq->max_wqe * qplib_srq->wqe_size);
bytes = PAGE_ALIGN(bytes);
umem = ib_umem_get(&rdev->ibdev, ureq.srqva, bytes,
IB_ACCESS_LOCAL_WRITE);
@@ -1534,15 +1538,20 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq,
struct ib_srq_init_attr *srq_init_attr,
struct ib_udata *udata)
{
- struct ib_pd *ib_pd = ib_srq->pd;
- struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd);
- struct bnxt_re_dev *rdev = pd->rdev;
- struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr;
- struct bnxt_re_srq *srq =
- container_of(ib_srq, struct bnxt_re_srq, ib_srq);
+ struct bnxt_qplib_dev_attr *dev_attr;
struct bnxt_qplib_nq *nq = NULL;
+ struct bnxt_re_dev *rdev;
+ struct bnxt_re_srq *srq;
+ struct bnxt_re_pd *pd;
+ struct ib_pd *ib_pd;
int rc, entries;
+ ib_pd = ib_srq->pd;
+ pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd);
+ rdev = pd->rdev;
+ dev_attr = &rdev->dev_attr;
+ srq = container_of(ib_srq, struct bnxt_re_srq, ib_srq);
+
if (srq_init_attr->attr.max_wr >= dev_attr->max_srq_wqes) {
ibdev_err(&rdev->ibdev, "Create CQ failed - max exceeded");
rc = -EINVAL;
@@ -1563,8 +1572,9 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq,
entries = roundup_pow_of_two(srq_init_attr->attr.max_wr + 1);
if (entries > dev_attr->max_srq_wqes + 1)
entries = dev_attr->max_srq_wqes + 1;
-
srq->qplib_srq.max_wqe = entries;
+
+ srq->qplib_srq.wqe_size = bnxt_re_get_rwqe_size();
srq->qplib_srq.max_sge = srq_init_attr->attr.max_sge;
srq->qplib_srq.threshold = srq_init_attr->attr.srq_limit;
srq->srq_limit = srq_init_attr->attr.srq_limit;
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
index 23d972da5652..e5fbbeba6d28 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
@@ -122,12 +122,6 @@ struct bnxt_re_frpl {
u64 *page_list;
};
-struct bnxt_re_fmr {
- struct bnxt_re_dev *rdev;
- struct ib_fmr ib_fmr;
- struct bnxt_qplib_mrw qplib_fmr;
-};
-
struct bnxt_re_mw {
struct bnxt_re_dev *rdev;
struct ib_mw ib_mw;
@@ -142,6 +136,16 @@ struct bnxt_re_ucontext {
spinlock_t sh_lock; /* protect shpg */
};
+static inline u16 bnxt_re_get_swqe_size(void)
+{
+ return sizeof(struct sq_send);
+}
+
+static inline u16 bnxt_re_get_rwqe_size(void)
+{
+ return sizeof(struct rq_wqe);
+}
+
int bnxt_re_query_device(struct ib_device *ibdev,
struct ib_device_attr *ib_attr,
struct ib_udata *udata);
@@ -160,7 +164,7 @@ enum rdma_link_layer bnxt_re_get_link_layer(struct ib_device *ibdev,
u8 port_num);
int bnxt_re_alloc_pd(struct ib_pd *pd, struct ib_udata *udata);
void bnxt_re_dealloc_pd(struct ib_pd *pd, struct ib_udata *udata);
-int bnxt_re_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr, u32 flags,
+int bnxt_re_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
int bnxt_re_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr);
int bnxt_re_query_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr);
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
index 899a5d2c100e..c5e29577cd43 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
@@ -300,12 +300,12 @@ static void bnxt_qplib_service_nq(unsigned long data)
{
struct bnxt_qplib_nq *nq = (struct bnxt_qplib_nq *)data;
struct bnxt_qplib_hwq *hwq = &nq->hwq;
- struct nq_base *nqe, **nq_ptr;
- struct bnxt_qplib_cq *cq;
- int num_cqne_processed = 0;
int num_srqne_processed = 0;
+ int num_cqne_processed = 0;
+ struct bnxt_qplib_cq *cq;
int budget = nq->budget;
u32 sw_cons, raw_cons;
+ struct nq_base *nqe;
uintptr_t q_handle;
u16 type;
@@ -314,8 +314,7 @@ static void bnxt_qplib_service_nq(unsigned long data)
raw_cons = hwq->cons;
while (budget--) {
sw_cons = HWQ_CMP(raw_cons, hwq);
- nq_ptr = (struct nq_base **)hwq->pbl_ptr;
- nqe = &nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)];
+ nqe = bnxt_qplib_get_qe(hwq, sw_cons, NULL);
if (!NQE_CMP_VALID(nqe, raw_cons, hwq->max_elements))
break;
@@ -392,13 +391,11 @@ static irqreturn_t bnxt_qplib_nq_irq(int irq, void *dev_instance)
{
struct bnxt_qplib_nq *nq = dev_instance;
struct bnxt_qplib_hwq *hwq = &nq->hwq;
- struct nq_base **nq_ptr;
u32 sw_cons;
/* Prefetch the NQ element */
sw_cons = HWQ_CMP(hwq->cons, hwq);
- nq_ptr = (struct nq_base **)nq->hwq.pbl_ptr;
- prefetch(&nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)]);
+ prefetch(bnxt_qplib_get_qe(hwq, sw_cons, NULL));
/* Fan out to CPU affinitized kthreads? */
tasklet_schedule(&nq->nq_tasklet);
@@ -612,12 +609,13 @@ int bnxt_qplib_create_srq(struct bnxt_qplib_res *res,
struct cmdq_create_srq req;
struct bnxt_qplib_pbl *pbl;
u16 cmd_flags = 0;
+ u16 pg_sz_lvl;
int rc, idx;
hwq_attr.res = res;
hwq_attr.sginfo = &srq->sg_info;
hwq_attr.depth = srq->max_wqe;
- hwq_attr.stride = BNXT_QPLIB_MAX_RQE_ENTRY_SIZE;
+ hwq_attr.stride = srq->wqe_size;
hwq_attr.type = HWQ_TYPE_QUEUE;
rc = bnxt_qplib_alloc_init_hwq(&srq->hwq, &hwq_attr);
if (rc)
@@ -638,22 +636,11 @@ int bnxt_qplib_create_srq(struct bnxt_qplib_res *res,
req.srq_size = cpu_to_le16((u16)srq->hwq.max_elements);
pbl = &srq->hwq.pbl[PBL_LVL_0];
- req.pg_size_lvl = cpu_to_le16((((u16)srq->hwq.level &
- CMDQ_CREATE_SRQ_LVL_MASK) <<
- CMDQ_CREATE_SRQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_CREATE_SRQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_SRQ_PG_SIZE_PG_4K));
+ pg_sz_lvl = ((u16)bnxt_qplib_base_pg_size(&srq->hwq) <<
+ CMDQ_CREATE_SRQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (srq->hwq.level & CMDQ_CREATE_SRQ_LVL_MASK) <<
+ CMDQ_CREATE_SRQ_LVL_SFT;
+ req.pg_size_lvl = cpu_to_le16(pg_sz_lvl);
req.pbl = cpu_to_le64(pbl->pg_map_arr[0]);
req.pd_id = cpu_to_le32(srq->pd->id);
req.eventq_id = cpu_to_le16(srq->eventq_hw_ring_id);
@@ -740,7 +727,7 @@ int bnxt_qplib_post_srq_recv(struct bnxt_qplib_srq *srq,
struct bnxt_qplib_swqe *wqe)
{
struct bnxt_qplib_hwq *srq_hwq = &srq->hwq;
- struct rq_wqe *srqe, **srqe_ptr;
+ struct rq_wqe *srqe;
struct sq_sge *hw_sge;
u32 sw_prod, sw_cons, count = 0;
int i, rc = 0, next;
@@ -758,9 +745,8 @@ int bnxt_qplib_post_srq_recv(struct bnxt_qplib_srq *srq,
spin_unlock(&srq_hwq->lock);
sw_prod = HWQ_CMP(srq_hwq->prod, srq_hwq);
- srqe_ptr = (struct rq_wqe **)srq_hwq->pbl_ptr;
- srqe = &srqe_ptr[RQE_PG(sw_prod)][RQE_IDX(sw_prod)];
- memset(srqe, 0, BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
+ srqe = bnxt_qplib_get_qe(srq_hwq, sw_prod, NULL);
+ memset(srqe, 0, srq->wqe_size);
/* Calculate wqe_size16 and data_len */
for (i = 0, hw_sge = (struct sq_sge *)srqe->data;
i < wqe->num_sge; i++, hw_sge++) {
@@ -809,6 +795,7 @@ int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
struct bnxt_qplib_pbl *pbl;
u16 cmd_flags = 0;
u32 qp_flags = 0;
+ u8 pg_sz_lvl;
int rc;
RCFW_CMD_PREP(req, CREATE_QP1, cmd_flags);
@@ -822,7 +809,7 @@ int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
hwq_attr.res = res;
hwq_attr.sginfo = &sq->sg_info;
hwq_attr.depth = sq->max_wqe;
- hwq_attr.stride = BNXT_QPLIB_MAX_SQE_ENTRY_SIZE;
+ hwq_attr.stride = sq->wqe_size;
hwq_attr.type = HWQ_TYPE_QUEUE;
rc = bnxt_qplib_alloc_init_hwq(&sq->hwq, &hwq_attr);
if (rc)
@@ -835,33 +822,18 @@ int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
}
pbl = &sq->hwq.pbl[PBL_LVL_0];
req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]);
- req.sq_pg_size_sq_lvl =
- ((sq->hwq.level & CMDQ_CREATE_QP1_SQ_LVL_MASK)
- << CMDQ_CREATE_QP1_SQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_4K);
+ pg_sz_lvl = (bnxt_qplib_base_pg_size(&sq->hwq) <<
+ CMDQ_CREATE_QP1_SQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (sq->hwq.level & CMDQ_CREATE_QP1_SQ_LVL_MASK);
+ req.sq_pg_size_sq_lvl = pg_sz_lvl;
if (qp->scq)
req.scq_cid = cpu_to_le32(qp->scq->id);
-
- qp_flags |= CMDQ_CREATE_QP1_QP_FLAGS_RESERVED_LKEY_ENABLE;
-
/* RQ */
if (rq->max_wqe) {
hwq_attr.res = res;
hwq_attr.sginfo = &rq->sg_info;
- hwq_attr.stride = BNXT_QPLIB_MAX_RQE_ENTRY_SIZE;
+ hwq_attr.stride = rq->wqe_size;
hwq_attr.depth = qp->rq.max_wqe;
hwq_attr.type = HWQ_TYPE_QUEUE;
rc = bnxt_qplib_alloc_init_hwq(&rq->hwq, &hwq_attr);
@@ -876,32 +848,20 @@ int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
}
pbl = &rq->hwq.pbl[PBL_LVL_0];
req.rq_pbl = cpu_to_le64(pbl->pg_map_arr[0]);
- req.rq_pg_size_rq_lvl =
- ((rq->hwq.level & CMDQ_CREATE_QP1_RQ_LVL_MASK) <<
- CMDQ_CREATE_QP1_RQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_4K);
+ pg_sz_lvl = (bnxt_qplib_base_pg_size(&rq->hwq) <<
+ CMDQ_CREATE_QP1_RQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (rq->hwq.level & CMDQ_CREATE_QP1_RQ_LVL_MASK);
+ req.rq_pg_size_rq_lvl = pg_sz_lvl;
if (qp->rcq)
req.rcq_cid = cpu_to_le32(qp->rcq->id);
}
-
/* Header buffer - allow hdr_buf pass in */
rc = bnxt_qplib_alloc_qp_hdr_buf(res, qp);
if (rc) {
rc = -ENOMEM;
goto fail;
}
+ qp_flags |= CMDQ_CREATE_QP1_QP_FLAGS_RESERVED_LKEY_ENABLE;
req.qp_flags = cpu_to_le32(qp_flags);
req.sq_size = cpu_to_le32(sq->hwq.max_elements);
req.rq_size = cpu_to_le32(rq->hwq.max_elements);
@@ -948,23 +908,47 @@ exit:
return rc;
}
+static void bnxt_qplib_init_psn_ptr(struct bnxt_qplib_qp *qp, int size)
+{
+ struct bnxt_qplib_hwq *hwq;
+ struct bnxt_qplib_q *sq;
+ u64 fpsne, psne, psn_pg;
+ u16 indx_pad = 0, indx;
+ u16 pg_num, pg_indx;
+ u64 *page;
+
+ sq = &qp->sq;
+ hwq = &sq->hwq;
+
+ fpsne = (u64)bnxt_qplib_get_qe(hwq, hwq->max_elements, &psn_pg);
+ if (!IS_ALIGNED(fpsne, PAGE_SIZE))
+ indx_pad = ALIGN(fpsne, PAGE_SIZE) / size;
+
+ page = (u64 *)psn_pg;
+ for (indx = 0; indx < hwq->max_elements; indx++) {
+ pg_num = (indx + indx_pad) / (PAGE_SIZE / size);
+ pg_indx = (indx + indx_pad) % (PAGE_SIZE / size);
+ psne = page[pg_num] + pg_indx * size;
+ sq->swq[indx].psn_ext = (struct sq_psn_search_ext *)psne;
+ sq->swq[indx].psn_search = (struct sq_psn_search *)psne;
+ }
+}
+
int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
{
struct bnxt_qplib_rcfw *rcfw = res->rcfw;
struct bnxt_qplib_hwq_attr hwq_attr = {};
- unsigned long int psn_search, poff = 0;
struct bnxt_qplib_sg_info sginfo = {};
- struct sq_psn_search **psn_search_ptr;
struct bnxt_qplib_q *sq = &qp->sq;
struct bnxt_qplib_q *rq = &qp->rq;
- int i, rc, req_size, psn_sz = 0;
- struct sq_send **hw_sq_send_ptr;
struct creq_create_qp_resp resp;
+ int rc, req_size, psn_sz = 0;
struct bnxt_qplib_hwq *xrrq;
u16 cmd_flags = 0, max_ssge;
- struct cmdq_create_qp req;
struct bnxt_qplib_pbl *pbl;
+ struct cmdq_create_qp req;
u32 qp_flags = 0;
+ u8 pg_sz_lvl;
u16 max_rsge;
RCFW_CMD_PREP(req, CREATE_QP, cmd_flags);
@@ -983,7 +967,7 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
hwq_attr.res = res;
hwq_attr.sginfo = &sq->sg_info;
- hwq_attr.stride = BNXT_QPLIB_MAX_SQE_ENTRY_SIZE;
+ hwq_attr.stride = sq->wqe_size;
hwq_attr.depth = sq->max_wqe;
hwq_attr.aux_stride = psn_sz;
hwq_attr.aux_depth = hwq_attr.depth;
@@ -997,64 +981,25 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
rc = -ENOMEM;
goto fail_sq;
}
- hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr;
- if (psn_sz) {
- psn_search_ptr = (struct sq_psn_search **)
- &hw_sq_send_ptr[get_sqe_pg
- (sq->hwq.max_elements)];
- psn_search = (unsigned long int)
- &hw_sq_send_ptr[get_sqe_pg(sq->hwq.max_elements)]
- [get_sqe_idx(sq->hwq.max_elements)];
- if (psn_search & ~PAGE_MASK) {
- /* If the psn_search does not start on a page boundary,
- * then calculate the offset
- */
- poff = (psn_search & ~PAGE_MASK) /
- BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE;
- }
- for (i = 0; i < sq->hwq.max_elements; i++) {
- sq->swq[i].psn_search =
- &psn_search_ptr[get_psne_pg(i + poff)]
- [get_psne_idx(i + poff)];
- /*psns_ext will be used only for P5 chips. */
- sq->swq[i].psn_ext =
- (struct sq_psn_search_ext *)
- &psn_search_ptr[get_psne_pg(i + poff)]
- [get_psne_idx(i + poff)];
- }
- }
+
+ if (psn_sz)
+ bnxt_qplib_init_psn_ptr(qp, psn_sz);
+
pbl = &sq->hwq.pbl[PBL_LVL_0];
req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]);
- req.sq_pg_size_sq_lvl =
- ((sq->hwq.level & CMDQ_CREATE_QP_SQ_LVL_MASK)
- << CMDQ_CREATE_QP_SQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K);
+ pg_sz_lvl = (bnxt_qplib_base_pg_size(&sq->hwq) <<
+ CMDQ_CREATE_QP_SQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (sq->hwq.level & CMDQ_CREATE_QP_SQ_LVL_MASK);
+ req.sq_pg_size_sq_lvl = pg_sz_lvl;
if (qp->scq)
req.scq_cid = cpu_to_le32(qp->scq->id);
- qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_RESERVED_LKEY_ENABLE;
- qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FR_PMR_ENABLED;
- if (qp->sig_type)
- qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FORCE_COMPLETION;
-
/* RQ */
if (rq->max_wqe) {
hwq_attr.res = res;
hwq_attr.sginfo = &rq->sg_info;
- hwq_attr.stride = BNXT_QPLIB_MAX_RQE_ENTRY_SIZE;
+ hwq_attr.stride = rq->wqe_size;
hwq_attr.depth = rq->max_wqe;
hwq_attr.aux_stride = 0;
hwq_attr.aux_depth = 0;
@@ -1071,22 +1016,10 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
}
pbl = &rq->hwq.pbl[PBL_LVL_0];
req.rq_pbl = cpu_to_le64(pbl->pg_map_arr[0]);
- req.rq_pg_size_rq_lvl =
- ((rq->hwq.level & CMDQ_CREATE_QP_RQ_LVL_MASK) <<
- CMDQ_CREATE_QP_RQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_QP_RQ_PG_SIZE_PG_4K);
+ pg_sz_lvl = (bnxt_qplib_base_pg_size(&rq->hwq) <<
+ CMDQ_CREATE_QP_RQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (rq->hwq.level & CMDQ_CREATE_QP_RQ_LVL_MASK);
+ req.rq_pg_size_rq_lvl = pg_sz_lvl;
} else {
/* SRQ */
if (qp->srq) {
@@ -1097,7 +1030,13 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
if (qp->rcq)
req.rcq_cid = cpu_to_le32(qp->rcq->id);
+
+ qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_RESERVED_LKEY_ENABLE;
+ qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FR_PMR_ENABLED;
+ if (qp->sig_type)
+ qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FORCE_COMPLETION;
req.qp_flags = cpu_to_le32(qp_flags);
+
req.sq_size = cpu_to_le32(sq->hwq.max_elements);
req.rq_size = cpu_to_le32(rq->hwq.max_elements);
qp->sq_hdr_buf = NULL;
@@ -1483,12 +1422,11 @@ bail:
static void __clean_cq(struct bnxt_qplib_cq *cq, u64 qp)
{
struct bnxt_qplib_hwq *cq_hwq = &cq->hwq;
- struct cq_base *hw_cqe, **hw_cqe_ptr;
+ struct cq_base *hw_cqe;
int i;
for (i = 0; i < cq_hwq->max_elements; i++) {
- hw_cqe_ptr = (struct cq_base **)cq_hwq->pbl_ptr;
- hw_cqe = &hw_cqe_ptr[CQE_PG(i)][CQE_IDX(i)];
+ hw_cqe = bnxt_qplib_get_qe(cq_hwq, i, NULL);
if (!CQE_CMP_VALID(hw_cqe, i, cq_hwq->max_elements))
continue;
/*
@@ -1615,6 +1553,34 @@ void *bnxt_qplib_get_qp1_rq_buf(struct bnxt_qplib_qp *qp,
return NULL;
}
+static void bnxt_qplib_fill_psn_search(struct bnxt_qplib_qp *qp,
+ struct bnxt_qplib_swqe *wqe,
+ struct bnxt_qplib_swq *swq)
+{
+ struct sq_psn_search_ext *psns_ext;
+ struct sq_psn_search *psns;
+ u32 flg_npsn;
+ u32 op_spsn;
+
+ psns = swq->psn_search;
+ psns_ext = swq->psn_ext;
+
+ op_spsn = ((swq->start_psn << SQ_PSN_SEARCH_START_PSN_SFT) &
+ SQ_PSN_SEARCH_START_PSN_MASK);
+ op_spsn |= ((wqe->type << SQ_PSN_SEARCH_OPCODE_SFT) &
+ SQ_PSN_SEARCH_OPCODE_MASK);
+ flg_npsn = ((swq->next_psn << SQ_PSN_SEARCH_NEXT_PSN_SFT) &
+ SQ_PSN_SEARCH_NEXT_PSN_MASK);
+
+ if (bnxt_qplib_is_chip_gen_p5(qp->cctx)) {
+ psns_ext->opcode_start_psn = cpu_to_le32(op_spsn);
+ psns_ext->flags_next_psn = cpu_to_le32(flg_npsn);
+ } else {
+ psns->opcode_start_psn = cpu_to_le32(op_spsn);
+ psns->flags_next_psn = cpu_to_le32(flg_npsn);
+ }
+}
+
void bnxt_qplib_post_send_db(struct bnxt_qplib_qp *qp)
{
struct bnxt_qplib_q *sq = &qp->sq;
@@ -1625,16 +1591,16 @@ void bnxt_qplib_post_send_db(struct bnxt_qplib_qp *qp)
int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
struct bnxt_qplib_swqe *wqe)
{
+ struct bnxt_qplib_nq_work *nq_work = NULL;
+ int i, rc = 0, data_len = 0, pkt_num = 0;
struct bnxt_qplib_q *sq = &qp->sq;
+ struct sq_send *hw_sq_send_hdr;
struct bnxt_qplib_swq *swq;
- struct sq_send *hw_sq_send_hdr, **hw_sq_send_ptr;
- struct sq_sge *hw_sge;
- struct bnxt_qplib_nq_work *nq_work = NULL;
bool sch_handler = false;
- u32 sw_prod;
+ struct sq_sge *hw_sge;
u8 wqe_size16;
- int i, rc = 0, data_len = 0, pkt_num = 0;
__le32 temp32;
+ u32 sw_prod;
if (qp->state != CMDQ_MODIFY_QP_NEW_STATE_RTS) {
if (qp->state == CMDQ_MODIFY_QP_NEW_STATE_ERR) {
@@ -1663,11 +1629,8 @@ int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
swq->flags |= SQ_SEND_FLAGS_SIGNAL_COMP;
swq->start_psn = sq->psn & BTH_PSN_MASK;
- hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr;
- hw_sq_send_hdr = &hw_sq_send_ptr[get_sqe_pg(sw_prod)]
- [get_sqe_idx(sw_prod)];
-
- memset(hw_sq_send_hdr, 0, BNXT_QPLIB_MAX_SQE_ENTRY_SIZE);
+ hw_sq_send_hdr = bnxt_qplib_get_qe(&sq->hwq, sw_prod, NULL);
+ memset(hw_sq_send_hdr, 0, sq->wqe_size);
if (wqe->flags & BNXT_QPLIB_SWQE_FLAGS_INLINE) {
/* Copy the inline data */
@@ -1854,28 +1817,8 @@ int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
goto done;
}
swq->next_psn = sq->psn & BTH_PSN_MASK;
- if (swq->psn_search) {
- u32 opcd_spsn;
- u32 flg_npsn;
-
- opcd_spsn = ((swq->start_psn << SQ_PSN_SEARCH_START_PSN_SFT) &
- SQ_PSN_SEARCH_START_PSN_MASK);
- opcd_spsn |= ((wqe->type << SQ_PSN_SEARCH_OPCODE_SFT) &
- SQ_PSN_SEARCH_OPCODE_MASK);
- flg_npsn = ((swq->next_psn << SQ_PSN_SEARCH_NEXT_PSN_SFT) &
- SQ_PSN_SEARCH_NEXT_PSN_MASK);
- if (bnxt_qplib_is_chip_gen_p5(qp->cctx)) {
- swq->psn_ext->opcode_start_psn =
- cpu_to_le32(opcd_spsn);
- swq->psn_ext->flags_next_psn =
- cpu_to_le32(flg_npsn);
- } else {
- swq->psn_search->opcode_start_psn =
- cpu_to_le32(opcd_spsn);
- swq->psn_search->flags_next_psn =
- cpu_to_le32(flg_npsn);
- }
- }
+ if (qp->type == CMDQ_CREATE_QP_TYPE_RC)
+ bnxt_qplib_fill_psn_search(qp, wqe, swq);
queue_err:
if (sch_handler) {
/* Store the ULP info in the software structures */
@@ -1918,13 +1861,13 @@ void bnxt_qplib_post_recv_db(struct bnxt_qplib_qp *qp)
int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp,
struct bnxt_qplib_swqe *wqe)
{
- struct bnxt_qplib_q *rq = &qp->rq;
- struct rq_wqe *rqe, **rqe_ptr;
- struct sq_sge *hw_sge;
struct bnxt_qplib_nq_work *nq_work = NULL;
+ struct bnxt_qplib_q *rq = &qp->rq;
bool sch_handler = false;
- u32 sw_prod;
+ struct sq_sge *hw_sge;
+ struct rq_wqe *rqe;
int i, rc = 0;
+ u32 sw_prod;
if (qp->state == CMDQ_MODIFY_QP_NEW_STATE_ERR) {
sch_handler = true;
@@ -1941,10 +1884,8 @@ int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp,
sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq);
rq->swq[sw_prod].wr_id = wqe->wr_id;
- rqe_ptr = (struct rq_wqe **)rq->hwq.pbl_ptr;
- rqe = &rqe_ptr[RQE_PG(sw_prod)][RQE_IDX(sw_prod)];
-
- memset(rqe, 0, BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
+ rqe = bnxt_qplib_get_qe(&rq->hwq, sw_prod, NULL);
+ memset(rqe, 0, rq->wqe_size);
/* Calculate wqe_size16 and data_len */
for (i = 0, hw_sge = (struct sq_sge *)rqe->data;
@@ -1997,9 +1938,10 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq)
struct bnxt_qplib_rcfw *rcfw = res->rcfw;
struct bnxt_qplib_hwq_attr hwq_attr = {};
struct creq_create_cq_resp resp;
- struct cmdq_create_cq req;
struct bnxt_qplib_pbl *pbl;
+ struct cmdq_create_cq req;
u16 cmd_flags = 0;
+ u32 pg_sz_lvl;
int rc;
hwq_attr.res = res;
@@ -2020,22 +1962,13 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq)
}
req.dpi = cpu_to_le32(cq->dpi->dpi);
req.cq_handle = cpu_to_le64(cq->cq_handle);
-
req.cq_size = cpu_to_le32(cq->hwq.max_elements);
pbl = &cq->hwq.pbl[PBL_LVL_0];
- req.pg_size_lvl = cpu_to_le32(
- ((cq->hwq.level & CMDQ_CREATE_CQ_LVL_MASK) <<
- CMDQ_CREATE_CQ_LVL_SFT) |
- (pbl->pg_size == ROCE_PG_SIZE_4K ? CMDQ_CREATE_CQ_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ? CMDQ_CREATE_CQ_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ? CMDQ_CREATE_CQ_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ? CMDQ_CREATE_CQ_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ? CMDQ_CREATE_CQ_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ? CMDQ_CREATE_CQ_PG_SIZE_PG_1G :
- CMDQ_CREATE_CQ_PG_SIZE_PG_4K));
-
+ pg_sz_lvl = (bnxt_qplib_base_pg_size(&cq->hwq) <<
+ CMDQ_CREATE_CQ_PG_SIZE_SFT);
+ pg_sz_lvl |= (cq->hwq.level & CMDQ_CREATE_CQ_LVL_MASK);
+ req.pg_size_lvl = cpu_to_le32(pg_sz_lvl);
req.pbl = cpu_to_le64(pbl->pg_map_arr[0]);
-
req.cq_fco_cnq_id = cpu_to_le32(
(cq->cnq_hw_ring_id & CMDQ_CREATE_CQ_CNQ_ID_MASK) <<
CMDQ_CREATE_CQ_CNQ_ID_SFT);
@@ -2194,13 +2127,13 @@ void bnxt_qplib_mark_qp_error(void *qp_handle)
static int do_wa9060(struct bnxt_qplib_qp *qp, struct bnxt_qplib_cq *cq,
u32 cq_cons, u32 sw_sq_cons, u32 cqe_sq_cons)
{
- struct bnxt_qplib_q *sq = &qp->sq;
- struct bnxt_qplib_swq *swq;
u32 peek_sw_cq_cons, peek_raw_cq_cons, peek_sq_cons_idx;
- struct cq_base *peek_hwcqe, **peek_hw_cqe_ptr;
+ struct bnxt_qplib_q *sq = &qp->sq;
struct cq_req *peek_req_hwcqe;
struct bnxt_qplib_qp *peek_qp;
struct bnxt_qplib_q *peek_sq;
+ struct bnxt_qplib_swq *swq;
+ struct cq_base *peek_hwcqe;
int i, rc = 0;
/* Normal mode */
@@ -2230,9 +2163,8 @@ static int do_wa9060(struct bnxt_qplib_qp *qp, struct bnxt_qplib_cq *cq,
i = cq->hwq.max_elements;
while (i--) {
peek_sw_cq_cons = HWQ_CMP((peek_sw_cq_cons), &cq->hwq);
- peek_hw_cqe_ptr = (struct cq_base **)cq->hwq.pbl_ptr;
- peek_hwcqe = &peek_hw_cqe_ptr[CQE_PG(peek_sw_cq_cons)]
- [CQE_IDX(peek_sw_cq_cons)];
+ peek_hwcqe = bnxt_qplib_get_qe(&cq->hwq,
+ peek_sw_cq_cons, NULL);
/* If the next hwcqe is VALID */
if (CQE_CMP_VALID(peek_hwcqe, peek_raw_cq_cons,
cq->hwq.max_elements)) {
@@ -2294,11 +2226,11 @@ static int bnxt_qplib_cq_process_req(struct bnxt_qplib_cq *cq,
struct bnxt_qplib_cqe **pcqe, int *budget,
u32 cq_cons, struct bnxt_qplib_qp **lib_qp)
{
- struct bnxt_qplib_qp *qp;
- struct bnxt_qplib_q *sq;
- struct bnxt_qplib_cqe *cqe;
u32 sw_sq_cons, cqe_sq_cons;
struct bnxt_qplib_swq *swq;
+ struct bnxt_qplib_cqe *cqe;
+ struct bnxt_qplib_qp *qp;
+ struct bnxt_qplib_q *sq;
int rc = 0;
qp = (struct bnxt_qplib_qp *)((unsigned long)
@@ -2408,10 +2340,10 @@ static int bnxt_qplib_cq_process_res_rc(struct bnxt_qplib_cq *cq,
struct bnxt_qplib_cqe **pcqe,
int *budget)
{
- struct bnxt_qplib_qp *qp;
- struct bnxt_qplib_q *rq;
struct bnxt_qplib_srq *srq;
struct bnxt_qplib_cqe *cqe;
+ struct bnxt_qplib_qp *qp;
+ struct bnxt_qplib_q *rq;
u32 wr_id_idx;
int rc = 0;
@@ -2483,10 +2415,10 @@ static int bnxt_qplib_cq_process_res_ud(struct bnxt_qplib_cq *cq,
struct bnxt_qplib_cqe **pcqe,
int *budget)
{
- struct bnxt_qplib_qp *qp;
- struct bnxt_qplib_q *rq;
struct bnxt_qplib_srq *srq;
struct bnxt_qplib_cqe *cqe;
+ struct bnxt_qplib_qp *qp;
+ struct bnxt_qplib_q *rq;
u32 wr_id_idx;
int rc = 0;
@@ -2561,15 +2493,13 @@ done:
bool bnxt_qplib_is_cq_empty(struct bnxt_qplib_cq *cq)
{
- struct cq_base *hw_cqe, **hw_cqe_ptr;
+ struct cq_base *hw_cqe;
u32 sw_cons, raw_cons;
bool rc = true;
raw_cons = cq->hwq.cons;
sw_cons = HWQ_CMP(raw_cons, &cq->hwq);
- hw_cqe_ptr = (struct cq_base **)cq->hwq.pbl_ptr;
- hw_cqe = &hw_cqe_ptr[CQE_PG(sw_cons)][CQE_IDX(sw_cons)];
-
+ hw_cqe = bnxt_qplib_get_qe(&cq->hwq, sw_cons, NULL);
/* Check for Valid bit. If the CQE is valid, return false */
rc = !CQE_CMP_VALID(hw_cqe, raw_cons, cq->hwq.max_elements);
return rc;
@@ -2813,7 +2743,7 @@ int bnxt_qplib_process_flush_list(struct bnxt_qplib_cq *cq,
int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe,
int num_cqes, struct bnxt_qplib_qp **lib_qp)
{
- struct cq_base *hw_cqe, **hw_cqe_ptr;
+ struct cq_base *hw_cqe;
u32 sw_cons, raw_cons;
int budget, rc = 0;
@@ -2822,8 +2752,7 @@ int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe,
while (budget) {
sw_cons = HWQ_CMP(raw_cons, &cq->hwq);
- hw_cqe_ptr = (struct cq_base **)cq->hwq.pbl_ptr;
- hw_cqe = &hw_cqe_ptr[CQE_PG(sw_cons)][CQE_IDX(sw_cons)];
+ hw_cqe = bnxt_qplib_get_qe(&cq->hwq, sw_cons, NULL);
/* Check for Valid bit */
if (!CQE_CMP_VALID(hw_cqe, raw_cons, cq->hwq.max_elements))
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
index 7edb70b6bb16..568ca390322c 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
@@ -45,6 +45,7 @@ struct bnxt_qplib_srq {
struct bnxt_qplib_db_info dbinfo;
u64 srq_handle;
u32 id;
+ u16 wqe_size;
u32 max_wqe;
u32 max_sge;
u32 threshold;
@@ -65,38 +66,7 @@ struct bnxt_qplib_sge {
u32 size;
};
-#define BNXT_QPLIB_MAX_SQE_ENTRY_SIZE sizeof(struct sq_send)
-
-#define SQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_SQE_ENTRY_SIZE)
-#define SQE_MAX_IDX_PER_PG (SQE_CNT_PER_PG - 1)
-
-static inline u32 get_sqe_pg(u32 val)
-{
- return ((val & ~SQE_MAX_IDX_PER_PG) / SQE_CNT_PER_PG);
-}
-
-static inline u32 get_sqe_idx(u32 val)
-{
- return (val & SQE_MAX_IDX_PER_PG);
-}
-
-#define BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE sizeof(struct sq_psn_search)
-
-#define PSNE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE)
-#define PSNE_MAX_IDX_PER_PG (PSNE_CNT_PER_PG - 1)
-
-static inline u32 get_psne_pg(u32 val)
-{
- return ((val & ~PSNE_MAX_IDX_PER_PG) / PSNE_CNT_PER_PG);
-}
-
-static inline u32 get_psne_idx(u32 val)
-{
- return (val & PSNE_MAX_IDX_PER_PG);
-}
-
#define BNXT_QPLIB_QP_MAX_SGL 6
-
struct bnxt_qplib_swq {
u64 wr_id;
int next_idx;
@@ -226,19 +196,13 @@ struct bnxt_qplib_swqe {
};
};
-#define BNXT_QPLIB_MAX_RQE_ENTRY_SIZE sizeof(struct rq_wqe)
-
-#define RQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_RQE_ENTRY_SIZE)
-#define RQE_MAX_IDX_PER_PG (RQE_CNT_PER_PG - 1)
-#define RQE_PG(x) (((x) & ~RQE_MAX_IDX_PER_PG) / RQE_CNT_PER_PG)
-#define RQE_IDX(x) ((x) & RQE_MAX_IDX_PER_PG)
-
struct bnxt_qplib_q {
struct bnxt_qplib_hwq hwq;
struct bnxt_qplib_swq *swq;
struct bnxt_qplib_db_info dbinfo;
struct bnxt_qplib_sg_info sg_info;
u32 max_wqe;
+ u16 wqe_size;
u16 q_full_delta;
u16 max_sge;
u32 psn;
@@ -256,7 +220,7 @@ struct bnxt_qplib_qp {
struct bnxt_qplib_dpi *dpi;
struct bnxt_qplib_chip_ctx *cctx;
u64 qp_handle;
-#define BNXT_QPLIB_QP_ID_INVALID 0xFFFFFFFF
+#define BNXT_QPLIB_QP_ID_INVALID 0xFFFFFFFF
u32 id;
u8 type;
u8 sig_type;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
index f01e864bb611..4e211162acee 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
@@ -89,10 +89,9 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw, struct cmdq_base *req,
struct creq_base *resp, void *sb, u8 is_block)
{
struct bnxt_qplib_cmdq_ctx *cmdq = &rcfw->cmdq;
- struct bnxt_qplib_cmdqe *cmdqe, **hwq_ptr;
struct bnxt_qplib_hwq *hwq = &cmdq->hwq;
struct bnxt_qplib_crsqe *crsqe;
- u32 cmdq_depth = rcfw->cmdq_depth;
+ struct bnxt_qplib_cmdqe *cmdqe;
u32 sw_prod, cmdq_prod;
struct pci_dev *pdev;
unsigned long flags;
@@ -163,13 +162,11 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw, struct cmdq_base *req,
BNXT_QPLIB_CMDQE_UNITS;
}
- hwq_ptr = (struct bnxt_qplib_cmdqe **)hwq->pbl_ptr;
preq = (u8 *)req;
do {
/* Locate the next cmdq slot */
sw_prod = HWQ_CMP(hwq->prod, hwq);
- cmdqe = &hwq_ptr[get_cmdq_pg(sw_prod, cmdq_depth)]
- [get_cmdq_idx(sw_prod, cmdq_depth)];
+ cmdqe = bnxt_qplib_get_qe(hwq, sw_prod, NULL);
if (!cmdqe) {
dev_err(&pdev->dev,
"RCFW request failed with no cmdqe!\n");
@@ -378,7 +375,7 @@ static void bnxt_qplib_service_creq(unsigned long data)
struct bnxt_qplib_creq_ctx *creq = &rcfw->creq;
u32 type, budget = CREQ_ENTRY_POLL_BUDGET;
struct bnxt_qplib_hwq *hwq = &creq->hwq;
- struct creq_base *creqe, **hwq_ptr;
+ struct creq_base *creqe;
u32 sw_cons, raw_cons;
unsigned long flags;
@@ -387,8 +384,7 @@ static void bnxt_qplib_service_creq(unsigned long data)
raw_cons = hwq->cons;
while (budget > 0) {
sw_cons = HWQ_CMP(raw_cons, hwq);
- hwq_ptr = (struct creq_base **)hwq->pbl_ptr;
- creqe = &hwq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)];
+ creqe = bnxt_qplib_get_qe(hwq, sw_cons, NULL);
if (!CREQ_CMP_VALID(creqe, raw_cons, hwq->max_elements))
break;
/* The valid test of the entry must be done first before
@@ -434,7 +430,6 @@ static irqreturn_t bnxt_qplib_creq_irq(int irq, void *dev_instance)
{
struct bnxt_qplib_rcfw *rcfw = dev_instance;
struct bnxt_qplib_creq_ctx *creq;
- struct creq_base **creq_ptr;
struct bnxt_qplib_hwq *hwq;
u32 sw_cons;
@@ -442,8 +437,7 @@ static irqreturn_t bnxt_qplib_creq_irq(int irq, void *dev_instance)
hwq = &creq->hwq;
/* Prefetch the CREQ element */
sw_cons = HWQ_CMP(hwq->cons, hwq);
- creq_ptr = (struct creq_base **)creq->hwq.pbl_ptr;
- prefetch(&creq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)]);
+ prefetch(bnxt_qplib_get_qe(hwq, sw_cons, NULL));
tasklet_schedule(&creq->creq_tasklet);
@@ -468,29 +462,13 @@ int bnxt_qplib_deinit_rcfw(struct bnxt_qplib_rcfw *rcfw)
return 0;
}
-static int __get_pbl_pg_idx(struct bnxt_qplib_pbl *pbl)
-{
- return (pbl->pg_size == ROCE_PG_SIZE_4K ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K :
- pbl->pg_size == ROCE_PG_SIZE_8K ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8K :
- pbl->pg_size == ROCE_PG_SIZE_64K ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_64K :
- pbl->pg_size == ROCE_PG_SIZE_2M ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_2M :
- pbl->pg_size == ROCE_PG_SIZE_8M ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8M :
- pbl->pg_size == ROCE_PG_SIZE_1G ?
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_1G :
- CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K);
-}
-
int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw,
struct bnxt_qplib_ctx *ctx, int is_virtfn)
{
- struct cmdq_initialize_fw req;
struct creq_initialize_fw_resp resp;
- u16 cmd_flags = 0, level;
+ struct cmdq_initialize_fw req;
+ u16 cmd_flags = 0;
+ u8 pgsz, lvl;
int rc;
RCFW_CMD_PREP(req, INITIALIZE_FW, cmd_flags);
@@ -511,32 +489,30 @@ int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw,
if (bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx))
goto config_vf_res;
- level = ctx->qpc_tbl.level;
- req.qpc_pg_size_qpc_lvl = (level << CMDQ_INITIALIZE_FW_QPC_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->qpc_tbl.pbl[level]);
- level = ctx->mrw_tbl.level;
- req.mrw_pg_size_mrw_lvl = (level << CMDQ_INITIALIZE_FW_MRW_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->mrw_tbl.pbl[level]);
- level = ctx->srqc_tbl.level;
- req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]);
- level = ctx->cq_tbl.level;
- req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]);
- level = ctx->srqc_tbl.level;
- req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]);
- level = ctx->cq_tbl.level;
- req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]);
- level = ctx->tim_tbl.level;
- req.tim_pg_size_tim_lvl = (level << CMDQ_INITIALIZE_FW_TIM_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->tim_tbl.pbl[level]);
- level = ctx->tqm_ctx.pde.level;
- req.tqm_pg_size_tqm_lvl =
- (level << CMDQ_INITIALIZE_FW_TQM_LVL_SFT) |
- __get_pbl_pg_idx(&ctx->tqm_ctx.pde.pbl[level]);
-
+ lvl = ctx->qpc_tbl.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->qpc_tbl);
+ req.qpc_pg_size_qpc_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
+ lvl = ctx->mrw_tbl.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->mrw_tbl);
+ req.mrw_pg_size_mrw_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
+ lvl = ctx->srqc_tbl.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->srqc_tbl);
+ req.srq_pg_size_srq_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
+ lvl = ctx->cq_tbl.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->cq_tbl);
+ req.cq_pg_size_cq_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
+ lvl = ctx->tim_tbl.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->tim_tbl);
+ req.tim_pg_size_tim_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
+ lvl = ctx->tqm_ctx.pde.level;
+ pgsz = bnxt_qplib_base_pg_size(&ctx->tqm_ctx.pde);
+ req.tqm_pg_size_tqm_lvl = (pgsz << CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT) |
+ lvl;
req.qpc_page_dir =
cpu_to_le64(ctx->qpc_tbl.pbl[PBL_LVL_0].pg_map_arr[0]);
req.mrw_page_dir =
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
index 411fce3493b6..157387636d00 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
@@ -87,12 +87,6 @@ static inline u32 bnxt_qplib_cmdqe_page_size(u32 depth)
return (bnxt_qplib_cmdqe_npages(depth) * PAGE_SIZE);
}
-static inline u32 bnxt_qplib_cmdqe_cnt_per_pg(u32 depth)
-{
- return (bnxt_qplib_cmdqe_page_size(depth) /
- BNXT_QPLIB_CMDQE_UNITS);
-}
-
/* Set the cmd_size to a factor of CMDQE unit */
static inline void bnxt_qplib_set_cmd_slots(struct cmdq_base *req)
{
@@ -100,30 +94,12 @@ static inline void bnxt_qplib_set_cmd_slots(struct cmdq_base *req)
BNXT_QPLIB_CMDQE_UNITS;
}
-#define MAX_CMDQ_IDX(depth) ((depth) - 1)
-
-static inline u32 bnxt_qplib_max_cmdq_idx_per_pg(u32 depth)
-{
- return (bnxt_qplib_cmdqe_cnt_per_pg(depth) - 1);
-}
-
#define RCFW_MAX_COOKIE_VALUE 0x7FFF
#define RCFW_CMD_IS_BLOCKING 0x8000
#define RCFW_BLOCKED_CMD_WAIT_COUNT 0x4E20
#define HWRM_VERSION_RCFW_CMDQ_DEPTH_CHECK 0x1000900020011ULL
-static inline u32 get_cmdq_pg(u32 val, u32 depth)
-{
- return (val & ~(bnxt_qplib_max_cmdq_idx_per_pg(depth))) /
- (bnxt_qplib_cmdqe_cnt_per_pg(depth));
-}
-
-static inline u32 get_cmdq_idx(u32 val, u32 depth)
-{
- return val & (bnxt_qplib_max_cmdq_idx_per_pg(depth));
-}
-
/* Crsq buf is 1024-Byte */
struct bnxt_qplib_crsbe {
u8 data[1024];
@@ -133,76 +109,9 @@ struct bnxt_qplib_crsbe {
/* Allocate 1 per QP for async error notification for now */
#define BNXT_QPLIB_CREQE_MAX_CNT (64 * 1024)
#define BNXT_QPLIB_CREQE_UNITS 16 /* 16-Bytes per prod unit */
-#define BNXT_QPLIB_CREQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CREQE_UNITS)
-
-#define MAX_CREQ_IDX (BNXT_QPLIB_CREQE_MAX_CNT - 1)
-#define MAX_CREQ_IDX_PER_PG (BNXT_QPLIB_CREQE_CNT_PER_PG - 1)
-
-static inline u32 get_creq_pg(u32 val)
-{
- return (val & ~MAX_CREQ_IDX_PER_PG) / BNXT_QPLIB_CREQE_CNT_PER_PG;
-}
-
-static inline u32 get_creq_idx(u32 val)
-{
- return val & MAX_CREQ_IDX_PER_PG;
-}
-
-#define BNXT_QPLIB_CREQE_PER_PG (PAGE_SIZE / sizeof(struct creq_base))
-
#define CREQ_CMP_VALID(hdr, raw_cons, cp_bit) \
(!!((hdr)->v & CREQ_BASE_V) == \
!((raw_cons) & (cp_bit)))
-
-#define CREQ_DB_KEY_CP (0x2 << CMPL_DOORBELL_KEY_SFT)
-#define CREQ_DB_IDX_VALID CMPL_DOORBELL_IDX_VALID
-#define CREQ_DB_IRQ_DIS CMPL_DOORBELL_MASK
-#define CREQ_DB_CP_FLAGS_REARM (CREQ_DB_KEY_CP | \
- CREQ_DB_IDX_VALID)
-#define CREQ_DB_CP_FLAGS (CREQ_DB_KEY_CP | \
- CREQ_DB_IDX_VALID | \
- CREQ_DB_IRQ_DIS)
-
-static inline void bnxt_qplib_ring_creq_db64(void __iomem *db, u32 index,
- u32 xid, bool arm)
-{
- u64 val = 0;
-
- val = xid & DBC_DBC_XID_MASK;
- val |= DBC_DBC_PATH_ROCE;
- val |= arm ? DBC_DBC_TYPE_NQ_ARM : DBC_DBC_TYPE_NQ;
- val <<= 32;
- val |= index & DBC_DBC_INDEX_MASK;
-
- writeq(val, db);
-}
-
-static inline void bnxt_qplib_ring_creq_db_rearm(void __iomem *db, u32 raw_cons,
- u32 max_elements, u32 xid,
- bool gen_p5)
-{
- u32 index = raw_cons & (max_elements - 1);
-
- if (gen_p5)
- bnxt_qplib_ring_creq_db64(db, index, xid, true);
- else
- writel(CREQ_DB_CP_FLAGS_REARM | (index & DBC_DBC32_XID_MASK),
- db);
-}
-
-static inline void bnxt_qplib_ring_creq_db(void __iomem *db, u32 raw_cons,
- u32 max_elements, u32 xid,
- bool gen_p5)
-{
- u32 index = raw_cons & (max_elements - 1);
-
- if (gen_p5)
- bnxt_qplib_ring_creq_db64(db, index, xid, true);
- else
- writel(CREQ_DB_CP_FLAGS | (index & DBC_DBC32_XID_MASK),
- db);
-}
-
#define CREQ_ENTRY_POLL_BUDGET 0x100
/* HWQ */
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c
index cab1adf1fed9..7efa6e5dce62 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c
@@ -347,6 +347,7 @@ done:
hwq->depth = hwq_attr->depth;
hwq->max_elements = depth;
hwq->element_size = stride;
+ hwq->qe_ppg = pg_size / stride;
/* For direct access to the elements */
lvl = hwq->level;
if (hwq_attr->sginfo->nopte && hwq->level)
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h
index 95b645dbbc2d..c29cbd3a2d7b 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h
@@ -80,6 +80,15 @@ enum bnxt_qplib_pbl_lvl {
#define ROCE_PG_SIZE_8M (8 * 1024 * 1024)
#define ROCE_PG_SIZE_1G (1024 * 1024 * 1024)
+enum bnxt_qplib_hwrm_pg_size {
+ BNXT_QPLIB_HWRM_PG_SIZE_4K = 0,
+ BNXT_QPLIB_HWRM_PG_SIZE_8K = 1,
+ BNXT_QPLIB_HWRM_PG_SIZE_64K = 2,
+ BNXT_QPLIB_HWRM_PG_SIZE_2M = 3,
+ BNXT_QPLIB_HWRM_PG_SIZE_8M = 4,
+ BNXT_QPLIB_HWRM_PG_SIZE_1G = 5,
+};
+
struct bnxt_qplib_reg_desc {
u8 bar_id;
resource_size_t bar_base;
@@ -126,6 +135,7 @@ struct bnxt_qplib_hwq {
u32 max_elements;
u32 depth;
u16 element_size; /* Size of each entry */
+ u16 qe_ppg; /* queue entry per page */
u32 prod; /* raw */
u32 cons; /* raw */
@@ -263,6 +273,49 @@ static inline u8 bnxt_qplib_get_ring_type(struct bnxt_qplib_chip_ctx *cctx)
RING_ALLOC_REQ_RING_TYPE_ROCE_CMPL;
}
+static inline u8 bnxt_qplib_base_pg_size(struct bnxt_qplib_hwq *hwq)
+{
+ u8 pg_size = BNXT_QPLIB_HWRM_PG_SIZE_4K;
+ struct bnxt_qplib_pbl *pbl;
+
+ pbl = &hwq->pbl[PBL_LVL_0];
+ switch (pbl->pg_size) {
+ case ROCE_PG_SIZE_4K:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_4K;
+ break;
+ case ROCE_PG_SIZE_8K:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_8K;
+ break;
+ case ROCE_PG_SIZE_64K:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_64K;
+ break;
+ case ROCE_PG_SIZE_2M:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_2M;
+ break;
+ case ROCE_PG_SIZE_8M:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_8M;
+ break;
+ case ROCE_PG_SIZE_1G:
+ pg_size = BNXT_QPLIB_HWRM_PG_SIZE_1G;
+ break;
+ default:
+ break;
+ }
+
+ return pg_size;
+}
+
+static inline void *bnxt_qplib_get_qe(struct bnxt_qplib_hwq *hwq,
+ u32 indx, u64 *pg)
+{
+ u32 pg_num, pg_idx;
+
+ pg_num = (indx / hwq->qe_ppg);
+ pg_idx = (indx % hwq->qe_ppg);
+ if (pg)
+ *pg = (u64)&hwq->pbl_ptr[pg_num];
+ return (void *)(hwq->pbl_ptr[pg_num] + hwq->element_size * pg_idx);
+}
#define to_bnxt_qplib(ptr, type, member) \
container_of(ptr, type, member)
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
index 66954ff6a2f2..4cd475ea97a2 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
@@ -132,9 +132,6 @@ int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw,
attr->max_raw_ethy_qp = le32_to_cpu(sb->max_raw_eth_qp);
attr->max_ah = le32_to_cpu(sb->max_ah);
- attr->max_fmr = le32_to_cpu(sb->max_fmr);
- attr->max_map_per_fmr = sb->max_map_per_fmr;
-
attr->max_srq = le16_to_cpu(sb->max_srq);
attr->max_srq_wqes = le32_to_cpu(sb->max_srq_wr) - 1;
attr->max_srq_sges = sb->max_srq_sge;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.h b/drivers/infiniband/hw/bnxt_re/qplib_sp.h
index 13d9432d5ce2..6404f0da1051 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_sp.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.h
@@ -64,8 +64,6 @@ struct bnxt_qplib_dev_attr {
u32 max_mw;
u32 max_raw_ethy_qp;
u32 max_ah;
- u32 max_fmr;
- u32 max_map_per_fmr;
u32 max_srq;
u32 max_srq_wqes;
u32 max_srq_sges;
diff --git a/drivers/infiniband/hw/bnxt_re/roce_hsi.h b/drivers/infiniband/hw/bnxt_re/roce_hsi.h
index e4b09e7c2175..6f00f07420b7 100644
--- a/drivers/infiniband/hw/bnxt_re/roce_hsi.h
+++ b/drivers/infiniband/hw/bnxt_re/roce_hsi.h
@@ -210,6 +210,20 @@ struct sq_send {
__le32 data[24];
};
+/* sq_send_hdr (size:256b/32B) */
+struct sq_send_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 wqe_size;
+ u8 reserved8_1;
+ __le32 inv_key_or_imm_data;
+ __le32 length;
+ __le32 q_key;
+ __le32 dst_qp;
+ __le32 avid;
+ __le64 reserved64;
+};
+
/* Send Raw Ethernet and QP1 SQ WQE (40 bytes) */
struct sq_send_raweth_qp1 {
u8 wqe_type;
@@ -265,6 +279,21 @@ struct sq_send_raweth_qp1 {
__le32 data[24];
};
+/* sq_send_raweth_qp1_hdr (size:256b/32B) */
+struct sq_send_raweth_qp1_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 wqe_size;
+ u8 reserved8;
+ __le16 lflags;
+ __le16 cfa_action;
+ __le32 length;
+ __le32 reserved32_1;
+ __le32 cfa_meta;
+ __le32 reserved32_2;
+ __le64 reserved64;
+};
+
/* RDMA SQ WQE (40 bytes) */
struct sq_rdma {
u8 wqe_type;
@@ -288,6 +317,20 @@ struct sq_rdma {
__le32 data[24];
};
+/* sq_rdma_hdr (size:256b/32B) */
+struct sq_rdma_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 wqe_size;
+ u8 reserved8;
+ __le32 imm_data;
+ __le32 length;
+ __le32 reserved32_1;
+ __le64 remote_va;
+ __le32 remote_key;
+ __le32 reserved32_2;
+};
+
/* Atomic SQ WQE (40 bytes) */
struct sq_atomic {
u8 wqe_type;
@@ -307,6 +350,17 @@ struct sq_atomic {
__le32 data[24];
};
+/* sq_atomic_hdr (size:256b/32B) */
+struct sq_atomic_hdr {
+ u8 wqe_type;
+ u8 flags;
+ __le16 reserved16;
+ __le32 remote_key;
+ __le64 remote_va;
+ __le64 swap_data;
+ __le64 cmp_data;
+};
+
/* Local Invalidate SQ WQE (40 bytes) */
struct sq_localinvalidate {
u8 wqe_type;
@@ -324,6 +378,16 @@ struct sq_localinvalidate {
__le32 data[24];
};
+/* sq_localinvalidate_hdr (size:256b/32B) */
+struct sq_localinvalidate_hdr {
+ u8 wqe_type;
+ u8 flags;
+ __le16 reserved16;
+ __le32 inv_l_key;
+ __le64 reserved64;
+ u8 reserved128[16];
+};
+
/* FR-PMR SQ WQE (40 bytes) */
struct sq_fr_pmr {
u8 wqe_type;
@@ -380,6 +444,21 @@ struct sq_fr_pmr {
__le32 data[24];
};
+/* sq_fr_pmr_hdr (size:256b/32B) */
+struct sq_fr_pmr_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 access_cntl;
+ u8 zero_based_page_size_log;
+ __le32 l_key;
+ u8 length[5];
+ u8 reserved8_1;
+ u8 reserved8_2;
+ u8 numlevels_pbl_page_size_log;
+ __le64 pblptr;
+ __le64 va;
+};
+
/* Bind SQ WQE (40 bytes) */
struct sq_bind {
u8 wqe_type;
@@ -417,6 +496,22 @@ struct sq_bind {
#define SQ_BIND_DATA_SFT 0
};
+/* sq_bind_hdr (size:256b/32B) */
+struct sq_bind_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 access_cntl;
+ u8 reserved8_1;
+ u8 mw_type_zero_based;
+ u8 reserved8_2;
+ __le16 reserved16;
+ __le32 parent_l_key;
+ __le32 l_key;
+ __le64 va;
+ u8 length[5];
+ u8 reserved24[3];
+};
+
/* RQ/SRQ WQE Structures */
/* RQ/SRQ WQE (40 bytes) */
struct rq_wqe {
@@ -435,6 +530,17 @@ struct rq_wqe {
__le32 data[24];
};
+/* rq_wqe_hdr (size:256b/32B) */
+struct rq_wqe_hdr {
+ u8 wqe_type;
+ u8 flags;
+ u8 wqe_size;
+ u8 reserved8;
+ __le32 reserved32;
+ __le32 wr_id[2];
+ u8 reserved128[16];
+};
+
/* CQ CQE Structures */
/* Base CQE (32 bytes) */
struct cq_base {
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 599340c1f0b8..541dbcf22d0e 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -953,6 +953,7 @@ void c4iw_dealloc(struct uld_ctx *ctx)
static void c4iw_remove(struct uld_ctx *ctx)
{
pr_debug("c4iw_dev %p\n", ctx->dev);
+ debugfs_remove_recursive(ctx->dev->debugfs_root);
c4iw_unregister_device(ctx->dev);
c4iw_dealloc(ctx);
}
diff --git a/drivers/infiniband/hw/efa/efa.h b/drivers/infiniband/hw/efa/efa.h
index aa7396a1588a..1889dd172a25 100644
--- a/drivers/infiniband/hw/efa/efa.h
+++ b/drivers/infiniband/hw/efa/efa.h
@@ -1,6 +1,6 @@
/* SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause */
/*
- * Copyright 2018-2019 Amazon.com, Inc. or its affiliates. All rights reserved.
+ * Copyright 2018-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
*/
#ifndef _EFA_H_
@@ -40,6 +40,7 @@ struct efa_sw_stats {
atomic64_t reg_mr_err;
atomic64_t alloc_ucontext_err;
atomic64_t create_ah_err;
+ atomic64_t mmap_err;
};
/* Don't use anything other than atomic64 */
@@ -153,8 +154,7 @@ int efa_mmap(struct ib_ucontext *ibucontext,
struct vm_area_struct *vma);
void efa_mmap_free(struct rdma_user_mmap_entry *rdma_entry);
int efa_create_ah(struct ib_ah *ibah,
- struct rdma_ah_attr *ah_attr,
- u32 flags,
+ struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
void efa_destroy_ah(struct ib_ah *ibah, u32 flags);
int efa_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
diff --git a/drivers/infiniband/hw/efa/efa_admin_cmds_defs.h b/drivers/infiniband/hw/efa/efa_admin_cmds_defs.h
index 96b104ab5415..bef2bd291054 100644
--- a/drivers/infiniband/hw/efa/efa_admin_cmds_defs.h
+++ b/drivers/infiniband/hw/efa/efa_admin_cmds_defs.h
@@ -37,7 +37,7 @@ enum efa_admin_aq_feature_id {
EFA_ADMIN_NETWORK_ATTR = 3,
EFA_ADMIN_QUEUE_ATTR = 4,
EFA_ADMIN_HW_HINTS = 5,
- EFA_ADMIN_FEATURES_OPCODE_NUM = 8,
+ EFA_ADMIN_HOST_INFO = 6,
};
/* QP transport type */
@@ -799,6 +799,54 @@ struct efa_admin_mmio_req_read_less_resp {
u32 reg_val;
};
+enum efa_admin_os_type {
+ EFA_ADMIN_OS_LINUX = 0,
+};
+
+struct efa_admin_host_info {
+ /* OS distribution string format */
+ u8 os_dist_str[128];
+
+ /* Defined in enum efa_admin_os_type */
+ u32 os_type;
+
+ /* Kernel version string format */
+ u8 kernel_ver_str[32];
+
+ /* Kernel version numeric format */
+ u32 kernel_ver;
+
+ /*
+ * 7:0 : driver_module_type
+ * 15:8 : driver_sub_minor
+ * 23:16 : driver_minor
+ * 31:24 : driver_major
+ */
+ u32 driver_ver;
+
+ /*
+ * Device's Bus, Device and Function
+ * 2:0 : function
+ * 7:3 : device
+ * 15:8 : bus
+ */
+ u16 bdf;
+
+ /*
+ * Spec version
+ * 7:0 : spec_minor
+ * 15:8 : spec_major
+ */
+ u16 spec_ver;
+
+ /*
+ * 0 : intree - Intree driver
+ * 1 : gdr - GPUDirect RDMA supported
+ * 31:2 : reserved2
+ */
+ u32 flags;
+};
+
/* create_qp_cmd */
#define EFA_ADMIN_CREATE_QP_CMD_SQ_VIRT_MASK BIT(0)
#define EFA_ADMIN_CREATE_QP_CMD_RQ_VIRT_MASK BIT(1)
@@ -820,4 +868,17 @@ struct efa_admin_mmio_req_read_less_resp {
/* feature_device_attr_desc */
#define EFA_ADMIN_FEATURE_DEVICE_ATTR_DESC_RDMA_READ_MASK BIT(0)
+/* host_info */
+#define EFA_ADMIN_HOST_INFO_DRIVER_MODULE_TYPE_MASK GENMASK(7, 0)
+#define EFA_ADMIN_HOST_INFO_DRIVER_SUB_MINOR_MASK GENMASK(15, 8)
+#define EFA_ADMIN_HOST_INFO_DRIVER_MINOR_MASK GENMASK(23, 16)
+#define EFA_ADMIN_HOST_INFO_DRIVER_MAJOR_MASK GENMASK(31, 24)
+#define EFA_ADMIN_HOST_INFO_FUNCTION_MASK GENMASK(2, 0)
+#define EFA_ADMIN_HOST_INFO_DEVICE_MASK GENMASK(7, 3)
+#define EFA_ADMIN_HOST_INFO_BUS_MASK GENMASK(15, 8)
+#define EFA_ADMIN_HOST_INFO_SPEC_MINOR_MASK GENMASK(7, 0)
+#define EFA_ADMIN_HOST_INFO_SPEC_MAJOR_MASK GENMASK(15, 8)
+#define EFA_ADMIN_HOST_INFO_INTREE_MASK BIT(0)
+#define EFA_ADMIN_HOST_INFO_GDR_MASK BIT(1)
+
#endif /* _EFA_ADMIN_CMDS_H_ */
diff --git a/drivers/infiniband/hw/efa/efa_com.c b/drivers/infiniband/hw/efa/efa_com.c
index 7fce69f5568f..336bc2c57bb1 100644
--- a/drivers/infiniband/hw/efa/efa_com.c
+++ b/drivers/infiniband/hw/efa/efa_com.c
@@ -631,17 +631,20 @@ int efa_com_cmd_exec(struct efa_com_admin_queue *aq,
cmd->aq_common_descriptor.opcode, PTR_ERR(comp_ctx));
up(&aq->avail_cmds);
+ atomic64_inc(&aq->stats.cmd_err);
return PTR_ERR(comp_ctx);
}
err = efa_com_wait_and_process_admin_cq(comp_ctx, aq);
- if (err)
+ if (err) {
ibdev_err_ratelimited(
aq->efa_dev,
"Failed to process command %s (opcode %u) comp_status %d err %d\n",
efa_com_cmd_str(cmd->aq_common_descriptor.opcode),
cmd->aq_common_descriptor.opcode, comp_ctx->comp_status,
err);
+ atomic64_inc(&aq->stats.cmd_err);
+ }
up(&aq->avail_cmds);
diff --git a/drivers/infiniband/hw/efa/efa_com.h b/drivers/infiniband/hw/efa/efa_com.h
index c67dd8109d1c..5e4c88877ddb 100644
--- a/drivers/infiniband/hw/efa/efa_com.h
+++ b/drivers/infiniband/hw/efa/efa_com.h
@@ -1,6 +1,6 @@
/* SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause */
/*
- * Copyright 2018-2019 Amazon.com, Inc. or its affiliates. All rights reserved.
+ * Copyright 2018-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
*/
#ifndef _EFA_COM_H_
@@ -47,6 +47,7 @@ struct efa_com_admin_sq {
struct efa_com_stats_admin {
atomic64_t submitted_cmd;
atomic64_t completed_cmd;
+ atomic64_t cmd_err;
atomic64_t no_completion;
};
diff --git a/drivers/infiniband/hw/efa/efa_com_cmd.c b/drivers/infiniband/hw/efa/efa_com_cmd.c
index eea5574a62e8..fabd8df2e78f 100644
--- a/drivers/infiniband/hw/efa/efa_com_cmd.c
+++ b/drivers/infiniband/hw/efa/efa_com_cmd.c
@@ -351,7 +351,7 @@ int efa_com_destroy_ah(struct efa_com_dev *edev,
return 0;
}
-static bool
+bool
efa_com_check_supported_feature_id(struct efa_com_dev *edev,
enum efa_admin_aq_feature_id feature_id)
{
@@ -388,7 +388,7 @@ static int efa_com_get_feature_ex(struct efa_com_dev *edev,
if (control_buff_size)
EFA_SET(&get_cmd.aq_common_descriptor.flags,
- EFA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT, 1);
+ EFA_ADMIN_AQ_COMMON_DESC_CTRL_DATA, 1);
efa_com_set_dma_addr(control_buf_dma_addr,
&get_cmd.control_buffer.address.mem_addr_high,
@@ -517,12 +517,12 @@ int efa_com_get_hw_hints(struct efa_com_dev *edev,
return 0;
}
-static int efa_com_set_feature_ex(struct efa_com_dev *edev,
- struct efa_admin_set_feature_resp *set_resp,
- struct efa_admin_set_feature_cmd *set_cmd,
- enum efa_admin_aq_feature_id feature_id,
- dma_addr_t control_buf_dma_addr,
- u32 control_buff_size)
+int efa_com_set_feature_ex(struct efa_com_dev *edev,
+ struct efa_admin_set_feature_resp *set_resp,
+ struct efa_admin_set_feature_cmd *set_cmd,
+ enum efa_admin_aq_feature_id feature_id,
+ dma_addr_t control_buf_dma_addr,
+ u32 control_buff_size)
{
struct efa_com_admin_queue *aq;
int err;
@@ -540,7 +540,7 @@ static int efa_com_set_feature_ex(struct efa_com_dev *edev,
if (control_buff_size) {
set_cmd->aq_common_descriptor.flags = 0;
EFA_SET(&set_cmd->aq_common_descriptor.flags,
- EFA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT, 1);
+ EFA_ADMIN_AQ_COMMON_DESC_CTRL_DATA, 1);
efa_com_set_dma_addr(control_buf_dma_addr,
&set_cmd->control_buffer.address.mem_addr_high,
&set_cmd->control_buffer.address.mem_addr_low);
diff --git a/drivers/infiniband/hw/efa/efa_com_cmd.h b/drivers/infiniband/hw/efa/efa_com_cmd.h
index 31db5a0cbd5b..41ce4a476ee6 100644
--- a/drivers/infiniband/hw/efa/efa_com_cmd.h
+++ b/drivers/infiniband/hw/efa/efa_com_cmd.h
@@ -1,6 +1,6 @@
/* SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause */
/*
- * Copyright 2018-2019 Amazon.com, Inc. or its affiliates. All rights reserved.
+ * Copyright 2018-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
*/
#ifndef _EFA_COM_CMD_H_
@@ -270,6 +270,15 @@ int efa_com_get_device_attr(struct efa_com_dev *edev,
struct efa_com_get_device_attr_result *result);
int efa_com_get_hw_hints(struct efa_com_dev *edev,
struct efa_com_get_hw_hints_result *result);
+bool
+efa_com_check_supported_feature_id(struct efa_com_dev *edev,
+ enum efa_admin_aq_feature_id feature_id);
+int efa_com_set_feature_ex(struct efa_com_dev *edev,
+ struct efa_admin_set_feature_resp *set_resp,
+ struct efa_admin_set_feature_cmd *set_cmd,
+ enum efa_admin_aq_feature_id feature_id,
+ dma_addr_t control_buf_dma_addr,
+ u32 control_buff_size);
int efa_com_set_aenq_config(struct efa_com_dev *edev, u32 groups);
int efa_com_alloc_pd(struct efa_com_dev *edev,
struct efa_com_alloc_pd_result *result);
diff --git a/drivers/infiniband/hw/efa/efa_main.c b/drivers/infiniband/hw/efa/efa_main.c
index faf3ff1bca2a..82145574c928 100644
--- a/drivers/infiniband/hw/efa/efa_main.c
+++ b/drivers/infiniband/hw/efa/efa_main.c
@@ -1,10 +1,12 @@
// SPDX-License-Identifier: GPL-2.0 OR BSD-2-Clause
/*
- * Copyright 2018-2019 Amazon.com, Inc. or its affiliates. All rights reserved.
+ * Copyright 2018-2020 Amazon.com, Inc. or its affiliates. All rights reserved.
*/
#include <linux/module.h>
#include <linux/pci.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
#include <rdma/ib_user_verbs.h>
@@ -187,6 +189,52 @@ static void efa_stats_init(struct efa_dev *dev)
atomic64_set(s, 0);
}
+static void efa_set_host_info(struct efa_dev *dev)
+{
+ struct efa_admin_set_feature_resp resp = {};
+ struct efa_admin_set_feature_cmd cmd = {};
+ struct efa_admin_host_info *hinf;
+ u32 bufsz = sizeof(*hinf);
+ dma_addr_t hinf_dma;
+
+ if (!efa_com_check_supported_feature_id(&dev->edev,
+ EFA_ADMIN_HOST_INFO))
+ return;
+
+ /* Failures in host info set shall not disturb probe */
+ hinf = dma_alloc_coherent(&dev->pdev->dev, bufsz, &hinf_dma,
+ GFP_KERNEL);
+ if (!hinf)
+ return;
+
+ strlcpy(hinf->os_dist_str, utsname()->release,
+ min(sizeof(hinf->os_dist_str), sizeof(utsname()->release)));
+ hinf->os_type = EFA_ADMIN_OS_LINUX;
+ strlcpy(hinf->kernel_ver_str, utsname()->version,
+ min(sizeof(hinf->kernel_ver_str), sizeof(utsname()->version)));
+ hinf->kernel_ver = LINUX_VERSION_CODE;
+ EFA_SET(&hinf->driver_ver, EFA_ADMIN_HOST_INFO_DRIVER_MAJOR, 0);
+ EFA_SET(&hinf->driver_ver, EFA_ADMIN_HOST_INFO_DRIVER_MINOR, 0);
+ EFA_SET(&hinf->driver_ver, EFA_ADMIN_HOST_INFO_DRIVER_SUB_MINOR, 0);
+ EFA_SET(&hinf->driver_ver, EFA_ADMIN_HOST_INFO_DRIVER_MODULE_TYPE, 0);
+ EFA_SET(&hinf->bdf, EFA_ADMIN_HOST_INFO_BUS, dev->pdev->bus->number);
+ EFA_SET(&hinf->bdf, EFA_ADMIN_HOST_INFO_DEVICE,
+ PCI_SLOT(dev->pdev->devfn));
+ EFA_SET(&hinf->bdf, EFA_ADMIN_HOST_INFO_FUNCTION,
+ PCI_FUNC(dev->pdev->devfn));
+ EFA_SET(&hinf->spec_ver, EFA_ADMIN_HOST_INFO_SPEC_MAJOR,
+ EFA_COMMON_SPEC_VERSION_MAJOR);
+ EFA_SET(&hinf->spec_ver, EFA_ADMIN_HOST_INFO_SPEC_MINOR,
+ EFA_COMMON_SPEC_VERSION_MINOR);
+ EFA_SET(&hinf->flags, EFA_ADMIN_HOST_INFO_INTREE, 1);
+ EFA_SET(&hinf->flags, EFA_ADMIN_HOST_INFO_GDR, 0);
+
+ efa_com_set_feature_ex(&dev->edev, &resp, &cmd, EFA_ADMIN_HOST_INFO,
+ hinf_dma, bufsz);
+
+ dma_free_coherent(&dev->pdev->dev, bufsz, hinf, hinf_dma);
+}
+
static const struct ib_device_ops efa_dev_ops = {
.owner = THIS_MODULE,
.driver_id = RDMA_DRIVER_EFA,
@@ -251,6 +299,8 @@ static int efa_ib_device_add(struct efa_dev *dev)
if (err)
goto err_release_doorbell_bar;
+ efa_set_host_info(dev);
+
dev->ibdev.node_type = RDMA_NODE_UNSPECIFIED;
dev->ibdev.phys_port_cnt = 1;
dev->ibdev.num_comp_vectors = 1;
diff --git a/drivers/infiniband/hw/efa/efa_verbs.c b/drivers/infiniband/hw/efa/efa_verbs.c
index 5c57098a4aee..08313f7c73bc 100644
--- a/drivers/infiniband/hw/efa/efa_verbs.c
+++ b/drivers/infiniband/hw/efa/efa_verbs.c
@@ -37,13 +37,16 @@ struct efa_user_mmap_entry {
op(EFA_RX_DROPS, "rx_drops") \
op(EFA_SUBMITTED_CMDS, "submitted_cmds") \
op(EFA_COMPLETED_CMDS, "completed_cmds") \
+ op(EFA_CMDS_ERR, "cmds_err") \
op(EFA_NO_COMPLETION_CMDS, "no_completion_cmds") \
op(EFA_KEEP_ALIVE_RCVD, "keep_alive_rcvd") \
op(EFA_ALLOC_PD_ERR, "alloc_pd_err") \
op(EFA_CREATE_QP_ERR, "create_qp_err") \
+ op(EFA_CREATE_CQ_ERR, "create_cq_err") \
op(EFA_REG_MR_ERR, "reg_mr_err") \
op(EFA_ALLOC_UCONTEXT_ERR, "alloc_ucontext_err") \
- op(EFA_CREATE_AH_ERR, "create_ah_err")
+ op(EFA_CREATE_AH_ERR, "create_ah_err") \
+ op(EFA_MMAP_ERR, "mmap_err")
#define EFA_STATS_ENUM(ename, name) ename,
#define EFA_STATS_STR(ename, name) [ename] = name,
@@ -1568,6 +1571,7 @@ static int __efa_mmap(struct efa_dev *dev, struct efa_ucontext *ucontext,
ibdev_dbg(&dev->ibdev,
"pgoff[%#lx] does not have valid entry\n",
vma->vm_pgoff);
+ atomic64_inc(&dev->stats.sw_stats.mmap_err);
return -EINVAL;
}
entry = to_emmap(rdma_entry);
@@ -1603,12 +1607,14 @@ static int __efa_mmap(struct efa_dev *dev, struct efa_ucontext *ucontext,
err = -EINVAL;
}
- if (err)
+ if (err) {
ibdev_dbg(
&dev->ibdev,
"Couldn't mmap address[%#llx] length[%#zx] mmap_flag[%d] err[%d]\n",
entry->address, rdma_entry->npages * PAGE_SIZE,
entry->mmap_flag, err);
+ atomic64_inc(&dev->stats.sw_stats.mmap_err);
+ }
rdma_user_mmap_entry_put(rdma_entry);
return err;
@@ -1639,10 +1645,10 @@ static int efa_ah_destroy(struct efa_dev *dev, struct efa_ah *ah)
}
int efa_create_ah(struct ib_ah *ibah,
- struct rdma_ah_attr *ah_attr,
- u32 flags,
+ struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata)
{
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
struct efa_dev *dev = to_edev(ibah->device);
struct efa_com_create_ah_params params = {};
struct efa_ibv_create_ah_resp resp = {};
@@ -1650,7 +1656,7 @@ int efa_create_ah(struct ib_ah *ibah,
struct efa_ah *ah = to_eah(ibah);
int err;
- if (!(flags & RDMA_CREATE_AH_SLEEPABLE)) {
+ if (!(init_attr->flags & RDMA_CREATE_AH_SLEEPABLE)) {
ibdev_dbg(&dev->ibdev,
"Create address handle is not supported in atomic context\n");
err = -EOPNOTSUPP;
@@ -1747,15 +1753,18 @@ int efa_get_hw_stats(struct ib_device *ibdev, struct rdma_hw_stats *stats,
as = &dev->edev.aq.stats;
stats->value[EFA_SUBMITTED_CMDS] = atomic64_read(&as->submitted_cmd);
stats->value[EFA_COMPLETED_CMDS] = atomic64_read(&as->completed_cmd);
+ stats->value[EFA_CMDS_ERR] = atomic64_read(&as->cmd_err);
stats->value[EFA_NO_COMPLETION_CMDS] = atomic64_read(&as->no_completion);
s = &dev->stats;
stats->value[EFA_KEEP_ALIVE_RCVD] = atomic64_read(&s->keep_alive_rcvd);
stats->value[EFA_ALLOC_PD_ERR] = atomic64_read(&s->sw_stats.alloc_pd_err);
stats->value[EFA_CREATE_QP_ERR] = atomic64_read(&s->sw_stats.create_qp_err);
+ stats->value[EFA_CREATE_CQ_ERR] = atomic64_read(&s->sw_stats.create_cq_err);
stats->value[EFA_REG_MR_ERR] = atomic64_read(&s->sw_stats.reg_mr_err);
stats->value[EFA_ALLOC_UCONTEXT_ERR] = atomic64_read(&s->sw_stats.alloc_ucontext_err);
stats->value[EFA_CREATE_AH_ERR] = atomic64_read(&s->sw_stats.create_ah_err);
+ stats->value[EFA_MMAP_ERR] = atomic64_read(&s->sw_stats.mmap_err);
return ARRAY_SIZE(efa_stats_names);
}
diff --git a/drivers/infiniband/hw/hfi1/Makefile b/drivers/infiniband/hw/hfi1/Makefile
index 0405d26d0833..2e89ec10efed 100644
--- a/drivers/infiniband/hw/hfi1/Makefile
+++ b/drivers/infiniband/hw/hfi1/Makefile
@@ -22,9 +22,13 @@ hfi1-y := \
init.o \
intr.o \
iowait.o \
+ ipoib_main.o \
+ ipoib_rx.o \
+ ipoib_tx.o \
mad.o \
mmu_rb.o \
msix.o \
+ netdev_rx.o \
opfn.o \
pcie.o \
pio.o \
diff --git a/drivers/infiniband/hw/hfi1/affinity.c b/drivers/infiniband/hw/hfi1/affinity.c
index 1aeea5d65c01..2a91b8d95e12 100644
--- a/drivers/infiniband/hw/hfi1/affinity.c
+++ b/drivers/infiniband/hw/hfi1/affinity.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -64,6 +64,7 @@ struct hfi1_affinity_node_list node_affinity = {
static const char * const irq_type_names[] = {
"SDMA",
"RCVCTXT",
+ "NETDEVCTXT",
"GENERAL",
"OTHER",
};
@@ -915,6 +916,11 @@ static int get_irq_affinity(struct hfi1_devdata *dd,
set = &entry->rcv_intr;
scnprintf(extra, 64, "ctxt %u", rcd->ctxt);
break;
+ case IRQ_NETDEVCTXT:
+ rcd = (struct hfi1_ctxtdata *)msix->arg;
+ set = &entry->def_intr;
+ scnprintf(extra, 64, "ctxt %u", rcd->ctxt);
+ break;
default:
dd_dev_err(dd, "Invalid IRQ type %d\n", msix->type);
return -EINVAL;
@@ -987,6 +993,10 @@ void hfi1_put_irq_affinity(struct hfi1_devdata *dd,
if (rcd->ctxt != HFI1_CTRL_CTXT)
set = &entry->rcv_intr;
break;
+ case IRQ_NETDEVCTXT:
+ rcd = (struct hfi1_ctxtdata *)msix->arg;
+ set = &entry->def_intr;
+ break;
default:
mutex_unlock(&node_affinity.lock);
return;
diff --git a/drivers/infiniband/hw/hfi1/affinity.h b/drivers/infiniband/hw/hfi1/affinity.h
index 6a7e6ea4e426..f94ed5d7c7a3 100644
--- a/drivers/infiniband/hw/hfi1/affinity.h
+++ b/drivers/infiniband/hw/hfi1/affinity.h
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -52,6 +52,7 @@
enum irq_type {
IRQ_SDMA,
IRQ_RCVCTXT,
+ IRQ_NETDEVCTXT,
IRQ_GENERAL,
IRQ_OTHER
};
diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c
index e0b1238d31df..15f9c635f292 100644
--- a/drivers/infiniband/hw/hfi1/chip.c
+++ b/drivers/infiniband/hw/hfi1/chip.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -66,10 +66,7 @@
#include "affinity.h"
#include "debugfs.h"
#include "fault.h"
-
-uint kdeth_qp;
-module_param_named(kdeth_qp, kdeth_qp, uint, S_IRUGO);
-MODULE_PARM_DESC(kdeth_qp, "Set the KDETH queue pair prefix");
+#include "netdev.h"
uint num_vls = HFI1_MAX_VLS_SUPPORTED;
module_param(num_vls, uint, S_IRUGO);
@@ -128,13 +125,15 @@ struct flag_table {
/*
* RSM instance allocation
- * 0 - Verbs
- * 1 - User Fecn Handling
- * 2 - Vnic
+ * 0 - User Fecn Handling
+ * 1 - Vnic
+ * 2 - AIP
+ * 3 - Verbs
*/
-#define RSM_INS_VERBS 0
-#define RSM_INS_FECN 1
-#define RSM_INS_VNIC 2
+#define RSM_INS_FECN 0
+#define RSM_INS_VNIC 1
+#define RSM_INS_AIP 2
+#define RSM_INS_VERBS 3
/* Bit offset into the GUID which carries HFI id information */
#define GUID_HFI_INDEX_SHIFT 39
@@ -175,6 +174,25 @@ struct flag_table {
/* QPN[m+n:1] QW 1, OFFSET 1 */
#define QPN_SELECT_OFFSET ((1ull << QW_SHIFT) | (1ull))
+/* RSM fields for AIP */
+/* LRH.BTH above is reused for this rule */
+
+/* BTH.DESTQP: QW 1, OFFSET 16 for match */
+#define BTH_DESTQP_QW 1ull
+#define BTH_DESTQP_BIT_OFFSET 16ull
+#define BTH_DESTQP_OFFSET(off) ((BTH_DESTQP_QW << QW_SHIFT) | (off))
+#define BTH_DESTQP_MATCH_OFFSET BTH_DESTQP_OFFSET(BTH_DESTQP_BIT_OFFSET)
+#define BTH_DESTQP_MASK 0xFFull
+#define BTH_DESTQP_VALUE 0x81ull
+
+/* DETH.SQPN: QW 1 Offset 56 for select */
+/* We use 8 most significant Soure QPN bits as entropy fpr AIP */
+#define DETH_AIP_SQPN_QW 3ull
+#define DETH_AIP_SQPN_BIT_OFFSET 56ull
+#define DETH_AIP_SQPN_OFFSET(off) ((DETH_AIP_SQPN_QW << QW_SHIFT) | (off))
+#define DETH_AIP_SQPN_SELECT_OFFSET \
+ DETH_AIP_SQPN_OFFSET(DETH_AIP_SQPN_BIT_OFFSET)
+
/* RSM fields for Vnic */
/* L2_TYPE: QW 0, OFFSET 61 - for match */
#define L2_TYPE_QW 0ull
@@ -8463,6 +8481,49 @@ static void hfi1_rcd_eoi_intr(struct hfi1_ctxtdata *rcd)
local_irq_restore(flags);
}
+/**
+ * hfi1_netdev_rx_napi - napi poll function to move eoi inline
+ * @napi - pointer to napi object
+ * @budget - netdev budget
+ */
+int hfi1_netdev_rx_napi(struct napi_struct *napi, int budget)
+{
+ struct hfi1_netdev_rxq *rxq = container_of(napi,
+ struct hfi1_netdev_rxq, napi);
+ struct hfi1_ctxtdata *rcd = rxq->rcd;
+ int work_done = 0;
+
+ work_done = rcd->do_interrupt(rcd, budget);
+
+ if (work_done < budget) {
+ napi_complete_done(napi, work_done);
+ hfi1_rcd_eoi_intr(rcd);
+ }
+
+ return work_done;
+}
+
+/* Receive packet napi handler for netdevs VNIC and AIP */
+irqreturn_t receive_context_interrupt_napi(int irq, void *data)
+{
+ struct hfi1_ctxtdata *rcd = data;
+
+ receive_interrupt_common(rcd);
+
+ if (likely(rcd->napi)) {
+ if (likely(napi_schedule_prep(rcd->napi)))
+ __napi_schedule_irqoff(rcd->napi);
+ else
+ __hfi1_rcd_eoi_intr(rcd);
+ } else {
+ WARN_ONCE(1, "Napi IRQ handler without napi set up ctxt=%d\n",
+ rcd->ctxt);
+ __hfi1_rcd_eoi_intr(rcd);
+ }
+
+ return IRQ_HANDLED;
+}
+
/*
* Receive packet IRQ handler. This routine expects to be on its own IRQ.
* This routine will try to handle packets immediately (latency), but if
@@ -13330,13 +13391,12 @@ static int set_up_interrupts(struct hfi1_devdata *dd)
* in array of contexts
* freectxts - number of free user contexts
* num_send_contexts - number of PIO send contexts being used
- * num_vnic_contexts - number of contexts reserved for VNIC
+ * num_netdev_contexts - number of contexts reserved for netdev
*/
static int set_up_context_variables(struct hfi1_devdata *dd)
{
unsigned long num_kernel_contexts;
- u16 num_vnic_contexts = HFI1_NUM_VNIC_CTXT;
- int total_contexts;
+ u16 num_netdev_contexts;
int ret;
unsigned ngroups;
int rmt_count;
@@ -13373,13 +13433,6 @@ static int set_up_context_variables(struct hfi1_devdata *dd)
num_kernel_contexts = send_contexts - num_vls - 1;
}
- /* Accommodate VNIC contexts if possible */
- if ((num_kernel_contexts + num_vnic_contexts) > rcv_contexts) {
- dd_dev_err(dd, "No receive contexts available for VNIC\n");
- num_vnic_contexts = 0;
- }
- total_contexts = num_kernel_contexts + num_vnic_contexts;
-
/*
* User contexts:
* - default to 1 user context per real (non-HT) CPU core if
@@ -13392,28 +13445,32 @@ static int set_up_context_variables(struct hfi1_devdata *dd)
/*
* Adjust the counts given a global max.
*/
- if (total_contexts + n_usr_ctxts > rcv_contexts) {
+ if (num_kernel_contexts + n_usr_ctxts > rcv_contexts) {
dd_dev_err(dd,
- "Reducing # user receive contexts to: %d, from %u\n",
- rcv_contexts - total_contexts,
+ "Reducing # user receive contexts to: %u, from %u\n",
+ (u32)(rcv_contexts - num_kernel_contexts),
n_usr_ctxts);
/* recalculate */
- n_usr_ctxts = rcv_contexts - total_contexts;
+ n_usr_ctxts = rcv_contexts - num_kernel_contexts;
}
+ num_netdev_contexts =
+ hfi1_num_netdev_contexts(dd, rcv_contexts -
+ (num_kernel_contexts + n_usr_ctxts),
+ &node_affinity.real_cpu_mask);
/*
* The RMT entries are currently allocated as shown below:
* 1. QOS (0 to 128 entries);
* 2. FECN (num_kernel_context - 1 + num_user_contexts +
- * num_vnic_contexts);
- * 3. VNIC (num_vnic_contexts).
- * It should be noted that FECN oversubscribe num_vnic_contexts
- * entries of RMT because both VNIC and PSM could allocate any receive
+ * num_netdev_contexts);
+ * 3. netdev (num_netdev_contexts).
+ * It should be noted that FECN oversubscribe num_netdev_contexts
+ * entries of RMT because both netdev and PSM could allocate any receive
* context between dd->first_dyn_alloc_text and dd->num_rcv_contexts,
* and PSM FECN must reserve an RMT entry for each possible PSM receive
* context.
*/
- rmt_count = qos_rmt_entries(dd, NULL, NULL) + (num_vnic_contexts * 2);
+ rmt_count = qos_rmt_entries(dd, NULL, NULL) + (num_netdev_contexts * 2);
if (HFI1_CAP_IS_KSET(TID_RDMA))
rmt_count += num_kernel_contexts - 1;
if (rmt_count + n_usr_ctxts > NUM_MAP_ENTRIES) {
@@ -13426,21 +13483,20 @@ static int set_up_context_variables(struct hfi1_devdata *dd)
n_usr_ctxts = user_rmt_reduced;
}
- total_contexts += n_usr_ctxts;
-
- /* the first N are kernel contexts, the rest are user/vnic contexts */
- dd->num_rcv_contexts = total_contexts;
+ /* the first N are kernel contexts, the rest are user/netdev contexts */
+ dd->num_rcv_contexts =
+ num_kernel_contexts + n_usr_ctxts + num_netdev_contexts;
dd->n_krcv_queues = num_kernel_contexts;
dd->first_dyn_alloc_ctxt = num_kernel_contexts;
- dd->num_vnic_contexts = num_vnic_contexts;
+ dd->num_netdev_contexts = num_netdev_contexts;
dd->num_user_contexts = n_usr_ctxts;
dd->freectxts = n_usr_ctxts;
dd_dev_info(dd,
- "rcv contexts: chip %d, used %d (kernel %d, vnic %u, user %u)\n",
+ "rcv contexts: chip %d, used %d (kernel %d, netdev %u, user %u)\n",
rcv_contexts,
(int)dd->num_rcv_contexts,
(int)dd->n_krcv_queues,
- dd->num_vnic_contexts,
+ dd->num_netdev_contexts,
dd->num_user_contexts);
/*
@@ -14119,21 +14175,12 @@ static void init_early_variables(struct hfi1_devdata *dd)
static void init_kdeth_qp(struct hfi1_devdata *dd)
{
- /* user changed the KDETH_QP */
- if (kdeth_qp != 0 && kdeth_qp >= 0xff) {
- /* out of range or illegal value */
- dd_dev_err(dd, "Invalid KDETH queue pair prefix, ignoring");
- kdeth_qp = 0;
- }
- if (kdeth_qp == 0) /* not set, or failed range check */
- kdeth_qp = DEFAULT_KDETH_QP;
-
write_csr(dd, SEND_BTH_QP,
- (kdeth_qp & SEND_BTH_QP_KDETH_QP_MASK) <<
+ (RVT_KDETH_QP_PREFIX & SEND_BTH_QP_KDETH_QP_MASK) <<
SEND_BTH_QP_KDETH_QP_SHIFT);
write_csr(dd, RCV_BTH_QP,
- (kdeth_qp & RCV_BTH_QP_KDETH_QP_MASK) <<
+ (RVT_KDETH_QP_PREFIX & RCV_BTH_QP_KDETH_QP_MASK) <<
RCV_BTH_QP_KDETH_QP_SHIFT);
}
@@ -14249,6 +14296,12 @@ static void complete_rsm_map_table(struct hfi1_devdata *dd,
}
}
+/* Is a receive side mapping rule */
+static bool has_rsm_rule(struct hfi1_devdata *dd, u8 rule_index)
+{
+ return read_csr(dd, RCV_RSM_CFG + (8 * rule_index)) != 0;
+}
+
/*
* Add a receive side mapping rule.
*/
@@ -14485,77 +14538,138 @@ static void init_fecn_handling(struct hfi1_devdata *dd,
rmt->used += total_cnt;
}
-/* Initialize RSM for VNIC */
-void hfi1_init_vnic_rsm(struct hfi1_devdata *dd)
+static inline bool hfi1_is_rmt_full(int start, int spare)
+{
+ return (start + spare) > NUM_MAP_ENTRIES;
+}
+
+static bool hfi1_netdev_update_rmt(struct hfi1_devdata *dd)
{
u8 i, j;
u8 ctx_id = 0;
u64 reg;
u32 regoff;
- struct rsm_rule_data rrd;
+ int rmt_start = hfi1_netdev_get_free_rmt_idx(dd);
+ int ctxt_count = hfi1_netdev_ctxt_count(dd);
- if (hfi1_vnic_is_rsm_full(dd, NUM_VNIC_MAP_ENTRIES)) {
- dd_dev_err(dd, "Vnic RSM disabled, rmt entries used = %d\n",
- dd->vnic.rmt_start);
- return;
+ /* We already have contexts mapped in RMT */
+ if (has_rsm_rule(dd, RSM_INS_VNIC) || has_rsm_rule(dd, RSM_INS_AIP)) {
+ dd_dev_info(dd, "Contexts are already mapped in RMT\n");
+ return true;
+ }
+
+ if (hfi1_is_rmt_full(rmt_start, NUM_NETDEV_MAP_ENTRIES)) {
+ dd_dev_err(dd, "Not enough RMT entries used = %d\n",
+ rmt_start);
+ return false;
}
- dev_dbg(&(dd)->pcidev->dev, "Vnic rsm start = %d, end %d\n",
- dd->vnic.rmt_start,
- dd->vnic.rmt_start + NUM_VNIC_MAP_ENTRIES);
+ dev_dbg(&(dd)->pcidev->dev, "RMT start = %d, end %d\n",
+ rmt_start,
+ rmt_start + NUM_NETDEV_MAP_ENTRIES);
/* Update RSM mapping table, 32 regs, 256 entries - 1 ctx per byte */
- regoff = RCV_RSM_MAP_TABLE + (dd->vnic.rmt_start / 8) * 8;
+ regoff = RCV_RSM_MAP_TABLE + (rmt_start / 8) * 8;
reg = read_csr(dd, regoff);
- for (i = 0; i < NUM_VNIC_MAP_ENTRIES; i++) {
- /* Update map register with vnic context */
- j = (dd->vnic.rmt_start + i) % 8;
+ for (i = 0; i < NUM_NETDEV_MAP_ENTRIES; i++) {
+ /* Update map register with netdev context */
+ j = (rmt_start + i) % 8;
reg &= ~(0xffllu << (j * 8));
- reg |= (u64)dd->vnic.ctxt[ctx_id++]->ctxt << (j * 8);
- /* Wrap up vnic ctx index */
- ctx_id %= dd->vnic.num_ctxt;
+ reg |= (u64)hfi1_netdev_get_ctxt(dd, ctx_id++)->ctxt << (j * 8);
+ /* Wrap up netdev ctx index */
+ ctx_id %= ctxt_count;
/* Write back map register */
- if (j == 7 || ((i + 1) == NUM_VNIC_MAP_ENTRIES)) {
+ if (j == 7 || ((i + 1) == NUM_NETDEV_MAP_ENTRIES)) {
dev_dbg(&(dd)->pcidev->dev,
- "Vnic rsm map reg[%d] =0x%llx\n",
+ "RMT[%d] =0x%llx\n",
regoff - RCV_RSM_MAP_TABLE, reg);
write_csr(dd, regoff, reg);
regoff += 8;
- if (i < (NUM_VNIC_MAP_ENTRIES - 1))
+ if (i < (NUM_NETDEV_MAP_ENTRIES - 1))
reg = read_csr(dd, regoff);
}
}
- /* Add rule for vnic */
- rrd.offset = dd->vnic.rmt_start;
- rrd.pkt_type = 4;
- /* Match 16B packets */
- rrd.field1_off = L2_TYPE_MATCH_OFFSET;
- rrd.mask1 = L2_TYPE_MASK;
- rrd.value1 = L2_16B_VALUE;
- /* Match ETH L4 packets */
- rrd.field2_off = L4_TYPE_MATCH_OFFSET;
- rrd.mask2 = L4_16B_TYPE_MASK;
- rrd.value2 = L4_16B_ETH_VALUE;
- /* Calc context from veswid and entropy */
- rrd.index1_off = L4_16B_HDR_VESWID_OFFSET;
- rrd.index1_width = ilog2(NUM_VNIC_MAP_ENTRIES);
- rrd.index2_off = L2_16B_ENTROPY_OFFSET;
- rrd.index2_width = ilog2(NUM_VNIC_MAP_ENTRIES);
- add_rsm_rule(dd, RSM_INS_VNIC, &rrd);
-
- /* Enable RSM if not already enabled */
+ return true;
+}
+
+static void hfi1_enable_rsm_rule(struct hfi1_devdata *dd,
+ int rule, struct rsm_rule_data *rrd)
+{
+ if (!hfi1_netdev_update_rmt(dd)) {
+ dd_dev_err(dd, "Failed to update RMT for RSM%d rule\n", rule);
+ return;
+ }
+
+ add_rsm_rule(dd, rule, rrd);
add_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK);
}
+void hfi1_init_aip_rsm(struct hfi1_devdata *dd)
+{
+ /*
+ * go through with the initialisation only if this rule actually doesn't
+ * exist yet
+ */
+ if (atomic_fetch_inc(&dd->ipoib_rsm_usr_num) == 0) {
+ int rmt_start = hfi1_netdev_get_free_rmt_idx(dd);
+ struct rsm_rule_data rrd = {
+ .offset = rmt_start,
+ .pkt_type = IB_PACKET_TYPE,
+ .field1_off = LRH_BTH_MATCH_OFFSET,
+ .mask1 = LRH_BTH_MASK,
+ .value1 = LRH_BTH_VALUE,
+ .field2_off = BTH_DESTQP_MATCH_OFFSET,
+ .mask2 = BTH_DESTQP_MASK,
+ .value2 = BTH_DESTQP_VALUE,
+ .index1_off = DETH_AIP_SQPN_SELECT_OFFSET +
+ ilog2(NUM_NETDEV_MAP_ENTRIES),
+ .index1_width = ilog2(NUM_NETDEV_MAP_ENTRIES),
+ .index2_off = DETH_AIP_SQPN_SELECT_OFFSET,
+ .index2_width = ilog2(NUM_NETDEV_MAP_ENTRIES)
+ };
+
+ hfi1_enable_rsm_rule(dd, RSM_INS_AIP, &rrd);
+ }
+}
+
+/* Initialize RSM for VNIC */
+void hfi1_init_vnic_rsm(struct hfi1_devdata *dd)
+{
+ int rmt_start = hfi1_netdev_get_free_rmt_idx(dd);
+ struct rsm_rule_data rrd = {
+ /* Add rule for vnic */
+ .offset = rmt_start,
+ .pkt_type = 4,
+ /* Match 16B packets */
+ .field1_off = L2_TYPE_MATCH_OFFSET,
+ .mask1 = L2_TYPE_MASK,
+ .value1 = L2_16B_VALUE,
+ /* Match ETH L4 packets */
+ .field2_off = L4_TYPE_MATCH_OFFSET,
+ .mask2 = L4_16B_TYPE_MASK,
+ .value2 = L4_16B_ETH_VALUE,
+ /* Calc context from veswid and entropy */
+ .index1_off = L4_16B_HDR_VESWID_OFFSET,
+ .index1_width = ilog2(NUM_NETDEV_MAP_ENTRIES),
+ .index2_off = L2_16B_ENTROPY_OFFSET,
+ .index2_width = ilog2(NUM_NETDEV_MAP_ENTRIES)
+ };
+
+ hfi1_enable_rsm_rule(dd, RSM_INS_VNIC, &rrd);
+}
+
void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd)
{
clear_rsm_rule(dd, RSM_INS_VNIC);
+}
- /* Disable RSM if used only by vnic */
- if (dd->vnic.rmt_start == 0)
- clear_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK);
+void hfi1_deinit_aip_rsm(struct hfi1_devdata *dd)
+{
+ /* only actually clear the rule if it's the last user asking to do so */
+ if (atomic_fetch_add_unless(&dd->ipoib_rsm_usr_num, -1, 0) == 1)
+ clear_rsm_rule(dd, RSM_INS_AIP);
}
static int init_rxe(struct hfi1_devdata *dd)
@@ -14574,8 +14688,8 @@ static int init_rxe(struct hfi1_devdata *dd)
init_qos(dd, rmt);
init_fecn_handling(dd, rmt);
complete_rsm_map_table(dd, rmt);
- /* record number of used rsm map entries for vnic */
- dd->vnic.rmt_start = rmt->used;
+ /* record number of used rsm map entries for netdev */
+ hfi1_netdev_set_free_rmt_idx(dd, rmt->used);
kfree(rmt);
/*
@@ -15129,6 +15243,10 @@ int hfi1_init_dd(struct hfi1_devdata *dd)
(dd->revision >> CCE_REVISION_SW_SHIFT)
& CCE_REVISION_SW_MASK);
+ /* alloc netdev data */
+ if (hfi1_netdev_alloc(dd))
+ goto bail_cleanup;
+
ret = set_up_context_variables(dd);
if (ret)
goto bail_cleanup;
@@ -15229,6 +15347,7 @@ bail_clear_intr:
hfi1_comp_vectors_clean_up(dd);
msix_clean_up_interrupts(dd);
bail_cleanup:
+ hfi1_netdev_free(dd);
hfi1_pcie_ddcleanup(dd);
bail_free:
hfi1_free_devdata(dd);
diff --git a/drivers/infiniband/hw/hfi1/chip.h b/drivers/infiniband/hw/hfi1/chip.h
index 725509261016..2c6f2de74d4d 100644
--- a/drivers/infiniband/hw/hfi1/chip.h
+++ b/drivers/infiniband/hw/hfi1/chip.h
@@ -1,7 +1,7 @@
#ifndef _CHIP_H
#define _CHIP_H
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -1447,6 +1447,7 @@ irqreturn_t general_interrupt(int irq, void *data);
irqreturn_t sdma_interrupt(int irq, void *data);
irqreturn_t receive_context_interrupt(int irq, void *data);
irqreturn_t receive_context_thread(int irq, void *data);
+irqreturn_t receive_context_interrupt_napi(int irq, void *data);
int set_intr_bits(struct hfi1_devdata *dd, u16 first, u16 last, bool set);
void init_qsfp_int(struct hfi1_devdata *dd);
@@ -1455,6 +1456,8 @@ void remap_intr(struct hfi1_devdata *dd, int isrc, int msix_intr);
void remap_sdma_interrupts(struct hfi1_devdata *dd, int engine, int msix_intr);
void reset_interrupts(struct hfi1_devdata *dd);
u8 hfi1_get_qp_map(struct hfi1_devdata *dd, u8 idx);
+void hfi1_init_aip_rsm(struct hfi1_devdata *dd);
+void hfi1_deinit_aip_rsm(struct hfi1_devdata *dd);
/*
* Interrupt source table.
diff --git a/drivers/infiniband/hw/hfi1/common.h b/drivers/infiniband/hw/hfi1/common.h
index 40a1ff0c8a8e..ff423e546b80 100644
--- a/drivers/infiniband/hw/hfi1/common.h
+++ b/drivers/infiniband/hw/hfi1/common.h
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -72,13 +72,6 @@
* compilation unit
*/
-/*
- * If a packet's QP[23:16] bits match this value, then it is
- * a PSM packet and the hardware will expect a KDETH header
- * following the BTH.
- */
-#define DEFAULT_KDETH_QP 0x80
-
/* driver/hw feature set bitmask */
#define HFI1_CAP_USER_SHIFT 24
#define HFI1_CAP_MASK ((1UL << HFI1_CAP_USER_SHIFT) - 1)
@@ -149,7 +142,8 @@
HFI1_CAP_NO_INTEGRITY | \
HFI1_CAP_PKEY_CHECK | \
HFI1_CAP_TID_RDMA | \
- HFI1_CAP_OPFN) << \
+ HFI1_CAP_OPFN | \
+ HFI1_CAP_AIP) << \
HFI1_CAP_USER_SHIFT)
/*
* Set of capabilities that need to be enabled for kernel context in
@@ -166,6 +160,7 @@
HFI1_CAP_PKEY_CHECK | \
HFI1_CAP_MULTI_PKT_EGR | \
HFI1_CAP_EXTENDED_PSN | \
+ HFI1_CAP_AIP | \
((HFI1_CAP_HDRSUPP | \
HFI1_CAP_MULTI_PKT_EGR | \
HFI1_CAP_STATIC_RATE_CTRL | \
diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c
index 049d15befe58..a40701a6e1b6 100644
--- a/drivers/infiniband/hw/hfi1/driver.c
+++ b/drivers/infiniband/hw/hfi1/driver.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015-2018 Intel Corporation.
+ * Copyright(c) 2015-2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -54,6 +54,7 @@
#include <linux/module.h>
#include <linux/prefetch.h>
#include <rdma/ib_verbs.h>
+#include <linux/etherdevice.h>
#include "hfi.h"
#include "trace.h"
@@ -63,6 +64,9 @@
#include "vnic.h"
#include "fault.h"
+#include "ipoib.h"
+#include "netdev.h"
+
#undef pr_fmt
#define pr_fmt(fmt) DRIVER_NAME ": " fmt
@@ -748,6 +752,39 @@ static noinline int skip_rcv_packet(struct hfi1_packet *packet, int thread)
return ret;
}
+static void process_rcv_packet_napi(struct hfi1_packet *packet)
+{
+ packet->etype = rhf_rcv_type(packet->rhf);
+
+ /* total length */
+ packet->tlen = rhf_pkt_len(packet->rhf); /* in bytes */
+ /* retrieve eager buffer details */
+ packet->etail = rhf_egr_index(packet->rhf);
+ packet->ebuf = get_egrbuf(packet->rcd, packet->rhf,
+ &packet->updegr);
+ /*
+ * Prefetch the contents of the eager buffer. It is
+ * OK to send a negative length to prefetch_range().
+ * The +2 is the size of the RHF.
+ */
+ prefetch_range(packet->ebuf,
+ packet->tlen - ((packet->rcd->rcvhdrqentsize -
+ (rhf_hdrq_offset(packet->rhf)
+ + 2)) * 4));
+
+ packet->rcd->rhf_rcv_function_map[packet->etype](packet);
+ packet->numpkt++;
+
+ /* Set up for the next packet */
+ packet->rhqoff += packet->rsize;
+ if (packet->rhqoff >= packet->maxcnt)
+ packet->rhqoff = 0;
+
+ packet->rhf_addr = (__le32 *)packet->rcd->rcvhdrq + packet->rhqoff +
+ packet->rcd->rhf_offset;
+ packet->rhf = rhf_to_cpu(packet->rhf_addr);
+}
+
static inline int process_rcv_packet(struct hfi1_packet *packet, int thread)
{
int ret;
@@ -827,6 +864,36 @@ static inline void finish_packet(struct hfi1_packet *packet)
}
/*
+ * handle_receive_interrupt_napi_fp - receive a packet
+ * @rcd: the context
+ * @budget: polling budget
+ *
+ * Called from interrupt handler for receive interrupt.
+ * This is the fast path interrupt handler
+ * when executing napi soft irq environment.
+ */
+int handle_receive_interrupt_napi_fp(struct hfi1_ctxtdata *rcd, int budget)
+{
+ struct hfi1_packet packet;
+
+ init_packet(rcd, &packet);
+ if (last_rcv_seq(rcd, rhf_rcv_seq(packet.rhf)))
+ goto bail;
+
+ while (packet.numpkt < budget) {
+ process_rcv_packet_napi(&packet);
+ if (hfi1_seq_incr(rcd, rhf_rcv_seq(packet.rhf)))
+ break;
+
+ process_rcv_update(0, &packet);
+ }
+ hfi1_set_rcd_head(rcd, packet.rhqoff);
+bail:
+ finish_packet(&packet);
+ return packet.numpkt;
+}
+
+/*
* Handle receive interrupts when using the no dma rtail option.
*/
int handle_receive_interrupt_nodma_rtail(struct hfi1_ctxtdata *rcd, int thread)
@@ -1074,6 +1141,63 @@ bail:
}
/*
+ * handle_receive_interrupt_napi_sp - receive a packet
+ * @rcd: the context
+ * @budget: polling budget
+ *
+ * Called from interrupt handler for errors or receive interrupt.
+ * This is the slow path interrupt handler
+ * when executing napi soft irq environment.
+ */
+int handle_receive_interrupt_napi_sp(struct hfi1_ctxtdata *rcd, int budget)
+{
+ struct hfi1_devdata *dd = rcd->dd;
+ int last = RCV_PKT_OK;
+ bool needset = true;
+ struct hfi1_packet packet;
+
+ init_packet(rcd, &packet);
+ if (last_rcv_seq(rcd, rhf_rcv_seq(packet.rhf)))
+ goto bail;
+
+ while (last != RCV_PKT_DONE && packet.numpkt < budget) {
+ if (hfi1_need_drop(dd)) {
+ /* On to the next packet */
+ packet.rhqoff += packet.rsize;
+ packet.rhf_addr = (__le32 *)rcd->rcvhdrq +
+ packet.rhqoff +
+ rcd->rhf_offset;
+ packet.rhf = rhf_to_cpu(packet.rhf_addr);
+
+ } else {
+ if (set_armed_to_active(&packet))
+ goto bail;
+ process_rcv_packet_napi(&packet);
+ }
+
+ if (hfi1_seq_incr(rcd, rhf_rcv_seq(packet.rhf)))
+ last = RCV_PKT_DONE;
+
+ if (needset) {
+ needset = false;
+ set_all_fastpath(dd, rcd);
+ }
+
+ process_rcv_update(last, &packet);
+ }
+
+ hfi1_set_rcd_head(rcd, packet.rhqoff);
+
+bail:
+ /*
+ * Always write head at end, and setup rcv interrupt, even
+ * if no packets were processed.
+ */
+ finish_packet(&packet);
+ return packet.numpkt;
+}
+
+/*
* We may discover in the interrupt that the hardware link state has
* changed from ARMED to ACTIVE (due to the arrival of a non-SC15 packet),
* and we need to update the driver's notion of the link state. We cannot
@@ -1550,6 +1674,82 @@ void handle_eflags(struct hfi1_packet *packet)
show_eflags_errs(packet);
}
+static void hfi1_ipoib_ib_rcv(struct hfi1_packet *packet)
+{
+ struct hfi1_ibport *ibp;
+ struct net_device *netdev;
+ struct hfi1_ctxtdata *rcd = packet->rcd;
+ struct napi_struct *napi = rcd->napi;
+ struct sk_buff *skb;
+ struct hfi1_netdev_rxq *rxq = container_of(napi,
+ struct hfi1_netdev_rxq, napi);
+ u32 extra_bytes;
+ u32 tlen, qpnum;
+ bool do_work, do_cnp;
+ struct hfi1_ipoib_dev_priv *priv;
+
+ trace_hfi1_rcvhdr(packet);
+
+ hfi1_setup_ib_header(packet);
+
+ packet->ohdr = &((struct ib_header *)packet->hdr)->u.oth;
+ packet->grh = NULL;
+
+ if (unlikely(rhf_err_flags(packet->rhf))) {
+ handle_eflags(packet);
+ return;
+ }
+
+ qpnum = ib_bth_get_qpn(packet->ohdr);
+ netdev = hfi1_netdev_get_data(rcd->dd, qpnum);
+ if (!netdev)
+ goto drop_no_nd;
+
+ trace_input_ibhdr(rcd->dd, packet, !!(rhf_dc_info(packet->rhf)));
+ trace_ctxt_rsm_hist(rcd->ctxt);
+
+ /* handle congestion notifications */
+ do_work = hfi1_may_ecn(packet);
+ if (unlikely(do_work)) {
+ do_cnp = (packet->opcode != IB_OPCODE_CNP);
+ (void)hfi1_process_ecn_slowpath(hfi1_ipoib_priv(netdev)->qp,
+ packet, do_cnp);
+ }
+
+ /*
+ * We have split point after last byte of DETH
+ * lets strip padding and CRC and ICRC.
+ * tlen is whole packet len so we need to
+ * subtract header size as well.
+ */
+ tlen = packet->tlen;
+ extra_bytes = ib_bth_get_pad(packet->ohdr) + (SIZE_OF_CRC << 2) +
+ packet->hlen;
+ if (unlikely(tlen < extra_bytes))
+ goto drop;
+
+ tlen -= extra_bytes;
+
+ skb = hfi1_ipoib_prepare_skb(rxq, tlen, packet->ebuf);
+ if (unlikely(!skb))
+ goto drop;
+
+ priv = hfi1_ipoib_priv(netdev);
+ hfi1_ipoib_update_rx_netstats(priv, 1, skb->len);
+
+ skb->dev = netdev;
+ skb->pkt_type = PACKET_HOST;
+ netif_receive_skb(skb);
+
+ return;
+
+drop:
+ ++netdev->stats.rx_dropped;
+drop_no_nd:
+ ibp = rcd_to_iport(packet->rcd);
+ ++ibp->rvp.n_pkt_drops;
+}
+
/*
* The following functions are called by the interrupt handler. They are type
* specific handlers for each packet type.
@@ -1572,28 +1772,10 @@ static void process_receive_ib(struct hfi1_packet *packet)
hfi1_ib_rcv(packet);
}
-static inline bool hfi1_is_vnic_packet(struct hfi1_packet *packet)
-{
- /* Packet received in VNIC context via RSM */
- if (packet->rcd->is_vnic)
- return true;
-
- if ((hfi1_16B_get_l2(packet->ebuf) == OPA_16B_L2_TYPE) &&
- (hfi1_16B_get_l4(packet->ebuf) == OPA_16B_L4_ETHR))
- return true;
-
- return false;
-}
-
static void process_receive_bypass(struct hfi1_packet *packet)
{
struct hfi1_devdata *dd = packet->rcd->dd;
- if (hfi1_is_vnic_packet(packet)) {
- hfi1_vnic_bypass_rcv(packet);
- return;
- }
-
if (hfi1_setup_bypass_packet(packet))
return;
@@ -1757,3 +1939,14 @@ const rhf_rcv_function_ptr normal_rhf_rcv_functions[] = {
[RHF_RCV_TYPE_INVALID6] = process_receive_invalid,
[RHF_RCV_TYPE_INVALID7] = process_receive_invalid,
};
+
+const rhf_rcv_function_ptr netdev_rhf_rcv_functions[] = {
+ [RHF_RCV_TYPE_EXPECTED] = process_receive_invalid,
+ [RHF_RCV_TYPE_EAGER] = process_receive_invalid,
+ [RHF_RCV_TYPE_IB] = hfi1_ipoib_ib_rcv,
+ [RHF_RCV_TYPE_ERROR] = process_receive_error,
+ [RHF_RCV_TYPE_BYPASS] = hfi1_vnic_bypass_rcv,
+ [RHF_RCV_TYPE_INVALID5] = process_receive_invalid,
+ [RHF_RCV_TYPE_INVALID6] = process_receive_invalid,
+ [RHF_RCV_TYPE_INVALID7] = process_receive_invalid,
+};
diff --git a/drivers/infiniband/hw/hfi1/file_ops.c b/drivers/infiniband/hw/hfi1/file_ops.c
index e7fdd70c6e78..8ca51e43cf53 100644
--- a/drivers/infiniband/hw/hfi1/file_ops.c
+++ b/drivers/infiniband/hw/hfi1/file_ops.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015-2017 Intel Corporation.
+ * Copyright(c) 2015-2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -1264,7 +1264,7 @@ static int get_base_info(struct hfi1_filedata *fd, unsigned long arg, u32 len)
memset(&binfo, 0, sizeof(binfo));
binfo.hw_version = dd->revision;
binfo.sw_version = HFI1_KERN_SWVERSION;
- binfo.bthqp = kdeth_qp;
+ binfo.bthqp = RVT_KDETH_QP_PREFIX;
binfo.jkey = uctxt->jkey;
/*
* If more than 64 contexts are enabled the allocated credit
diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h
index b06c2594105a..b4c6bff60a4e 100644
--- a/drivers/infiniband/hw/hfi1/hfi.h
+++ b/drivers/infiniband/hw/hfi1/hfi.h
@@ -1,7 +1,7 @@
#ifndef _HFI1_KERNEL_H
#define _HFI1_KERNEL_H
/*
- * Copyright(c) 2015-2018 Intel Corporation.
+ * Copyright(c) 2015-2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -233,6 +233,8 @@ struct hfi1_ctxtdata {
intr_handler fast_handler;
/** slow handler */
intr_handler slow_handler;
+ /* napi pointer assiociated with netdev */
+ struct napi_struct *napi;
/* verbs rx_stats per rcd */
struct hfi1_opcode_stats_perctx *opstats;
/* clear interrupt mask */
@@ -383,11 +385,11 @@ struct hfi1_packet {
u32 rhqoff;
u32 dlid;
u32 slid;
+ int numpkt;
u16 tlen;
s16 etail;
u16 pkey;
u8 hlen;
- u8 numpkt;
u8 rsize;
u8 updegr;
u8 etype;
@@ -985,7 +987,7 @@ typedef void (*hfi1_make_req)(struct rvt_qp *qp,
struct hfi1_pkt_state *ps,
struct rvt_swqe *wqe);
extern const rhf_rcv_function_ptr normal_rhf_rcv_functions[];
-
+extern const rhf_rcv_function_ptr netdev_rhf_rcv_functions[];
/* return values for the RHF receive functions */
#define RHF_RCV_CONTINUE 0 /* keep going */
@@ -1045,23 +1047,10 @@ struct hfi1_asic_data {
#define NUM_MAP_ENTRIES 256
#define NUM_MAP_REGS 32
-/*
- * Number of VNIC contexts used. Ensure it is less than or equal to
- * max queues supported by VNIC (HFI1_VNIC_MAX_QUEUE).
- */
-#define HFI1_NUM_VNIC_CTXT 8
-
-/* Number of VNIC RSM entries */
-#define NUM_VNIC_MAP_ENTRIES 8
-
/* Virtual NIC information */
struct hfi1_vnic_data {
- struct hfi1_ctxtdata *ctxt[HFI1_NUM_VNIC_CTXT];
struct kmem_cache *txreq_cache;
- struct xarray vesws;
u8 num_vports;
- u8 rmt_start;
- u8 num_ctxt;
};
struct hfi1_vnic_vport_info;
@@ -1167,8 +1156,8 @@ struct hfi1_devdata {
u64 z_send_schedule;
u64 __percpu *send_schedule;
- /* number of reserved contexts for VNIC usage */
- u16 num_vnic_contexts;
+ /* number of reserved contexts for netdev usage */
+ u16 num_netdev_contexts;
/* number of receive contexts in use by the driver */
u32 num_rcv_contexts;
/* number of pio send contexts in use by the driver */
@@ -1417,12 +1406,12 @@ struct hfi1_devdata {
struct hfi1_vnic_data vnic;
/* Lock to protect IRQ SRC register access */
spinlock_t irq_src_lock;
-};
+ int vnic_num_vports;
+ struct net_device *dummy_netdev;
-static inline bool hfi1_vnic_is_rsm_full(struct hfi1_devdata *dd, int spare)
-{
- return (dd->vnic.rmt_start + spare) > NUM_MAP_ENTRIES;
-}
+ /* Keeps track of IPoIB RSM rule users */
+ atomic_t ipoib_rsm_usr_num;
+};
/* 8051 firmware version helper */
#define dc8051_ver(a, b, c) ((a) << 16 | (b) << 8 | (c))
@@ -1500,6 +1489,8 @@ struct hfi1_ctxtdata *hfi1_rcd_get_by_index(struct hfi1_devdata *dd, u16 ctxt);
int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread);
int handle_receive_interrupt_nodma_rtail(struct hfi1_ctxtdata *rcd, int thread);
int handle_receive_interrupt_dma_rtail(struct hfi1_ctxtdata *rcd, int thread);
+int handle_receive_interrupt_napi_fp(struct hfi1_ctxtdata *rcd, int budget);
+int handle_receive_interrupt_napi_sp(struct hfi1_ctxtdata *rcd, int budget);
void set_all_slowpath(struct hfi1_devdata *dd);
extern const struct pci_device_id hfi1_pci_tbl[];
@@ -2250,7 +2241,6 @@ extern int num_user_contexts;
extern unsigned long n_krcvqs;
extern uint krcvqs[];
extern int krcvqsset;
-extern uint kdeth_qp;
extern uint loopback;
extern uint quick_linkup;
extern uint rcv_intr_timeout;
diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c
index 3759d9233a1c..5eed4360695f 100644
--- a/drivers/infiniband/hw/hfi1/init.c
+++ b/drivers/infiniband/hw/hfi1/init.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -69,6 +69,7 @@
#include "affinity.h"
#include "vnic.h"
#include "exp_rcv.h"
+#include "netdev.h"
#undef pr_fmt
#define pr_fmt(fmt) DRIVER_NAME ": " fmt
@@ -374,6 +375,7 @@ int hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, int numa,
rcd->numa_id = numa;
rcd->rcv_array_groups = dd->rcv_entries.ngroups;
rcd->rhf_rcv_function_map = normal_rhf_rcv_functions;
+ rcd->msix_intr = CCE_NUM_MSIX_VECTORS;
mutex_init(&rcd->exp_mutex);
spin_lock_init(&rcd->exp_lock);
@@ -1316,6 +1318,7 @@ static struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev,
goto bail;
}
+ atomic_set(&dd->ipoib_rsm_usr_num, 0);
return dd;
bail:
@@ -1663,9 +1666,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
/* do the generic initialization */
initfail = hfi1_init(dd, 0);
- /* setup vnic */
- hfi1_vnic_setup(dd);
-
ret = hfi1_register_ib_device(dd);
/*
@@ -1704,7 +1704,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
hfi1_device_remove(dd);
if (!ret)
hfi1_unregister_ib_device(dd);
- hfi1_vnic_cleanup(dd);
postinit_cleanup(dd);
if (initfail)
ret = initfail;
@@ -1749,8 +1748,8 @@ static void remove_one(struct pci_dev *pdev)
/* unregister from IB core */
hfi1_unregister_ib_device(dd);
- /* cleanup vnic */
- hfi1_vnic_cleanup(dd);
+ /* free netdev data */
+ hfi1_netdev_free(dd);
/*
* Disable the IB link, disable interrupts on the device,
diff --git a/drivers/infiniband/hw/hfi1/ipoib.h b/drivers/infiniband/hw/hfi1/ipoib.h
new file mode 100644
index 000000000000..185c9b02c974
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/ipoib.h
@@ -0,0 +1,171 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) */
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+/*
+ * This file contains HFI1 support for IPOIB functionality
+ */
+
+#ifndef HFI1_IPOIB_H
+#define HFI1_IPOIB_H
+
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/atomic.h>
+#include <linux/netdevice.h>
+#include <linux/slab.h>
+#include <linux/skbuff.h>
+#include <linux/list.h>
+#include <linux/if_infiniband.h>
+
+#include "hfi.h"
+#include "iowait.h"
+#include "netdev.h"
+
+#include <rdma/ib_verbs.h>
+
+#define HFI1_IPOIB_ENTROPY_SHIFT 24
+
+#define HFI1_IPOIB_TXREQ_NAME_LEN 32
+
+#define HFI1_IPOIB_PSEUDO_LEN 20
+#define HFI1_IPOIB_ENCAP_LEN 4
+
+struct hfi1_ipoib_dev_priv;
+
+union hfi1_ipoib_flow {
+ u16 as_int;
+ struct {
+ u8 tx_queue;
+ u8 sc5;
+ } __attribute__((__packed__));
+};
+
+/**
+ * struct hfi1_ipoib_circ_buf - List of items to be processed
+ * @items: ring of items
+ * @head: ring head
+ * @tail: ring tail
+ * @max_items: max items + 1 that the ring can contain
+ * @producer_lock: producer sync lock
+ * @consumer_lock: consumer sync lock
+ */
+struct hfi1_ipoib_circ_buf {
+ void **items;
+ unsigned long head;
+ unsigned long tail;
+ unsigned long max_items;
+ spinlock_t producer_lock; /* head sync lock */
+ spinlock_t consumer_lock; /* tail sync lock */
+};
+
+/**
+ * struct hfi1_ipoib_txq - IPOIB per Tx queue information
+ * @priv: private pointer
+ * @sde: sdma engine
+ * @tx_list: tx request list
+ * @sent_txreqs: count of txreqs posted to sdma
+ * @flow: tracks when list needs to be flushed for a flow change
+ * @q_idx: ipoib Tx queue index
+ * @pkts_sent: indicator packets have been sent from this queue
+ * @wait: iowait structure
+ * @complete_txreqs: count of txreqs completed by sdma
+ * @napi: pointer to tx napi interface
+ * @tx_ring: ring of ipoib txreqs to be reaped by napi callback
+ */
+struct hfi1_ipoib_txq {
+ struct hfi1_ipoib_dev_priv *priv;
+ struct sdma_engine *sde;
+ struct list_head tx_list;
+ u64 sent_txreqs;
+ union hfi1_ipoib_flow flow;
+ u8 q_idx;
+ bool pkts_sent;
+ struct iowait wait;
+
+ atomic64_t ____cacheline_aligned_in_smp complete_txreqs;
+ struct napi_struct *napi;
+ struct hfi1_ipoib_circ_buf tx_ring;
+};
+
+struct hfi1_ipoib_dev_priv {
+ struct hfi1_devdata *dd;
+ struct net_device *netdev;
+ struct ib_device *device;
+ struct hfi1_ipoib_txq *txqs;
+ struct kmem_cache *txreq_cache;
+ struct napi_struct *tx_napis;
+ u16 pkey;
+ u16 pkey_index;
+ u32 qkey;
+ u8 port_num;
+
+ const struct net_device_ops *netdev_ops;
+ struct rvt_qp *qp;
+ struct pcpu_sw_netstats __percpu *netstats;
+};
+
+/* hfi1 ipoib rdma netdev's private data structure */
+struct hfi1_ipoib_rdma_netdev {
+ struct rdma_netdev rn; /* keep this first */
+ /* followed by device private data */
+ struct hfi1_ipoib_dev_priv dev_priv;
+};
+
+static inline struct hfi1_ipoib_dev_priv *
+hfi1_ipoib_priv(const struct net_device *dev)
+{
+ return &((struct hfi1_ipoib_rdma_netdev *)netdev_priv(dev))->dev_priv;
+}
+
+static inline void
+hfi1_ipoib_update_rx_netstats(struct hfi1_ipoib_dev_priv *priv,
+ u64 packets,
+ u64 bytes)
+{
+ struct pcpu_sw_netstats *netstats = this_cpu_ptr(priv->netstats);
+
+ u64_stats_update_begin(&netstats->syncp);
+ netstats->rx_packets += packets;
+ netstats->rx_bytes += bytes;
+ u64_stats_update_end(&netstats->syncp);
+}
+
+static inline void
+hfi1_ipoib_update_tx_netstats(struct hfi1_ipoib_dev_priv *priv,
+ u64 packets,
+ u64 bytes)
+{
+ struct pcpu_sw_netstats *netstats = this_cpu_ptr(priv->netstats);
+
+ u64_stats_update_begin(&netstats->syncp);
+ netstats->tx_packets += packets;
+ netstats->tx_bytes += bytes;
+ u64_stats_update_end(&netstats->syncp);
+}
+
+int hfi1_ipoib_send_dma(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ib_ah *address,
+ u32 dqpn);
+
+int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv);
+void hfi1_ipoib_txreq_deinit(struct hfi1_ipoib_dev_priv *priv);
+
+int hfi1_ipoib_rxq_init(struct net_device *dev);
+void hfi1_ipoib_rxq_deinit(struct net_device *dev);
+
+void hfi1_ipoib_napi_tx_enable(struct net_device *dev);
+void hfi1_ipoib_napi_tx_disable(struct net_device *dev);
+
+struct sk_buff *hfi1_ipoib_prepare_skb(struct hfi1_netdev_rxq *rxq,
+ int size, void *data);
+
+int hfi1_ipoib_rn_get_params(struct ib_device *device,
+ u8 port_num,
+ enum rdma_netdev_t type,
+ struct rdma_netdev_alloc_params *params);
+
+#endif /* _IPOIB_H */
diff --git a/drivers/infiniband/hw/hfi1/ipoib_main.c b/drivers/infiniband/hw/hfi1/ipoib_main.c
new file mode 100644
index 000000000000..014351ebbefa
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/ipoib_main.c
@@ -0,0 +1,309 @@
+// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+/*
+ * This file contains HFI1 support for ipoib functionality
+ */
+
+#include "ipoib.h"
+#include "hfi.h"
+
+static u32 qpn_from_mac(u8 *mac_arr)
+{
+ return (u32)mac_arr[1] << 16 | mac_arr[2] << 8 | mac_arr[3];
+}
+
+static int hfi1_ipoib_dev_init(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ int ret;
+
+ priv->netstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
+
+ ret = priv->netdev_ops->ndo_init(dev);
+ if (ret)
+ return ret;
+
+ ret = hfi1_netdev_add_data(priv->dd,
+ qpn_from_mac(priv->netdev->dev_addr),
+ dev);
+ if (ret < 0) {
+ priv->netdev_ops->ndo_uninit(dev);
+ return ret;
+ }
+
+ return 0;
+}
+
+static void hfi1_ipoib_dev_uninit(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+
+ hfi1_netdev_remove_data(priv->dd, qpn_from_mac(priv->netdev->dev_addr));
+
+ priv->netdev_ops->ndo_uninit(dev);
+}
+
+static int hfi1_ipoib_dev_open(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ int ret;
+
+ ret = priv->netdev_ops->ndo_open(dev);
+ if (!ret) {
+ struct hfi1_ibport *ibp = to_iport(priv->device,
+ priv->port_num);
+ struct rvt_qp *qp;
+ u32 qpn = qpn_from_mac(priv->netdev->dev_addr);
+
+ rcu_read_lock();
+ qp = rvt_lookup_qpn(ib_to_rvt(priv->device), &ibp->rvp, qpn);
+ if (!qp) {
+ rcu_read_unlock();
+ priv->netdev_ops->ndo_stop(dev);
+ return -EINVAL;
+ }
+ rvt_get_qp(qp);
+ priv->qp = qp;
+ rcu_read_unlock();
+
+ hfi1_netdev_enable_queues(priv->dd);
+ hfi1_ipoib_napi_tx_enable(dev);
+ }
+
+ return ret;
+}
+
+static int hfi1_ipoib_dev_stop(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+
+ if (!priv->qp)
+ return 0;
+
+ hfi1_ipoib_napi_tx_disable(dev);
+ hfi1_netdev_disable_queues(priv->dd);
+
+ rvt_put_qp(priv->qp);
+ priv->qp = NULL;
+
+ return priv->netdev_ops->ndo_stop(dev);
+}
+
+static void hfi1_ipoib_dev_get_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *storage)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ u64 rx_packets = 0ull;
+ u64 rx_bytes = 0ull;
+ u64 tx_packets = 0ull;
+ u64 tx_bytes = 0ull;
+ int i;
+
+ netdev_stats_to_stats64(storage, &dev->stats);
+
+ for_each_possible_cpu(i) {
+ const struct pcpu_sw_netstats *stats;
+ unsigned int start;
+ u64 trx_packets;
+ u64 trx_bytes;
+ u64 ttx_packets;
+ u64 ttx_bytes;
+
+ stats = per_cpu_ptr(priv->netstats, i);
+ do {
+ start = u64_stats_fetch_begin_irq(&stats->syncp);
+ trx_packets = stats->rx_packets;
+ trx_bytes = stats->rx_bytes;
+ ttx_packets = stats->tx_packets;
+ ttx_bytes = stats->tx_bytes;
+ } while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+ rx_packets += trx_packets;
+ rx_bytes += trx_bytes;
+ tx_packets += ttx_packets;
+ tx_bytes += ttx_bytes;
+ }
+
+ storage->rx_packets += rx_packets;
+ storage->rx_bytes += rx_bytes;
+ storage->tx_packets += tx_packets;
+ storage->tx_bytes += tx_bytes;
+}
+
+static const struct net_device_ops hfi1_ipoib_netdev_ops = {
+ .ndo_init = hfi1_ipoib_dev_init,
+ .ndo_uninit = hfi1_ipoib_dev_uninit,
+ .ndo_open = hfi1_ipoib_dev_open,
+ .ndo_stop = hfi1_ipoib_dev_stop,
+ .ndo_get_stats64 = hfi1_ipoib_dev_get_stats64,
+};
+
+static int hfi1_ipoib_send(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ib_ah *address,
+ u32 dqpn)
+{
+ return hfi1_ipoib_send_dma(dev, skb, address, dqpn);
+}
+
+static int hfi1_ipoib_mcast_attach(struct net_device *dev,
+ struct ib_device *device,
+ union ib_gid *mgid,
+ u16 mlid,
+ int set_qkey,
+ u32 qkey)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ u32 qpn = (u32)qpn_from_mac(priv->netdev->dev_addr);
+ struct hfi1_ibport *ibp = to_iport(priv->device, priv->port_num);
+ struct rvt_qp *qp;
+ int ret = -EINVAL;
+
+ rcu_read_lock();
+
+ qp = rvt_lookup_qpn(ib_to_rvt(priv->device), &ibp->rvp, qpn);
+ if (qp) {
+ rvt_get_qp(qp);
+ rcu_read_unlock();
+ if (set_qkey)
+ priv->qkey = qkey;
+
+ /* attach QP to multicast group */
+ ret = ib_attach_mcast(&qp->ibqp, mgid, mlid);
+ rvt_put_qp(qp);
+ } else {
+ rcu_read_unlock();
+ }
+
+ return ret;
+}
+
+static int hfi1_ipoib_mcast_detach(struct net_device *dev,
+ struct ib_device *device,
+ union ib_gid *mgid,
+ u16 mlid)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ u32 qpn = (u32)qpn_from_mac(priv->netdev->dev_addr);
+ struct hfi1_ibport *ibp = to_iport(priv->device, priv->port_num);
+ struct rvt_qp *qp;
+ int ret = -EINVAL;
+
+ rcu_read_lock();
+
+ qp = rvt_lookup_qpn(ib_to_rvt(priv->device), &ibp->rvp, qpn);
+ if (qp) {
+ rvt_get_qp(qp);
+ rcu_read_unlock();
+ ret = ib_detach_mcast(&qp->ibqp, mgid, mlid);
+ rvt_put_qp(qp);
+ } else {
+ rcu_read_unlock();
+ }
+ return ret;
+}
+
+static void hfi1_ipoib_netdev_dtor(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+
+ hfi1_ipoib_txreq_deinit(priv);
+ hfi1_ipoib_rxq_deinit(priv->netdev);
+
+ free_percpu(priv->netstats);
+}
+
+static void hfi1_ipoib_free_rdma_netdev(struct net_device *dev)
+{
+ hfi1_ipoib_netdev_dtor(dev);
+ free_netdev(dev);
+}
+
+static void hfi1_ipoib_set_id(struct net_device *dev, int id)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+
+ priv->pkey_index = (u16)id;
+ ib_query_pkey(priv->device,
+ priv->port_num,
+ priv->pkey_index,
+ &priv->pkey);
+}
+
+static int hfi1_ipoib_setup_rn(struct ib_device *device,
+ u8 port_num,
+ struct net_device *netdev,
+ void *param)
+{
+ struct hfi1_devdata *dd = dd_from_ibdev(device);
+ struct rdma_netdev *rn = netdev_priv(netdev);
+ struct hfi1_ipoib_dev_priv *priv;
+ int rc;
+
+ rn->send = hfi1_ipoib_send;
+ rn->attach_mcast = hfi1_ipoib_mcast_attach;
+ rn->detach_mcast = hfi1_ipoib_mcast_detach;
+ rn->set_id = hfi1_ipoib_set_id;
+ rn->hca = device;
+ rn->port_num = port_num;
+ rn->mtu = netdev->mtu;
+
+ priv = hfi1_ipoib_priv(netdev);
+ priv->dd = dd;
+ priv->netdev = netdev;
+ priv->device = device;
+ priv->port_num = port_num;
+ priv->netdev_ops = netdev->netdev_ops;
+
+ netdev->netdev_ops = &hfi1_ipoib_netdev_ops;
+
+ ib_query_pkey(device, port_num, priv->pkey_index, &priv->pkey);
+
+ rc = hfi1_ipoib_txreq_init(priv);
+ if (rc) {
+ dd_dev_err(dd, "IPoIB netdev TX init - failed(%d)\n", rc);
+ hfi1_ipoib_free_rdma_netdev(netdev);
+ return rc;
+ }
+
+ rc = hfi1_ipoib_rxq_init(netdev);
+ if (rc) {
+ dd_dev_err(dd, "IPoIB netdev RX init - failed(%d)\n", rc);
+ hfi1_ipoib_free_rdma_netdev(netdev);
+ return rc;
+ }
+
+ netdev->priv_destructor = hfi1_ipoib_netdev_dtor;
+ netdev->needs_free_netdev = true;
+
+ return 0;
+}
+
+int hfi1_ipoib_rn_get_params(struct ib_device *device,
+ u8 port_num,
+ enum rdma_netdev_t type,
+ struct rdma_netdev_alloc_params *params)
+{
+ struct hfi1_devdata *dd = dd_from_ibdev(device);
+
+ if (type != RDMA_NETDEV_IPOIB)
+ return -EOPNOTSUPP;
+
+ if (!HFI1_CAP_IS_KSET(AIP) || !dd->num_netdev_contexts)
+ return -EOPNOTSUPP;
+
+ if (!port_num || port_num > dd->num_pports)
+ return -EINVAL;
+
+ params->sizeof_priv = sizeof(struct hfi1_ipoib_rdma_netdev);
+ params->txqs = dd->num_sdma;
+ params->rxqs = dd->num_netdev_contexts;
+ params->param = NULL;
+ params->initialize_rdma_netdev = hfi1_ipoib_setup_rn;
+
+ return 0;
+}
diff --git a/drivers/infiniband/hw/hfi1/ipoib_rx.c b/drivers/infiniband/hw/hfi1/ipoib_rx.c
new file mode 100644
index 000000000000..3afa7545242c
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/ipoib_rx.c
@@ -0,0 +1,95 @@
+// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+#include "netdev.h"
+#include "ipoib.h"
+
+#define HFI1_IPOIB_SKB_PAD ((NET_SKB_PAD) + (NET_IP_ALIGN))
+
+static void copy_ipoib_buf(struct sk_buff *skb, void *data, int size)
+{
+ void *dst_data;
+
+ skb_checksum_none_assert(skb);
+ skb->protocol = *((__be16 *)data);
+
+ dst_data = skb_put(skb, size);
+ memcpy(dst_data, data, size);
+ skb->mac_header = HFI1_IPOIB_PSEUDO_LEN;
+ skb_pull(skb, HFI1_IPOIB_ENCAP_LEN);
+}
+
+static struct sk_buff *prepare_frag_skb(struct napi_struct *napi, int size)
+{
+ struct sk_buff *skb;
+ int skb_size = SKB_DATA_ALIGN(size + HFI1_IPOIB_SKB_PAD);
+ void *frag;
+
+ skb_size += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+ skb_size = SKB_DATA_ALIGN(skb_size);
+ frag = napi_alloc_frag(skb_size);
+
+ if (unlikely(!frag))
+ return napi_alloc_skb(napi, size);
+
+ skb = build_skb(frag, skb_size);
+
+ if (unlikely(!skb)) {
+ skb_free_frag(frag);
+ return NULL;
+ }
+
+ skb_reserve(skb, HFI1_IPOIB_SKB_PAD);
+ return skb;
+}
+
+struct sk_buff *hfi1_ipoib_prepare_skb(struct hfi1_netdev_rxq *rxq,
+ int size, void *data)
+{
+ struct napi_struct *napi = &rxq->napi;
+ int skb_size = size + HFI1_IPOIB_ENCAP_LEN;
+ struct sk_buff *skb;
+
+ /*
+ * For smaller(4k + skb overhead) allocations we will go using
+ * napi cache. Otherwise we will try to use napi frag cache.
+ */
+ if (size <= SKB_WITH_OVERHEAD(PAGE_SIZE))
+ skb = napi_alloc_skb(napi, skb_size);
+ else
+ skb = prepare_frag_skb(napi, skb_size);
+
+ if (unlikely(!skb))
+ return NULL;
+
+ copy_ipoib_buf(skb, data, size);
+
+ return skb;
+}
+
+int hfi1_ipoib_rxq_init(struct net_device *netdev)
+{
+ struct hfi1_ipoib_dev_priv *ipoib_priv = hfi1_ipoib_priv(netdev);
+ struct hfi1_devdata *dd = ipoib_priv->dd;
+ int ret;
+
+ ret = hfi1_netdev_rx_init(dd);
+ if (ret)
+ return ret;
+
+ hfi1_init_aip_rsm(dd);
+
+ return ret;
+}
+
+void hfi1_ipoib_rxq_deinit(struct net_device *netdev)
+{
+ struct hfi1_ipoib_dev_priv *ipoib_priv = hfi1_ipoib_priv(netdev);
+ struct hfi1_devdata *dd = ipoib_priv->dd;
+
+ hfi1_deinit_aip_rsm(dd);
+ hfi1_netdev_rx_destroy(dd);
+}
diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
new file mode 100644
index 000000000000..883cb9d48022
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
@@ -0,0 +1,828 @@
+// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+/*
+ * This file contains HFI1 support for IPOIB SDMA functionality
+ */
+
+#include <linux/log2.h>
+#include <linux/circ_buf.h>
+
+#include "sdma.h"
+#include "verbs.h"
+#include "trace_ibhdrs.h"
+#include "ipoib.h"
+
+/* Add a convenience helper */
+#define CIRC_ADD(val, add, size) (((val) + (add)) & ((size) - 1))
+#define CIRC_NEXT(val, size) CIRC_ADD(val, 1, size)
+#define CIRC_PREV(val, size) CIRC_ADD(val, -1, size)
+
+/**
+ * struct ipoib_txreq - IPOIB transmit descriptor
+ * @txreq: sdma transmit request
+ * @sdma_hdr: 9b ib headers
+ * @sdma_status: status returned by sdma engine
+ * @priv: ipoib netdev private data
+ * @txq: txq on which skb was output
+ * @skb: skb to send
+ */
+struct ipoib_txreq {
+ struct sdma_txreq txreq;
+ struct hfi1_sdma_header sdma_hdr;
+ int sdma_status;
+ struct hfi1_ipoib_dev_priv *priv;
+ struct hfi1_ipoib_txq *txq;
+ struct sk_buff *skb;
+};
+
+struct ipoib_txparms {
+ struct hfi1_devdata *dd;
+ struct rdma_ah_attr *ah_attr;
+ struct hfi1_ibport *ibp;
+ struct hfi1_ipoib_txq *txq;
+ union hfi1_ipoib_flow flow;
+ u32 dqpn;
+ u8 hdr_dwords;
+ u8 entropy;
+};
+
+static u64 hfi1_ipoib_txreqs(const u64 sent, const u64 completed)
+{
+ return sent - completed;
+}
+
+static void hfi1_ipoib_check_queue_depth(struct hfi1_ipoib_txq *txq)
+{
+ if (unlikely(hfi1_ipoib_txreqs(++txq->sent_txreqs,
+ atomic64_read(&txq->complete_txreqs)) >=
+ min_t(unsigned int, txq->priv->netdev->tx_queue_len,
+ txq->tx_ring.max_items - 1)))
+ netif_stop_subqueue(txq->priv->netdev, txq->q_idx);
+}
+
+static void hfi1_ipoib_check_queue_stopped(struct hfi1_ipoib_txq *txq)
+{
+ struct net_device *dev = txq->priv->netdev;
+
+ /* If the queue is already running just return */
+ if (likely(!__netif_subqueue_stopped(dev, txq->q_idx)))
+ return;
+
+ /* If shutting down just return as queue state is irrelevant */
+ if (unlikely(dev->reg_state != NETREG_REGISTERED))
+ return;
+
+ /*
+ * When the queue has been drained to less than half full it will be
+ * restarted.
+ * The size of the txreq ring is fixed at initialization.
+ * The tx queue len can be adjusted upward while the interface is
+ * running.
+ * The tx queue len can be large enough to overflow the txreq_ring.
+ * Use the minimum of the current tx_queue_len or the rings max txreqs
+ * to protect against ring overflow.
+ */
+ if (hfi1_ipoib_txreqs(txq->sent_txreqs,
+ atomic64_read(&txq->complete_txreqs))
+ < min_t(unsigned int, dev->tx_queue_len,
+ txq->tx_ring.max_items) >> 1)
+ netif_wake_subqueue(dev, txq->q_idx);
+}
+
+static void hfi1_ipoib_free_tx(struct ipoib_txreq *tx, int budget)
+{
+ struct hfi1_ipoib_dev_priv *priv = tx->priv;
+
+ if (likely(!tx->sdma_status)) {
+ hfi1_ipoib_update_tx_netstats(priv, 1, tx->skb->len);
+ } else {
+ ++priv->netdev->stats.tx_errors;
+ dd_dev_warn(priv->dd,
+ "%s: Status = 0x%x pbc 0x%llx txq = %d sde = %d\n",
+ __func__, tx->sdma_status,
+ le64_to_cpu(tx->sdma_hdr.pbc), tx->txq->q_idx,
+ tx->txq->sde->this_idx);
+ }
+
+ napi_consume_skb(tx->skb, budget);
+ sdma_txclean(priv->dd, &tx->txreq);
+ kmem_cache_free(priv->txreq_cache, tx);
+}
+
+static int hfi1_ipoib_drain_tx_ring(struct hfi1_ipoib_txq *txq, int budget)
+{
+ struct hfi1_ipoib_circ_buf *tx_ring = &txq->tx_ring;
+ unsigned long head;
+ unsigned long tail;
+ unsigned int max_tx;
+ int work_done;
+ int tx_count;
+
+ spin_lock_bh(&tx_ring->consumer_lock);
+
+ /* Read index before reading contents at that index. */
+ head = smp_load_acquire(&tx_ring->head);
+ tail = tx_ring->tail;
+ max_tx = tx_ring->max_items;
+
+ work_done = min_t(int, CIRC_CNT(head, tail, max_tx), budget);
+
+ for (tx_count = work_done; tx_count; tx_count--) {
+ hfi1_ipoib_free_tx(tx_ring->items[tail], budget);
+ tail = CIRC_NEXT(tail, max_tx);
+ }
+
+ atomic64_add(work_done, &txq->complete_txreqs);
+
+ /* Finished freeing tx items so store the tail value. */
+ smp_store_release(&tx_ring->tail, tail);
+
+ spin_unlock_bh(&tx_ring->consumer_lock);
+
+ hfi1_ipoib_check_queue_stopped(txq);
+
+ return work_done;
+}
+
+static int hfi1_ipoib_process_tx_ring(struct napi_struct *napi, int budget)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(napi->dev);
+ struct hfi1_ipoib_txq *txq = &priv->txqs[napi - priv->tx_napis];
+
+ int work_done = hfi1_ipoib_drain_tx_ring(txq, budget);
+
+ if (work_done < budget)
+ napi_complete_done(napi, work_done);
+
+ return work_done;
+}
+
+static void hfi1_ipoib_add_tx(struct ipoib_txreq *tx)
+{
+ struct hfi1_ipoib_circ_buf *tx_ring = &tx->txq->tx_ring;
+ unsigned long head;
+ unsigned long tail;
+ size_t max_tx;
+
+ spin_lock(&tx_ring->producer_lock);
+
+ head = tx_ring->head;
+ tail = READ_ONCE(tx_ring->tail);
+ max_tx = tx_ring->max_items;
+
+ if (likely(CIRC_SPACE(head, tail, max_tx))) {
+ tx_ring->items[head] = tx;
+
+ /* Finish storing txreq before incrementing head. */
+ smp_store_release(&tx_ring->head, CIRC_ADD(head, 1, max_tx));
+ napi_schedule(tx->txq->napi);
+ } else {
+ struct hfi1_ipoib_txq *txq = tx->txq;
+ struct hfi1_ipoib_dev_priv *priv = tx->priv;
+
+ /* Ring was full */
+ hfi1_ipoib_free_tx(tx, 0);
+ atomic64_inc(&txq->complete_txreqs);
+ dd_dev_dbg(priv->dd, "txq %d full.\n", txq->q_idx);
+ }
+
+ spin_unlock(&tx_ring->producer_lock);
+}
+
+static void hfi1_ipoib_sdma_complete(struct sdma_txreq *txreq, int status)
+{
+ struct ipoib_txreq *tx = container_of(txreq, struct ipoib_txreq, txreq);
+
+ tx->sdma_status = status;
+
+ hfi1_ipoib_add_tx(tx);
+}
+
+static int hfi1_ipoib_build_ulp_payload(struct ipoib_txreq *tx,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_devdata *dd = txp->dd;
+ struct sdma_txreq *txreq = &tx->txreq;
+ struct sk_buff *skb = tx->skb;
+ int ret = 0;
+ int i;
+
+ if (skb_headlen(skb)) {
+ ret = sdma_txadd_kvaddr(dd, txreq, skb->data, skb_headlen(skb));
+ if (unlikely(ret))
+ return ret;
+ }
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+ const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+ ret = sdma_txadd_page(dd,
+ txreq,
+ skb_frag_page(frag),
+ frag->bv_offset,
+ skb_frag_size(frag));
+ if (unlikely(ret))
+ break;
+ }
+
+ return ret;
+}
+
+static int hfi1_ipoib_build_tx_desc(struct ipoib_txreq *tx,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_devdata *dd = txp->dd;
+ struct sdma_txreq *txreq = &tx->txreq;
+ struct hfi1_sdma_header *sdma_hdr = &tx->sdma_hdr;
+ u16 pkt_bytes =
+ sizeof(sdma_hdr->pbc) + (txp->hdr_dwords << 2) + tx->skb->len;
+ int ret;
+
+ ret = sdma_txinit(txreq, 0, pkt_bytes, hfi1_ipoib_sdma_complete);
+ if (unlikely(ret))
+ return ret;
+
+ /* add pbc + headers */
+ ret = sdma_txadd_kvaddr(dd,
+ txreq,
+ sdma_hdr,
+ sizeof(sdma_hdr->pbc) + (txp->hdr_dwords << 2));
+ if (unlikely(ret))
+ return ret;
+
+ /* add the ulp payload */
+ return hfi1_ipoib_build_ulp_payload(tx, txp);
+}
+
+static void hfi1_ipoib_build_ib_tx_headers(struct ipoib_txreq *tx,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_ipoib_dev_priv *priv = tx->priv;
+ struct hfi1_sdma_header *sdma_hdr = &tx->sdma_hdr;
+ struct sk_buff *skb = tx->skb;
+ struct hfi1_pportdata *ppd = ppd_from_ibp(txp->ibp);
+ struct rdma_ah_attr *ah_attr = txp->ah_attr;
+ struct ib_other_headers *ohdr;
+ struct ib_grh *grh;
+ u16 dwords;
+ u16 slid;
+ u16 dlid;
+ u16 lrh0;
+ u32 bth0;
+ u32 sqpn = (u32)(priv->netdev->dev_addr[1] << 16 |
+ priv->netdev->dev_addr[2] << 8 |
+ priv->netdev->dev_addr[3]);
+ u16 payload_dwords;
+ u8 pad_cnt;
+
+ pad_cnt = -skb->len & 3;
+
+ /* Includes ICRC */
+ payload_dwords = ((skb->len + pad_cnt) >> 2) + SIZE_OF_CRC;
+
+ /* header size in dwords LRH+BTH+DETH = (8+12+8)/4. */
+ txp->hdr_dwords = 7;
+
+ if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) {
+ grh = &sdma_hdr->hdr.ibh.u.l.grh;
+ txp->hdr_dwords +=
+ hfi1_make_grh(txp->ibp,
+ grh,
+ rdma_ah_read_grh(ah_attr),
+ txp->hdr_dwords - LRH_9B_DWORDS,
+ payload_dwords);
+ lrh0 = HFI1_LRH_GRH;
+ ohdr = &sdma_hdr->hdr.ibh.u.l.oth;
+ } else {
+ lrh0 = HFI1_LRH_BTH;
+ ohdr = &sdma_hdr->hdr.ibh.u.oth;
+ }
+
+ lrh0 |= (rdma_ah_get_sl(ah_attr) & 0xf) << 4;
+ lrh0 |= (txp->flow.sc5 & 0xf) << 12;
+
+ dlid = opa_get_lid(rdma_ah_get_dlid(ah_attr), 9B);
+ if (dlid == be16_to_cpu(IB_LID_PERMISSIVE)) {
+ slid = be16_to_cpu(IB_LID_PERMISSIVE);
+ } else {
+ u16 lid = (u16)ppd->lid;
+
+ if (lid) {
+ lid |= rdma_ah_get_path_bits(ah_attr) &
+ ((1 << ppd->lmc) - 1);
+ slid = lid;
+ } else {
+ slid = be16_to_cpu(IB_LID_PERMISSIVE);
+ }
+ }
+
+ /* Includes ICRC */
+ dwords = txp->hdr_dwords + payload_dwords;
+
+ /* Build the lrh */
+ sdma_hdr->hdr.hdr_type = HFI1_PKT_TYPE_9B;
+ hfi1_make_ib_hdr(&sdma_hdr->hdr.ibh, lrh0, dwords, dlid, slid);
+
+ /* Build the bth */
+ bth0 = (IB_OPCODE_UD_SEND_ONLY << 24) | (pad_cnt << 20) | priv->pkey;
+
+ ohdr->bth[0] = cpu_to_be32(bth0);
+ ohdr->bth[1] = cpu_to_be32(txp->dqpn);
+ ohdr->bth[2] = cpu_to_be32(mask_psn((u32)txp->txq->sent_txreqs));
+
+ /* Build the deth */
+ ohdr->u.ud.deth[0] = cpu_to_be32(priv->qkey);
+ ohdr->u.ud.deth[1] = cpu_to_be32((txp->entropy <<
+ HFI1_IPOIB_ENTROPY_SHIFT) | sqpn);
+
+ /* Construct the pbc. */
+ sdma_hdr->pbc =
+ cpu_to_le64(create_pbc(ppd,
+ ib_is_sc5(txp->flow.sc5) <<
+ PBC_DC_INFO_SHIFT,
+ 0,
+ sc_to_vlt(priv->dd, txp->flow.sc5),
+ dwords - SIZE_OF_CRC +
+ (sizeof(sdma_hdr->pbc) >> 2)));
+}
+
+static struct ipoib_txreq *hfi1_ipoib_send_dma_common(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ struct ipoib_txreq *tx;
+ int ret;
+
+ tx = kmem_cache_alloc_node(priv->txreq_cache,
+ GFP_ATOMIC,
+ priv->dd->node);
+ if (unlikely(!tx))
+ return ERR_PTR(-ENOMEM);
+
+ /* so that we can test if the sdma decriptors are there */
+ tx->txreq.num_desc = 0;
+ tx->priv = priv;
+ tx->txq = txp->txq;
+ tx->skb = skb;
+
+ hfi1_ipoib_build_ib_tx_headers(tx, txp);
+
+ ret = hfi1_ipoib_build_tx_desc(tx, txp);
+ if (likely(!ret)) {
+ if (txp->txq->flow.as_int != txp->flow.as_int) {
+ txp->txq->flow.tx_queue = txp->flow.tx_queue;
+ txp->txq->flow.sc5 = txp->flow.sc5;
+ txp->txq->sde =
+ sdma_select_engine_sc(priv->dd,
+ txp->flow.tx_queue,
+ txp->flow.sc5);
+ }
+
+ return tx;
+ }
+
+ sdma_txclean(priv->dd, &tx->txreq);
+ kmem_cache_free(priv->txreq_cache, tx);
+
+ return ERR_PTR(ret);
+}
+
+static int hfi1_ipoib_submit_tx_list(struct net_device *dev,
+ struct hfi1_ipoib_txq *txq)
+{
+ int ret;
+ u16 count_out;
+
+ ret = sdma_send_txlist(txq->sde,
+ iowait_get_ib_work(&txq->wait),
+ &txq->tx_list,
+ &count_out);
+ if (likely(!ret) || ret == -EBUSY || ret == -ECOMM)
+ return ret;
+
+ dd_dev_warn(txq->priv->dd, "cannot send skb tx list, err %d.\n", ret);
+
+ return ret;
+}
+
+static int hfi1_ipoib_flush_tx_list(struct net_device *dev,
+ struct hfi1_ipoib_txq *txq)
+{
+ int ret = 0;
+
+ if (!list_empty(&txq->tx_list)) {
+ /* Flush the current list */
+ ret = hfi1_ipoib_submit_tx_list(dev, txq);
+
+ if (unlikely(ret))
+ if (ret != -EBUSY)
+ ++dev->stats.tx_carrier_errors;
+ }
+
+ return ret;
+}
+
+static int hfi1_ipoib_submit_tx(struct hfi1_ipoib_txq *txq,
+ struct ipoib_txreq *tx)
+{
+ int ret;
+
+ ret = sdma_send_txreq(txq->sde,
+ iowait_get_ib_work(&txq->wait),
+ &tx->txreq,
+ txq->pkts_sent);
+ if (likely(!ret)) {
+ txq->pkts_sent = true;
+ iowait_starve_clear(txq->pkts_sent, &txq->wait);
+ }
+
+ return ret;
+}
+
+static int hfi1_ipoib_send_dma_single(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ struct hfi1_ipoib_txq *txq = txp->txq;
+ struct ipoib_txreq *tx;
+ int ret;
+
+ tx = hfi1_ipoib_send_dma_common(dev, skb, txp);
+ if (IS_ERR(tx)) {
+ int ret = PTR_ERR(tx);
+
+ dev_kfree_skb_any(skb);
+
+ if (ret == -ENOMEM)
+ ++dev->stats.tx_errors;
+ else
+ ++dev->stats.tx_carrier_errors;
+
+ return NETDEV_TX_OK;
+ }
+
+ ret = hfi1_ipoib_submit_tx(txq, tx);
+ if (likely(!ret)) {
+ trace_sdma_output_ibhdr(tx->priv->dd,
+ &tx->sdma_hdr.hdr,
+ ib_is_sc5(txp->flow.sc5));
+ hfi1_ipoib_check_queue_depth(txq);
+ return NETDEV_TX_OK;
+ }
+
+ txq->pkts_sent = false;
+
+ if (ret == -EBUSY) {
+ list_add_tail(&tx->txreq.list, &txq->tx_list);
+
+ trace_sdma_output_ibhdr(tx->priv->dd,
+ &tx->sdma_hdr.hdr,
+ ib_is_sc5(txp->flow.sc5));
+ hfi1_ipoib_check_queue_depth(txq);
+ return NETDEV_TX_OK;
+ }
+
+ if (ret == -ECOMM) {
+ hfi1_ipoib_check_queue_depth(txq);
+ return NETDEV_TX_OK;
+ }
+
+ sdma_txclean(priv->dd, &tx->txreq);
+ dev_kfree_skb_any(skb);
+ kmem_cache_free(priv->txreq_cache, tx);
+ ++dev->stats.tx_carrier_errors;
+
+ return NETDEV_TX_OK;
+}
+
+static int hfi1_ipoib_send_dma_list(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ipoib_txparms *txp)
+{
+ struct hfi1_ipoib_txq *txq = txp->txq;
+ struct ipoib_txreq *tx;
+
+ /* Has the flow change ? */
+ if (txq->flow.as_int != txp->flow.as_int)
+ (void)hfi1_ipoib_flush_tx_list(dev, txq);
+
+ tx = hfi1_ipoib_send_dma_common(dev, skb, txp);
+ if (IS_ERR(tx)) {
+ int ret = PTR_ERR(tx);
+
+ dev_kfree_skb_any(skb);
+
+ if (ret == -ENOMEM)
+ ++dev->stats.tx_errors;
+ else
+ ++dev->stats.tx_carrier_errors;
+
+ return NETDEV_TX_OK;
+ }
+
+ list_add_tail(&tx->txreq.list, &txq->tx_list);
+
+ hfi1_ipoib_check_queue_depth(txq);
+
+ trace_sdma_output_ibhdr(tx->priv->dd,
+ &tx->sdma_hdr.hdr,
+ ib_is_sc5(txp->flow.sc5));
+
+ if (!netdev_xmit_more())
+ (void)hfi1_ipoib_flush_tx_list(dev, txq);
+
+ return NETDEV_TX_OK;
+}
+
+static u8 hfi1_ipoib_calc_entropy(struct sk_buff *skb)
+{
+ if (skb_transport_header_was_set(skb)) {
+ u8 *hdr = (u8 *)skb_transport_header(skb);
+
+ return (hdr[0] ^ hdr[1] ^ hdr[2] ^ hdr[3]);
+ }
+
+ return (u8)skb_get_queue_mapping(skb);
+}
+
+int hfi1_ipoib_send_dma(struct net_device *dev,
+ struct sk_buff *skb,
+ struct ib_ah *address,
+ u32 dqpn)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ struct ipoib_txparms txp;
+ struct rdma_netdev *rn = netdev_priv(dev);
+
+ if (unlikely(skb->len > rn->mtu + HFI1_IPOIB_ENCAP_LEN)) {
+ dd_dev_warn(priv->dd, "packet len %d (> %d) too long to send, dropping\n",
+ skb->len,
+ rn->mtu + HFI1_IPOIB_ENCAP_LEN);
+ ++dev->stats.tx_dropped;
+ ++dev->stats.tx_errors;
+ dev_kfree_skb_any(skb);
+ return NETDEV_TX_OK;
+ }
+
+ txp.dd = priv->dd;
+ txp.ah_attr = &ibah_to_rvtah(address)->attr;
+ txp.ibp = to_iport(priv->device, priv->port_num);
+ txp.txq = &priv->txqs[skb_get_queue_mapping(skb)];
+ txp.dqpn = dqpn;
+ txp.flow.sc5 = txp.ibp->sl_to_sc[rdma_ah_get_sl(txp.ah_attr)];
+ txp.flow.tx_queue = (u8)skb_get_queue_mapping(skb);
+ txp.entropy = hfi1_ipoib_calc_entropy(skb);
+
+ if (netdev_xmit_more() || !list_empty(&txp.txq->tx_list))
+ return hfi1_ipoib_send_dma_list(dev, skb, &txp);
+
+ return hfi1_ipoib_send_dma_single(dev, skb, &txp);
+}
+
+/*
+ * hfi1_ipoib_sdma_sleep - ipoib sdma sleep function
+ *
+ * This function gets called from sdma_send_txreq() when there are not enough
+ * sdma descriptors available to send the packet. It adds Tx queue's wait
+ * structure to sdma engine's dmawait list to be woken up when descriptors
+ * become available.
+ */
+static int hfi1_ipoib_sdma_sleep(struct sdma_engine *sde,
+ struct iowait_work *wait,
+ struct sdma_txreq *txreq,
+ uint seq,
+ bool pkts_sent)
+{
+ struct hfi1_ipoib_txq *txq =
+ container_of(wait->iow, struct hfi1_ipoib_txq, wait);
+
+ write_seqlock(&sde->waitlock);
+
+ if (likely(txq->priv->netdev->reg_state == NETREG_REGISTERED)) {
+ if (sdma_progress(sde, seq, txreq)) {
+ write_sequnlock(&sde->waitlock);
+ return -EAGAIN;
+ }
+
+ netif_stop_subqueue(txq->priv->netdev, txq->q_idx);
+
+ if (list_empty(&txq->wait.list))
+ iowait_queue(pkts_sent, wait->iow, &sde->dmawait);
+
+ write_sequnlock(&sde->waitlock);
+ return -EBUSY;
+ }
+
+ write_sequnlock(&sde->waitlock);
+ return -EINVAL;
+}
+
+/*
+ * hfi1_ipoib_sdma_wakeup - ipoib sdma wakeup function
+ *
+ * This function gets called when SDMA descriptors becomes available and Tx
+ * queue's wait structure was previously added to sdma engine's dmawait list.
+ */
+static void hfi1_ipoib_sdma_wakeup(struct iowait *wait, int reason)
+{
+ struct hfi1_ipoib_txq *txq =
+ container_of(wait, struct hfi1_ipoib_txq, wait);
+
+ if (likely(txq->priv->netdev->reg_state == NETREG_REGISTERED))
+ iowait_schedule(wait, system_highpri_wq, WORK_CPU_UNBOUND);
+}
+
+static void hfi1_ipoib_flush_txq(struct work_struct *work)
+{
+ struct iowait_work *ioww =
+ container_of(work, struct iowait_work, iowork);
+ struct iowait *wait = iowait_ioww_to_iow(ioww);
+ struct hfi1_ipoib_txq *txq =
+ container_of(wait, struct hfi1_ipoib_txq, wait);
+ struct net_device *dev = txq->priv->netdev;
+
+ if (likely(dev->reg_state == NETREG_REGISTERED) &&
+ likely(__netif_subqueue_stopped(dev, txq->q_idx)) &&
+ likely(!hfi1_ipoib_flush_tx_list(dev, txq)))
+ netif_wake_subqueue(dev, txq->q_idx);
+}
+
+int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
+{
+ struct net_device *dev = priv->netdev;
+ char buf[HFI1_IPOIB_TXREQ_NAME_LEN];
+ unsigned long tx_ring_size;
+ int i;
+
+ /*
+ * Ring holds 1 less than tx_ring_size
+ * Round up to next power of 2 in order to hold at least tx_queue_len
+ */
+ tx_ring_size = roundup_pow_of_two((unsigned long)dev->tx_queue_len + 1);
+
+ snprintf(buf, sizeof(buf), "hfi1_%u_ipoib_txreq_cache", priv->dd->unit);
+ priv->txreq_cache = kmem_cache_create(buf,
+ sizeof(struct ipoib_txreq),
+ 0,
+ 0,
+ NULL);
+ if (!priv->txreq_cache)
+ return -ENOMEM;
+
+ priv->tx_napis = kcalloc_node(dev->num_tx_queues,
+ sizeof(struct napi_struct),
+ GFP_ATOMIC,
+ priv->dd->node);
+ if (!priv->tx_napis)
+ goto free_txreq_cache;
+
+ priv->txqs = kcalloc_node(dev->num_tx_queues,
+ sizeof(struct hfi1_ipoib_txq),
+ GFP_ATOMIC,
+ priv->dd->node);
+ if (!priv->txqs)
+ goto free_tx_napis;
+
+ for (i = 0; i < dev->num_tx_queues; i++) {
+ struct hfi1_ipoib_txq *txq = &priv->txqs[i];
+
+ iowait_init(&txq->wait,
+ 0,
+ hfi1_ipoib_flush_txq,
+ NULL,
+ hfi1_ipoib_sdma_sleep,
+ hfi1_ipoib_sdma_wakeup,
+ NULL,
+ NULL);
+ txq->priv = priv;
+ txq->sde = NULL;
+ INIT_LIST_HEAD(&txq->tx_list);
+ atomic64_set(&txq->complete_txreqs, 0);
+ txq->q_idx = i;
+ txq->flow.tx_queue = 0xff;
+ txq->flow.sc5 = 0xff;
+ txq->pkts_sent = false;
+
+ netdev_queue_numa_node_write(netdev_get_tx_queue(dev, i),
+ priv->dd->node);
+
+ txq->tx_ring.items =
+ vzalloc_node(array_size(tx_ring_size,
+ sizeof(struct ipoib_txreq)),
+ priv->dd->node);
+ if (!txq->tx_ring.items)
+ goto free_txqs;
+
+ spin_lock_init(&txq->tx_ring.producer_lock);
+ spin_lock_init(&txq->tx_ring.consumer_lock);
+ txq->tx_ring.max_items = tx_ring_size;
+
+ txq->napi = &priv->tx_napis[i];
+ netif_tx_napi_add(dev, txq->napi,
+ hfi1_ipoib_process_tx_ring,
+ NAPI_POLL_WEIGHT);
+ }
+
+ return 0;
+
+free_txqs:
+ for (i--; i >= 0; i--) {
+ struct hfi1_ipoib_txq *txq = &priv->txqs[i];
+
+ netif_napi_del(txq->napi);
+ vfree(txq->tx_ring.items);
+ }
+
+ kfree(priv->txqs);
+ priv->txqs = NULL;
+
+free_tx_napis:
+ kfree(priv->tx_napis);
+ priv->tx_napis = NULL;
+
+free_txreq_cache:
+ kmem_cache_destroy(priv->txreq_cache);
+ priv->txreq_cache = NULL;
+ return -ENOMEM;
+}
+
+static void hfi1_ipoib_drain_tx_list(struct hfi1_ipoib_txq *txq)
+{
+ struct sdma_txreq *txreq;
+ struct sdma_txreq *txreq_tmp;
+ atomic64_t *complete_txreqs = &txq->complete_txreqs;
+
+ list_for_each_entry_safe(txreq, txreq_tmp, &txq->tx_list, list) {
+ struct ipoib_txreq *tx =
+ container_of(txreq, struct ipoib_txreq, txreq);
+
+ list_del(&txreq->list);
+ sdma_txclean(txq->priv->dd, &tx->txreq);
+ dev_kfree_skb_any(tx->skb);
+ kmem_cache_free(txq->priv->txreq_cache, tx);
+ atomic64_inc(complete_txreqs);
+ }
+
+ if (hfi1_ipoib_txreqs(txq->sent_txreqs, atomic64_read(complete_txreqs)))
+ dd_dev_warn(txq->priv->dd,
+ "txq %d not empty found %llu requests\n",
+ txq->q_idx,
+ hfi1_ipoib_txreqs(txq->sent_txreqs,
+ atomic64_read(complete_txreqs)));
+}
+
+void hfi1_ipoib_txreq_deinit(struct hfi1_ipoib_dev_priv *priv)
+{
+ int i;
+
+ for (i = 0; i < priv->netdev->num_tx_queues; i++) {
+ struct hfi1_ipoib_txq *txq = &priv->txqs[i];
+
+ iowait_cancel_work(&txq->wait);
+ iowait_sdma_drain(&txq->wait);
+ hfi1_ipoib_drain_tx_list(txq);
+ netif_napi_del(txq->napi);
+ (void)hfi1_ipoib_drain_tx_ring(txq, txq->tx_ring.max_items);
+ vfree(txq->tx_ring.items);
+ }
+
+ kfree(priv->txqs);
+ priv->txqs = NULL;
+
+ kfree(priv->tx_napis);
+ priv->tx_napis = NULL;
+
+ kmem_cache_destroy(priv->txreq_cache);
+ priv->txreq_cache = NULL;
+}
+
+void hfi1_ipoib_napi_tx_enable(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ int i;
+
+ for (i = 0; i < dev->num_tx_queues; i++) {
+ struct hfi1_ipoib_txq *txq = &priv->txqs[i];
+
+ napi_enable(txq->napi);
+ }
+}
+
+void hfi1_ipoib_napi_tx_disable(struct net_device *dev)
+{
+ struct hfi1_ipoib_dev_priv *priv = hfi1_ipoib_priv(dev);
+ int i;
+
+ for (i = 0; i < dev->num_tx_queues; i++) {
+ struct hfi1_ipoib_txq *txq = &priv->txqs[i];
+
+ napi_disable(txq->napi);
+ (void)hfi1_ipoib_drain_tx_ring(txq, txq->tx_ring.max_items);
+ }
+}
diff --git a/drivers/infiniband/hw/hfi1/msix.c b/drivers/infiniband/hw/hfi1/msix.c
index db82db497b2c..d61ee853d215 100644
--- a/drivers/infiniband/hw/hfi1/msix.c
+++ b/drivers/infiniband/hw/hfi1/msix.c
@@ -1,6 +1,6 @@
// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
/*
- * Copyright(c) 2018 Intel Corporation.
+ * Copyright(c) 2018 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -49,6 +49,7 @@
#include "hfi.h"
#include "affinity.h"
#include "sdma.h"
+#include "netdev.h"
/**
* msix_initialize() - Calculate, request and configure MSIx IRQs
@@ -69,7 +70,7 @@ int msix_initialize(struct hfi1_devdata *dd)
* one for each VNIC context
* ...any new IRQs should be added here.
*/
- total = 1 + dd->num_sdma + dd->n_krcv_queues + dd->num_vnic_contexts;
+ total = 1 + dd->num_sdma + dd->n_krcv_queues + dd->num_netdev_contexts;
if (total >= CCE_NUM_MSIX_VECTORS)
return -EINVAL;
@@ -140,7 +141,7 @@ static int msix_request_irq(struct hfi1_devdata *dd, void *arg,
ret = pci_request_irq(dd->pcidev, nr, handler, thread, arg, name);
if (ret) {
dd_dev_err(dd,
- "%s: request for IRQ %d failed, MSIx %lu, err %d\n",
+ "%s: request for IRQ %d failed, MSIx %lx, err %d\n",
name, irq, nr, ret);
spin_lock(&dd->msix_info.msix_lock);
__clear_bit(nr, dd->msix_info.in_use_msix);
@@ -160,7 +161,7 @@ static int msix_request_irq(struct hfi1_devdata *dd, void *arg,
/* This is a request, so a failure is not fatal */
ret = hfi1_get_irq_affinity(dd, me);
if (ret)
- dd_dev_err(dd, "unable to pin IRQ %d\n", ret);
+ dd_dev_err(dd, "%s: unable to pin IRQ %d\n", name, ret);
return nr;
}
@@ -171,7 +172,8 @@ static int msix_request_rcd_irq_common(struct hfi1_ctxtdata *rcd,
const char *name)
{
int nr = msix_request_irq(rcd->dd, rcd, handler, thread,
- IRQ_RCVCTXT, name);
+ rcd->is_vnic ? IRQ_NETDEVCTXT : IRQ_RCVCTXT,
+ name);
if (nr < 0)
return nr;
@@ -204,6 +206,21 @@ int msix_request_rcd_irq(struct hfi1_ctxtdata *rcd)
}
/**
+ * msix_request_rcd_irq() - Helper function for RCVAVAIL IRQs
+ * for netdev context
+ * @rcd: valid netdev contexti
+ */
+int msix_netdev_request_rcd_irq(struct hfi1_ctxtdata *rcd)
+{
+ char name[MAX_NAME_SIZE];
+
+ snprintf(name, sizeof(name), DRIVER_NAME "_%d nd kctxt%d",
+ rcd->dd->unit, rcd->ctxt);
+ return msix_request_rcd_irq_common(rcd, receive_context_interrupt_napi,
+ NULL, name);
+}
+
+/**
* msix_request_smda_ira() - Helper for getting SDMA IRQ resources
* @sde: valid sdma engine
*
@@ -355,15 +372,16 @@ void msix_clean_up_interrupts(struct hfi1_devdata *dd)
}
/**
- * msix_vnic_syncrhonize_irq() - Vnic IRQ synchronize
+ * msix_netdev_syncrhonize_irq() - netdev IRQ synchronize
* @dd: valid devdata
*/
-void msix_vnic_synchronize_irq(struct hfi1_devdata *dd)
+void msix_netdev_synchronize_irq(struct hfi1_devdata *dd)
{
int i;
+ int ctxt_count = hfi1_netdev_ctxt_count(dd);
- for (i = 0; i < dd->vnic.num_ctxt; i++) {
- struct hfi1_ctxtdata *rcd = dd->vnic.ctxt[i];
+ for (i = 0; i < ctxt_count; i++) {
+ struct hfi1_ctxtdata *rcd = hfi1_netdev_get_ctxt(dd, i);
struct hfi1_msix_entry *me;
me = &dd->msix_info.msix_entries[rcd->msix_intr];
diff --git a/drivers/infiniband/hw/hfi1/msix.h b/drivers/infiniband/hw/hfi1/msix.h
index 1a02ab7971c8..e63e944bf0fc 100644
--- a/drivers/infiniband/hw/hfi1/msix.h
+++ b/drivers/infiniband/hw/hfi1/msix.h
@@ -1,6 +1,6 @@
/* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) */
/*
- * Copyright(c) 2018 Intel Corporation.
+ * Copyright(c) 2018 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -59,7 +59,8 @@ int msix_request_rcd_irq(struct hfi1_ctxtdata *rcd);
int msix_request_sdma_irq(struct sdma_engine *sde);
void msix_free_irq(struct hfi1_devdata *dd, u8 msix_intr);
-/* VNIC interface */
-void msix_vnic_synchronize_irq(struct hfi1_devdata *dd);
+/* Netdev interface */
+void msix_netdev_synchronize_irq(struct hfi1_devdata *dd);
+int msix_netdev_request_rcd_irq(struct hfi1_ctxtdata *rcd);
#endif
diff --git a/drivers/infiniband/hw/hfi1/netdev.h b/drivers/infiniband/hw/hfi1/netdev.h
new file mode 100644
index 000000000000..947543a3e0c4
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/netdev.h
@@ -0,0 +1,118 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause) */
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+#ifndef HFI1_NETDEV_H
+#define HFI1_NETDEV_H
+
+#include "hfi.h"
+
+#include <linux/netdevice.h>
+#include <linux/xarray.h>
+
+/**
+ * struct hfi1_netdev_rxq - Receive Queue for HFI
+ * dummy netdev. Both IPoIB and VNIC netdevices will be working on
+ * top of this device.
+ * @napi: napi object
+ * @priv: ptr to netdev_priv
+ * @rcd: ptr to receive context data
+ */
+struct hfi1_netdev_rxq {
+ struct napi_struct napi;
+ struct hfi1_netdev_priv *priv;
+ struct hfi1_ctxtdata *rcd;
+};
+
+/*
+ * Number of netdev contexts used. Ensure it is less than or equal to
+ * max queues supported by VNIC (HFI1_VNIC_MAX_QUEUE).
+ */
+#define HFI1_MAX_NETDEV_CTXTS 8
+
+/* Number of NETDEV RSM entries */
+#define NUM_NETDEV_MAP_ENTRIES HFI1_MAX_NETDEV_CTXTS
+
+/**
+ * struct hfi1_netdev_priv: data required to setup and run HFI netdev.
+ * @dd: hfi1_devdata
+ * @rxq: pointer to dummy netdev receive queues.
+ * @num_rx_q: number of receive queues
+ * @rmt_index: first free index in RMT Array
+ * @msix_start: first free MSI-X interrupt vector.
+ * @dev_tbl: netdev table for unique identifier VNIC and IPoIb VLANs.
+ * @enabled: atomic counter of netdevs enabling receive queues.
+ * When 0 NAPI will be disabled.
+ * @netdevs: atomic counter of netdevs using dummy netdev.
+ * When 0 receive queues will be freed.
+ */
+struct hfi1_netdev_priv {
+ struct hfi1_devdata *dd;
+ struct hfi1_netdev_rxq *rxq;
+ int num_rx_q;
+ int rmt_start;
+ struct xarray dev_tbl;
+ /* count of enabled napi polls */
+ atomic_t enabled;
+ /* count of netdevs on top */
+ atomic_t netdevs;
+};
+
+static inline
+struct hfi1_netdev_priv *hfi1_netdev_priv(struct net_device *dev)
+{
+ return (struct hfi1_netdev_priv *)&dev[1];
+}
+
+static inline
+int hfi1_netdev_ctxt_count(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return priv->num_rx_q;
+}
+
+static inline
+struct hfi1_ctxtdata *hfi1_netdev_get_ctxt(struct hfi1_devdata *dd, int ctxt)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return priv->rxq[ctxt].rcd;
+}
+
+static inline
+int hfi1_netdev_get_free_rmt_idx(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return priv->rmt_start;
+}
+
+static inline
+void hfi1_netdev_set_free_rmt_idx(struct hfi1_devdata *dd, int rmt_idx)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ priv->rmt_start = rmt_idx;
+}
+
+u32 hfi1_num_netdev_contexts(struct hfi1_devdata *dd, u32 available_contexts,
+ struct cpumask *cpu_mask);
+
+void hfi1_netdev_enable_queues(struct hfi1_devdata *dd);
+void hfi1_netdev_disable_queues(struct hfi1_devdata *dd);
+int hfi1_netdev_rx_init(struct hfi1_devdata *dd);
+int hfi1_netdev_rx_destroy(struct hfi1_devdata *dd);
+int hfi1_netdev_alloc(struct hfi1_devdata *dd);
+void hfi1_netdev_free(struct hfi1_devdata *dd);
+int hfi1_netdev_add_data(struct hfi1_devdata *dd, int id, void *data);
+void *hfi1_netdev_remove_data(struct hfi1_devdata *dd, int id);
+void *hfi1_netdev_get_data(struct hfi1_devdata *dd, int id);
+void *hfi1_netdev_get_first_data(struct hfi1_devdata *dd, int *start_id);
+
+/* chip.c */
+int hfi1_netdev_rx_napi(struct napi_struct *napi, int budget);
+
+#endif /* HFI1_NETDEV_H */
diff --git a/drivers/infiniband/hw/hfi1/netdev_rx.c b/drivers/infiniband/hw/hfi1/netdev_rx.c
new file mode 100644
index 000000000000..63688e85e8da
--- /dev/null
+++ b/drivers/infiniband/hw/hfi1/netdev_rx.c
@@ -0,0 +1,481 @@
+// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+/*
+ * Copyright(c) 2020 Intel Corporation.
+ *
+ */
+
+/*
+ * This file contains HFI1 support for netdev RX functionality
+ */
+
+#include "sdma.h"
+#include "verbs.h"
+#include "netdev.h"
+#include "hfi.h"
+
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <rdma/ib_verbs.h>
+
+static int hfi1_netdev_setup_ctxt(struct hfi1_netdev_priv *priv,
+ struct hfi1_ctxtdata *uctxt)
+{
+ unsigned int rcvctrl_ops;
+ struct hfi1_devdata *dd = priv->dd;
+ int ret;
+
+ uctxt->rhf_rcv_function_map = netdev_rhf_rcv_functions;
+ uctxt->do_interrupt = &handle_receive_interrupt_napi_sp;
+
+ /* Now allocate the RcvHdr queue and eager buffers. */
+ ret = hfi1_create_rcvhdrq(dd, uctxt);
+ if (ret)
+ goto done;
+
+ ret = hfi1_setup_eagerbufs(uctxt);
+ if (ret)
+ goto done;
+
+ clear_rcvhdrtail(uctxt);
+
+ rcvctrl_ops = HFI1_RCVCTRL_CTXT_DIS;
+ rcvctrl_ops |= HFI1_RCVCTRL_INTRAVAIL_DIS;
+
+ if (!HFI1_CAP_KGET_MASK(uctxt->flags, MULTI_PKT_EGR))
+ rcvctrl_ops |= HFI1_RCVCTRL_ONE_PKT_EGR_ENB;
+ if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_EGR_FULL))
+ rcvctrl_ops |= HFI1_RCVCTRL_NO_EGR_DROP_ENB;
+ if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_RHQ_FULL))
+ rcvctrl_ops |= HFI1_RCVCTRL_NO_RHQ_DROP_ENB;
+ if (HFI1_CAP_KGET_MASK(uctxt->flags, DMA_RTAIL))
+ rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_ENB;
+
+ hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt);
+done:
+ return ret;
+}
+
+static int hfi1_netdev_allocate_ctxt(struct hfi1_devdata *dd,
+ struct hfi1_ctxtdata **ctxt)
+{
+ struct hfi1_ctxtdata *uctxt;
+ int ret;
+
+ if (dd->flags & HFI1_FROZEN)
+ return -EIO;
+
+ ret = hfi1_create_ctxtdata(dd->pport, dd->node, &uctxt);
+ if (ret < 0) {
+ dd_dev_err(dd, "Unable to create ctxtdata, failing open\n");
+ return -ENOMEM;
+ }
+
+ uctxt->flags = HFI1_CAP_KGET(MULTI_PKT_EGR) |
+ HFI1_CAP_KGET(NODROP_RHQ_FULL) |
+ HFI1_CAP_KGET(NODROP_EGR_FULL) |
+ HFI1_CAP_KGET(DMA_RTAIL);
+ /* Netdev contexts are always NO_RDMA_RTAIL */
+ uctxt->fast_handler = handle_receive_interrupt_napi_fp;
+ uctxt->slow_handler = handle_receive_interrupt_napi_sp;
+ hfi1_set_seq_cnt(uctxt, 1);
+ uctxt->is_vnic = true;
+
+ hfi1_stats.sps_ctxts++;
+
+ dd_dev_info(dd, "created netdev context %d\n", uctxt->ctxt);
+ *ctxt = uctxt;
+
+ return 0;
+}
+
+static void hfi1_netdev_deallocate_ctxt(struct hfi1_devdata *dd,
+ struct hfi1_ctxtdata *uctxt)
+{
+ flush_wc();
+
+ /*
+ * Disable receive context and interrupt available, reset all
+ * RcvCtxtCtrl bits to default values.
+ */
+ hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_DIS |
+ HFI1_RCVCTRL_TIDFLOW_DIS |
+ HFI1_RCVCTRL_INTRAVAIL_DIS |
+ HFI1_RCVCTRL_ONE_PKT_EGR_DIS |
+ HFI1_RCVCTRL_NO_RHQ_DROP_DIS |
+ HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt);
+
+ if (uctxt->msix_intr != CCE_NUM_MSIX_VECTORS)
+ msix_free_irq(dd, uctxt->msix_intr);
+
+ uctxt->msix_intr = CCE_NUM_MSIX_VECTORS;
+ uctxt->event_flags = 0;
+
+ hfi1_clear_tids(uctxt);
+ hfi1_clear_ctxt_pkey(dd, uctxt);
+
+ hfi1_stats.sps_ctxts--;
+
+ hfi1_free_ctxt(uctxt);
+}
+
+static int hfi1_netdev_allot_ctxt(struct hfi1_netdev_priv *priv,
+ struct hfi1_ctxtdata **ctxt)
+{
+ int rc;
+ struct hfi1_devdata *dd = priv->dd;
+
+ rc = hfi1_netdev_allocate_ctxt(dd, ctxt);
+ if (rc) {
+ dd_dev_err(dd, "netdev ctxt alloc failed %d\n", rc);
+ return rc;
+ }
+
+ rc = hfi1_netdev_setup_ctxt(priv, *ctxt);
+ if (rc) {
+ dd_dev_err(dd, "netdev ctxt setup failed %d\n", rc);
+ hfi1_netdev_deallocate_ctxt(dd, *ctxt);
+ *ctxt = NULL;
+ }
+
+ return rc;
+}
+
+/**
+ * hfi1_num_netdev_contexts - Count of netdev recv contexts to use.
+ * @dd: device on which to allocate netdev contexts
+ * @available_contexts: count of available receive contexts
+ * @cpu_mask: mask of possible cpus to include for contexts
+ *
+ * Return: count of physical cores on a node or the remaining available recv
+ * contexts for netdev recv context usage up to the maximum of
+ * HFI1_MAX_NETDEV_CTXTS.
+ * A value of 0 can be returned when acceleration is explicitly turned off,
+ * a memory allocation error occurs or when there are no available contexts.
+ *
+ */
+u32 hfi1_num_netdev_contexts(struct hfi1_devdata *dd, u32 available_contexts,
+ struct cpumask *cpu_mask)
+{
+ cpumask_var_t node_cpu_mask;
+ unsigned int available_cpus;
+
+ if (!HFI1_CAP_IS_KSET(AIP))
+ return 0;
+
+ /* Always give user contexts priority over netdev contexts */
+ if (available_contexts == 0) {
+ dd_dev_info(dd, "No receive contexts available for netdevs.\n");
+ return 0;
+ }
+
+ if (!zalloc_cpumask_var(&node_cpu_mask, GFP_KERNEL)) {
+ dd_dev_err(dd, "Unable to allocate cpu_mask for netdevs.\n");
+ return 0;
+ }
+
+ cpumask_and(node_cpu_mask, cpu_mask,
+ cpumask_of_node(pcibus_to_node(dd->pcidev->bus)));
+
+ available_cpus = cpumask_weight(node_cpu_mask);
+
+ free_cpumask_var(node_cpu_mask);
+
+ return min3(available_cpus, available_contexts,
+ (u32)HFI1_MAX_NETDEV_CTXTS);
+}
+
+static int hfi1_netdev_rxq_init(struct net_device *dev)
+{
+ int i;
+ int rc;
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dev);
+ struct hfi1_devdata *dd = priv->dd;
+
+ priv->num_rx_q = dd->num_netdev_contexts;
+ priv->rxq = kcalloc_node(priv->num_rx_q, sizeof(struct hfi1_netdev_rxq),
+ GFP_KERNEL, dd->node);
+
+ if (!priv->rxq) {
+ dd_dev_err(dd, "Unable to allocate netdev queue data\n");
+ return (-ENOMEM);
+ }
+
+ for (i = 0; i < priv->num_rx_q; i++) {
+ struct hfi1_netdev_rxq *rxq = &priv->rxq[i];
+
+ rc = hfi1_netdev_allot_ctxt(priv, &rxq->rcd);
+ if (rc)
+ goto bail_context_irq_failure;
+
+ hfi1_rcd_get(rxq->rcd);
+ rxq->priv = priv;
+ rxq->rcd->napi = &rxq->napi;
+ dd_dev_info(dd, "Setting rcv queue %d napi to context %d\n",
+ i, rxq->rcd->ctxt);
+ /*
+ * Disable BUSY_POLL on this NAPI as this is not supported
+ * right now.
+ */
+ set_bit(NAPI_STATE_NO_BUSY_POLL, &rxq->napi.state);
+ netif_napi_add(dev, &rxq->napi, hfi1_netdev_rx_napi, 64);
+ rc = msix_netdev_request_rcd_irq(rxq->rcd);
+ if (rc)
+ goto bail_context_irq_failure;
+ }
+
+ return 0;
+
+bail_context_irq_failure:
+ dd_dev_err(dd, "Unable to allot receive context\n");
+ for (; i >= 0; i--) {
+ struct hfi1_netdev_rxq *rxq = &priv->rxq[i];
+
+ if (rxq->rcd) {
+ hfi1_netdev_deallocate_ctxt(dd, rxq->rcd);
+ hfi1_rcd_put(rxq->rcd);
+ rxq->rcd = NULL;
+ }
+ }
+ kfree(priv->rxq);
+ priv->rxq = NULL;
+
+ return rc;
+}
+
+static void hfi1_netdev_rxq_deinit(struct net_device *dev)
+{
+ int i;
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dev);
+ struct hfi1_devdata *dd = priv->dd;
+
+ for (i = 0; i < priv->num_rx_q; i++) {
+ struct hfi1_netdev_rxq *rxq = &priv->rxq[i];
+
+ netif_napi_del(&rxq->napi);
+ hfi1_netdev_deallocate_ctxt(dd, rxq->rcd);
+ hfi1_rcd_put(rxq->rcd);
+ rxq->rcd = NULL;
+ }
+
+ kfree(priv->rxq);
+ priv->rxq = NULL;
+ priv->num_rx_q = 0;
+}
+
+static void enable_queues(struct hfi1_netdev_priv *priv)
+{
+ int i;
+
+ for (i = 0; i < priv->num_rx_q; i++) {
+ struct hfi1_netdev_rxq *rxq = &priv->rxq[i];
+
+ dd_dev_info(priv->dd, "enabling queue %d on context %d\n", i,
+ rxq->rcd->ctxt);
+ napi_enable(&rxq->napi);
+ hfi1_rcvctrl(priv->dd,
+ HFI1_RCVCTRL_CTXT_ENB | HFI1_RCVCTRL_INTRAVAIL_ENB,
+ rxq->rcd);
+ }
+}
+
+static void disable_queues(struct hfi1_netdev_priv *priv)
+{
+ int i;
+
+ msix_netdev_synchronize_irq(priv->dd);
+
+ for (i = 0; i < priv->num_rx_q; i++) {
+ struct hfi1_netdev_rxq *rxq = &priv->rxq[i];
+
+ dd_dev_info(priv->dd, "disabling queue %d on context %d\n", i,
+ rxq->rcd->ctxt);
+
+ /* wait for napi if it was scheduled */
+ hfi1_rcvctrl(priv->dd,
+ HFI1_RCVCTRL_CTXT_DIS | HFI1_RCVCTRL_INTRAVAIL_DIS,
+ rxq->rcd);
+ napi_synchronize(&rxq->napi);
+ napi_disable(&rxq->napi);
+ }
+}
+
+/**
+ * hfi1_netdev_rx_init - Incrememnts netdevs counter. When called first time,
+ * it allocates receive queue data and calls netif_napi_add
+ * for each queue.
+ *
+ * @dd: hfi1 dev data
+ */
+int hfi1_netdev_rx_init(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+ int res;
+
+ if (atomic_fetch_inc(&priv->netdevs))
+ return 0;
+
+ mutex_lock(&hfi1_mutex);
+ init_dummy_netdev(dd->dummy_netdev);
+ res = hfi1_netdev_rxq_init(dd->dummy_netdev);
+ mutex_unlock(&hfi1_mutex);
+ return res;
+}
+
+/**
+ * hfi1_netdev_rx_destroy - Decrements netdevs counter, when it reaches 0
+ * napi is deleted and receive queses memory is freed.
+ *
+ * @dd: hfi1 dev data
+ */
+int hfi1_netdev_rx_destroy(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ /* destroy the RX queues only if it is the last netdev going away */
+ if (atomic_fetch_add_unless(&priv->netdevs, -1, 0) == 1) {
+ mutex_lock(&hfi1_mutex);
+ hfi1_netdev_rxq_deinit(dd->dummy_netdev);
+ mutex_unlock(&hfi1_mutex);
+ }
+
+ return 0;
+}
+
+/**
+ * hfi1_netdev_alloc - Allocates netdev and private data. It is required
+ * because RMT index and MSI-X interrupt can be set only
+ * during driver initialization.
+ *
+ * @dd: hfi1 dev data
+ */
+int hfi1_netdev_alloc(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv;
+ const int netdev_size = sizeof(*dd->dummy_netdev) +
+ sizeof(struct hfi1_netdev_priv);
+
+ dd_dev_info(dd, "allocating netdev size %d\n", netdev_size);
+ dd->dummy_netdev = kcalloc_node(1, netdev_size, GFP_KERNEL, dd->node);
+
+ if (!dd->dummy_netdev)
+ return -ENOMEM;
+
+ priv = hfi1_netdev_priv(dd->dummy_netdev);
+ priv->dd = dd;
+ xa_init(&priv->dev_tbl);
+ atomic_set(&priv->enabled, 0);
+ atomic_set(&priv->netdevs, 0);
+
+ return 0;
+}
+
+void hfi1_netdev_free(struct hfi1_devdata *dd)
+{
+ if (dd->dummy_netdev) {
+ dd_dev_info(dd, "hfi1 netdev freed\n");
+ free_netdev(dd->dummy_netdev);
+ dd->dummy_netdev = NULL;
+ }
+}
+
+/**
+ * hfi1_netdev_enable_queues - This is napi enable function.
+ * It enables napi objects associated with queues.
+ * When at least one device has called it it increments atomic counter.
+ * Disable function decrements counter and when it is 0,
+ * calls napi_disable for every queue.
+ *
+ * @dd: hfi1 dev data
+ */
+void hfi1_netdev_enable_queues(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv;
+
+ if (!dd->dummy_netdev)
+ return;
+
+ priv = hfi1_netdev_priv(dd->dummy_netdev);
+ if (atomic_fetch_inc(&priv->enabled))
+ return;
+
+ mutex_lock(&hfi1_mutex);
+ enable_queues(priv);
+ mutex_unlock(&hfi1_mutex);
+}
+
+void hfi1_netdev_disable_queues(struct hfi1_devdata *dd)
+{
+ struct hfi1_netdev_priv *priv;
+
+ if (!dd->dummy_netdev)
+ return;
+
+ priv = hfi1_netdev_priv(dd->dummy_netdev);
+ if (atomic_dec_if_positive(&priv->enabled))
+ return;
+
+ mutex_lock(&hfi1_mutex);
+ disable_queues(priv);
+ mutex_unlock(&hfi1_mutex);
+}
+
+/**
+ * hfi1_netdev_add_data - Registers data with unique identifier
+ * to be requested later this is needed for VNIC and IPoIB VLANs
+ * implementations.
+ * This call is protected by mutex idr_lock.
+ *
+ * @dd: hfi1 dev data
+ * @id: requested integer id up to INT_MAX
+ * @data: data to be associated with index
+ */
+int hfi1_netdev_add_data(struct hfi1_devdata *dd, int id, void *data)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return xa_insert(&priv->dev_tbl, id, data, GFP_NOWAIT);
+}
+
+/**
+ * hfi1_netdev_remove_data - Removes data with previously given id.
+ * Returns the reference to removed entry.
+ *
+ * @dd: hfi1 dev data
+ * @id: requested integer id up to INT_MAX
+ */
+void *hfi1_netdev_remove_data(struct hfi1_devdata *dd, int id)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return xa_erase(&priv->dev_tbl, id);
+}
+
+/**
+ * hfi1_netdev_get_data - Gets data with given id
+ *
+ * @dd: hfi1 dev data
+ * @id: requested integer id up to INT_MAX
+ */
+void *hfi1_netdev_get_data(struct hfi1_devdata *dd, int id)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+
+ return xa_load(&priv->dev_tbl, id);
+}
+
+/**
+ * hfi1_netdev_get_first_dat - Gets first entry with greater or equal id.
+ *
+ * @dd: hfi1 dev data
+ * @id: requested integer id up to INT_MAX
+ */
+void *hfi1_netdev_get_first_data(struct hfi1_devdata *dd, int *start_id)
+{
+ struct hfi1_netdev_priv *priv = hfi1_netdev_priv(dd->dummy_netdev);
+ unsigned long index = *start_id;
+ void *ret;
+
+ ret = xa_find(&priv->dev_tbl, &index, UINT_MAX, XA_PRESENT);
+ *start_id = (int)index;
+ return ret;
+}
diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c
index f8e733aa3bb8..0c2ae9f7b3e8 100644
--- a/drivers/infiniband/hw/hfi1/qp.c
+++ b/drivers/infiniband/hw/hfi1/qp.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2019 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -186,15 +186,6 @@ static void flush_iowait(struct rvt_qp *qp)
write_sequnlock_irqrestore(lock, flags);
}
-static inline int opa_mtu_enum_to_int(int mtu)
-{
- switch (mtu) {
- case OPA_MTU_8192: return 8192;
- case OPA_MTU_10240: return 10240;
- default: return -1;
- }
-}
-
/**
* This function is what we would push to the core layer if we wanted to be a
* "first class citizen". Instead we hide this here and rely on Verbs ULPs
@@ -202,15 +193,10 @@ static inline int opa_mtu_enum_to_int(int mtu)
*/
static inline int verbs_mtu_enum_to_int(struct ib_device *dev, enum ib_mtu mtu)
{
- int val;
-
/* Constraining 10KB packets to 8KB packets */
if (mtu == (enum ib_mtu)OPA_MTU_10240)
mtu = OPA_MTU_8192;
- val = opa_mtu_enum_to_int((int)mtu);
- if (val > 0)
- return val;
- return ib_mtu_enum_to_int(mtu);
+ return opa_mtu_enum_to_int((enum opa_mtu)mtu);
}
int hfi1_check_modify_qp(struct rvt_qp *qp, struct ib_qp_attr *attr,
diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c
index 8a2e0d9351e9..243b4ba0b6f6 100644
--- a/drivers/infiniband/hw/hfi1/tid_rdma.c
+++ b/drivers/infiniband/hw/hfi1/tid_rdma.c
@@ -1,6 +1,6 @@
// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
/*
- * Copyright(c) 2018 Intel Corporation.
+ * Copyright(c) 2018 - 2020 Intel Corporation.
*
*/
@@ -194,7 +194,7 @@ void tid_rdma_opfn_init(struct rvt_qp *qp, struct tid_rdma_params *p)
{
struct hfi1_qp_priv *priv = qp->priv;
- p->qp = (kdeth_qp << 16) | priv->rcd->ctxt;
+ p->qp = (RVT_KDETH_QP_PREFIX << 16) | priv->rcd->ctxt;
p->max_len = TID_RDMA_MAX_SEGMENT_SIZE;
p->jkey = priv->rcd->jkey;
p->max_read = TID_RDMA_MAX_READ_SEGS_PER_REQ;
diff --git a/drivers/infiniband/hw/hfi1/trace.c b/drivers/infiniband/hw/hfi1/trace.c
index 9a3d236bcc88..b219ea90fd6f 100644
--- a/drivers/infiniband/hw/hfi1/trace.c
+++ b/drivers/infiniband/hw/hfi1/trace.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -47,6 +47,7 @@
#define CREATE_TRACE_POINTS
#include "trace.h"
#include "exp_rcv.h"
+#include "ipoib.h"
static u8 __get_ib_hdr_len(struct ib_header *hdr)
{
@@ -126,6 +127,7 @@ const char *hfi1_trace_get_packet_l2_str(u8 l2)
#define RETH_PRN "reth vaddr:0x%.16llx rkey:0x%.8x dlen:0x%.8x"
#define AETH_PRN "aeth syn:0x%.2x %s msn:0x%.8x"
#define DETH_PRN "deth qkey:0x%.8x sqpn:0x%.6x"
+#define DETH_ENTROPY_PRN "deth qkey:0x%.8x sqpn:0x%.6x entropy:0x%.2x"
#define IETH_PRN "ieth rkey:0x%.8x"
#define ATOMICACKETH_PRN "origdata:%llx"
#define ATOMICETH_PRN "vaddr:0x%llx rkey:0x%.8x sdata:%llx cdata:%llx"
@@ -444,6 +446,12 @@ const char *parse_everbs_hdrs(
break;
/* deth */
case OP(UD, SEND_ONLY):
+ trace_seq_printf(p, DETH_ENTROPY_PRN,
+ be32_to_cpu(eh->ud.deth[0]),
+ be32_to_cpu(eh->ud.deth[1]) & RVT_QPN_MASK,
+ be32_to_cpu(eh->ud.deth[1]) >>
+ HFI1_IPOIB_ENTROPY_SHIFT);
+ break;
case OP(UD, SEND_ONLY_WITH_IMMEDIATE):
trace_seq_printf(p, DETH_PRN,
be32_to_cpu(eh->ud.deth[0]),
@@ -512,6 +520,38 @@ u16 hfi1_trace_get_tid_idx(u32 ent)
return EXP_TID_GET(ent, IDX);
}
+struct hfi1_ctxt_hist {
+ atomic_t count;
+ atomic_t data[255];
+};
+
+struct hfi1_ctxt_hist hist = {
+ .count = ATOMIC_INIT(0)
+};
+
+const char *hfi1_trace_print_rsm_hist(struct trace_seq *p, unsigned int ctxt)
+{
+ int i, len = ARRAY_SIZE(hist.data);
+ const char *ret = trace_seq_buffer_ptr(p);
+ unsigned long packet_count = atomic_fetch_inc(&hist.count);
+
+ trace_seq_printf(p, "packet[%lu]", packet_count);
+ for (i = 0; i < len; ++i) {
+ unsigned long val;
+ atomic_t *count = &hist.data[i];
+
+ if (ctxt == i)
+ val = atomic_fetch_inc(count);
+ else
+ val = atomic_read(count);
+
+ if (val)
+ trace_seq_printf(p, "(%d:%lu)", i, val);
+ }
+ trace_seq_putc(p, 0);
+ return ret;
+}
+
__hfi1_trace_fn(AFFINITY);
__hfi1_trace_fn(PKT);
__hfi1_trace_fn(PROC);
diff --git a/drivers/infiniband/hw/hfi1/trace_ctxts.h b/drivers/infiniband/hw/hfi1/trace_ctxts.h
index b5fc5c6cd52f..d8c168dc3ea8 100644
--- a/drivers/infiniband/hw/hfi1/trace_ctxts.h
+++ b/drivers/infiniband/hw/hfi1/trace_ctxts.h
@@ -1,5 +1,5 @@
/*
-* Copyright(c) 2015, 2016 Intel Corporation.
+* Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -138,6 +138,15 @@ TRACE_EVENT(hfi1_ctxt_info,
)
);
+const char *hfi1_trace_print_rsm_hist(struct trace_seq *p, unsigned int ctxt);
+TRACE_EVENT(ctxt_rsm_hist,
+ TP_PROTO(unsigned int ctxt),
+ TP_ARGS(ctxt),
+ TP_STRUCT__entry(__field(unsigned int, ctxt)),
+ TP_fast_assign(__entry->ctxt = ctxt;),
+ TP_printk("%s", hfi1_trace_print_rsm_hist(p, __entry->ctxt))
+);
+
#endif /* __HFI1_TRACE_CTXTS_H */
#undef TRACE_INCLUDE_PATH
diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c
index 2f6323ad9c59..30865635b449 100644
--- a/drivers/infiniband/hw/hfi1/verbs.c
+++ b/drivers/infiniband/hw/hfi1/verbs.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2018 Intel Corporation.
+ * Copyright(c) 2015 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -66,6 +66,7 @@
#include "vnic.h"
#include "fault.h"
#include "affinity.h"
+#include "ipoib.h"
static unsigned int hfi1_lkey_table_size = 16;
module_param_named(lkey_table_size, hfi1_lkey_table_size, uint,
@@ -1342,7 +1343,7 @@ static void hfi1_fill_device_attr(struct hfi1_devdata *dd)
IB_DEVICE_SYS_IMAGE_GUID | IB_DEVICE_RC_RNR_NAK_GEN |
IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SRQ_RESIZE |
IB_DEVICE_MEM_MGT_EXTENSIONS |
- IB_DEVICE_RDMA_NETDEV_OPA_VNIC;
+ IB_DEVICE_RDMA_NETDEV_OPA;
rdi->dparms.props.page_size_cap = PAGE_SIZE;
rdi->dparms.props.vendor_id = dd->oui1 << 16 | dd->oui2 << 8 | dd->oui3;
rdi->dparms.props.vendor_part_id = dd->pcidev->device;
@@ -1360,7 +1361,6 @@ static void hfi1_fill_device_attr(struct hfi1_devdata *dd)
rdi->dparms.props.max_cq = hfi1_max_cqs;
rdi->dparms.props.max_ah = hfi1_max_ahs;
rdi->dparms.props.max_cqe = hfi1_max_cqes;
- rdi->dparms.props.max_map_per_fmr = 32767;
rdi->dparms.props.max_pd = hfi1_max_pds;
rdi->dparms.props.max_qp_rd_atom = HFI1_MAX_RDMA_ATOMIC;
rdi->dparms.props.max_qp_init_rd_atom = 255;
@@ -1439,6 +1439,8 @@ static int query_port(struct rvt_dev_info *rdi, u8 port_num,
4096 : hfi1_max_mtu), IB_MTU_4096);
props->active_mtu = !valid_ib_mtu(ppd->ibmtu) ? props->max_mtu :
mtu_to_enum(ppd->ibmtu, IB_MTU_4096);
+ props->phys_mtu = HFI1_CAP_IS_KSET(AIP) ? hfi1_max_mtu :
+ ib_mtu_enum_to_int(props->max_mtu);
return 0;
}
@@ -1793,6 +1795,7 @@ static const struct ib_device_ops hfi1_dev_ops = {
.modify_device = modify_device,
/* keep process mad in the driver */
.process_mad = hfi1_process_mad,
+ .rdma_netdev_get_params = hfi1_ipoib_rn_get_params,
};
/**
@@ -1863,9 +1866,8 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd)
dd->verbs_dev.rdi.dparms.qpn_start = 0;
dd->verbs_dev.rdi.dparms.qpn_inc = 1;
dd->verbs_dev.rdi.dparms.qos_shift = dd->qos_shift;
- dd->verbs_dev.rdi.dparms.qpn_res_start = kdeth_qp << 16;
- dd->verbs_dev.rdi.dparms.qpn_res_end =
- dd->verbs_dev.rdi.dparms.qpn_res_start + 65535;
+ dd->verbs_dev.rdi.dparms.qpn_res_start = RVT_KDETH_QP_BASE;
+ dd->verbs_dev.rdi.dparms.qpn_res_end = RVT_AIP_QP_MAX;
dd->verbs_dev.rdi.dparms.max_rdma_atomic = HFI1_MAX_RDMA_ATOMIC;
dd->verbs_dev.rdi.dparms.psn_mask = PSN_MASK;
dd->verbs_dev.rdi.dparms.psn_shift = PSN_SHIFT;
diff --git a/drivers/infiniband/hw/hfi1/vnic.h b/drivers/infiniband/hw/hfi1/vnic.h
index 5ae781514e32..66150a13f374 100644
--- a/drivers/infiniband/hw/hfi1/vnic.h
+++ b/drivers/infiniband/hw/hfi1/vnic.h
@@ -1,7 +1,7 @@
#ifndef _HFI1_VNIC_H
#define _HFI1_VNIC_H
/*
- * Copyright(c) 2017 Intel Corporation.
+ * Copyright(c) 2017 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -69,6 +69,7 @@
#define HFI1_VNIC_SC_SHIFT 4
#define HFI1_VNIC_MAX_QUEUE 16
+#define HFI1_NUM_VNIC_CTXT 8
/**
* struct hfi1_vnic_sdma - VNIC per Tx ring SDMA information
@@ -104,7 +105,6 @@ struct hfi1_vnic_rx_queue {
struct hfi1_vnic_vport_info *vinfo;
struct net_device *netdev;
struct napi_struct napi;
- struct sk_buff_head skbq;
};
/**
@@ -146,7 +146,6 @@ struct hfi1_vnic_vport_info {
/* vnic hfi1 internal functions */
void hfi1_vnic_setup(struct hfi1_devdata *dd);
-void hfi1_vnic_cleanup(struct hfi1_devdata *dd);
int hfi1_vnic_txreq_init(struct hfi1_devdata *dd);
void hfi1_vnic_txreq_deinit(struct hfi1_devdata *dd);
diff --git a/drivers/infiniband/hw/hfi1/vnic_main.c b/drivers/infiniband/hw/hfi1/vnic_main.c
index 6b14581b9965..a90824de0f57 100644
--- a/drivers/infiniband/hw/hfi1/vnic_main.c
+++ b/drivers/infiniband/hw/hfi1/vnic_main.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2017 - 2018 Intel Corporation.
+ * Copyright(c) 2017 - 2020 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -53,6 +53,7 @@
#include <linux/if_vlan.h>
#include "vnic.h"
+#include "netdev.h"
#define HFI_TX_TIMEOUT_MS 1000
@@ -62,114 +63,6 @@
static DEFINE_SPINLOCK(vport_cntr_lock);
-static int setup_vnic_ctxt(struct hfi1_devdata *dd, struct hfi1_ctxtdata *uctxt)
-{
- unsigned int rcvctrl_ops = 0;
- int ret;
-
- uctxt->do_interrupt = &handle_receive_interrupt;
-
- /* Now allocate the RcvHdr queue and eager buffers. */
- ret = hfi1_create_rcvhdrq(dd, uctxt);
- if (ret)
- goto done;
-
- ret = hfi1_setup_eagerbufs(uctxt);
- if (ret)
- goto done;
-
- if (hfi1_rcvhdrtail_kvaddr(uctxt))
- clear_rcvhdrtail(uctxt);
-
- rcvctrl_ops = HFI1_RCVCTRL_CTXT_ENB;
- rcvctrl_ops |= HFI1_RCVCTRL_INTRAVAIL_ENB;
-
- if (!HFI1_CAP_KGET_MASK(uctxt->flags, MULTI_PKT_EGR))
- rcvctrl_ops |= HFI1_RCVCTRL_ONE_PKT_EGR_ENB;
- if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_EGR_FULL))
- rcvctrl_ops |= HFI1_RCVCTRL_NO_EGR_DROP_ENB;
- if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_RHQ_FULL))
- rcvctrl_ops |= HFI1_RCVCTRL_NO_RHQ_DROP_ENB;
- if (HFI1_CAP_KGET_MASK(uctxt->flags, DMA_RTAIL))
- rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_ENB;
-
- hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt);
-done:
- return ret;
-}
-
-static int allocate_vnic_ctxt(struct hfi1_devdata *dd,
- struct hfi1_ctxtdata **vnic_ctxt)
-{
- struct hfi1_ctxtdata *uctxt;
- int ret;
-
- if (dd->flags & HFI1_FROZEN)
- return -EIO;
-
- ret = hfi1_create_ctxtdata(dd->pport, dd->node, &uctxt);
- if (ret < 0) {
- dd_dev_err(dd, "Unable to create ctxtdata, failing open\n");
- return -ENOMEM;
- }
-
- uctxt->flags = HFI1_CAP_KGET(MULTI_PKT_EGR) |
- HFI1_CAP_KGET(NODROP_RHQ_FULL) |
- HFI1_CAP_KGET(NODROP_EGR_FULL) |
- HFI1_CAP_KGET(DMA_RTAIL);
- uctxt->seq_cnt = 1;
- uctxt->is_vnic = true;
-
- msix_request_rcd_irq(uctxt);
-
- hfi1_stats.sps_ctxts++;
- dd_dev_dbg(dd, "created vnic context %d\n", uctxt->ctxt);
- *vnic_ctxt = uctxt;
-
- return 0;
-}
-
-static void deallocate_vnic_ctxt(struct hfi1_devdata *dd,
- struct hfi1_ctxtdata *uctxt)
-{
- dd_dev_dbg(dd, "closing vnic context %d\n", uctxt->ctxt);
- flush_wc();
-
- /*
- * Disable receive context and interrupt available, reset all
- * RcvCtxtCtrl bits to default values.
- */
- hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_DIS |
- HFI1_RCVCTRL_TIDFLOW_DIS |
- HFI1_RCVCTRL_INTRAVAIL_DIS |
- HFI1_RCVCTRL_ONE_PKT_EGR_DIS |
- HFI1_RCVCTRL_NO_RHQ_DROP_DIS |
- HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt);
-
- /* msix_intr will always be > 0, only clean up if this is true */
- if (uctxt->msix_intr)
- msix_free_irq(dd, uctxt->msix_intr);
-
- uctxt->event_flags = 0;
-
- hfi1_clear_tids(uctxt);
- hfi1_clear_ctxt_pkey(dd, uctxt);
-
- hfi1_stats.sps_ctxts--;
-
- hfi1_free_ctxt(uctxt);
-}
-
-void hfi1_vnic_setup(struct hfi1_devdata *dd)
-{
- xa_init(&dd->vnic.vesws);
-}
-
-void hfi1_vnic_cleanup(struct hfi1_devdata *dd)
-{
- WARN_ON(!xa_empty(&dd->vnic.vesws));
-}
-
#define SUM_GRP_COUNTERS(stats, qstats, x_grp) do { \
u64 *src64, *dst64; \
for (src64 = &qstats->x_grp.unicast, \
@@ -179,6 +72,9 @@ void hfi1_vnic_cleanup(struct hfi1_devdata *dd)
} \
} while (0)
+#define VNIC_MASK (0xFF)
+#define VNIC_ID(val) ((1ull << 24) | ((val) & VNIC_MASK))
+
/* hfi1_vnic_update_stats - update statistics */
static void hfi1_vnic_update_stats(struct hfi1_vnic_vport_info *vinfo,
struct opa_vnic_stats *stats)
@@ -454,71 +350,25 @@ static inline int hfi1_vnic_decap_skb(struct hfi1_vnic_rx_queue *rxq,
return rc;
}
-static inline struct sk_buff *hfi1_vnic_get_skb(struct hfi1_vnic_rx_queue *rxq)
+static struct hfi1_vnic_vport_info *get_vnic_port(struct hfi1_devdata *dd,
+ int vesw_id)
{
- unsigned char *pad_info;
- struct sk_buff *skb;
+ int vnic_id = VNIC_ID(vesw_id);
- skb = skb_dequeue(&rxq->skbq);
- if (unlikely(!skb))
- return NULL;
-
- /* remove tail padding and icrc */
- pad_info = skb->data + skb->len - 1;
- skb_trim(skb, (skb->len - OPA_VNIC_ICRC_TAIL_LEN -
- ((*pad_info) & 0x7)));
-
- return skb;
+ return hfi1_netdev_get_data(dd, vnic_id);
}
-/* hfi1_vnic_handle_rx - handle skb receive */
-static void hfi1_vnic_handle_rx(struct hfi1_vnic_rx_queue *rxq,
- int *work_done, int work_to_do)
+static struct hfi1_vnic_vport_info *get_first_vnic_port(struct hfi1_devdata *dd)
{
- struct hfi1_vnic_vport_info *vinfo = rxq->vinfo;
- struct sk_buff *skb;
- int rc;
-
- while (1) {
- if (*work_done >= work_to_do)
- break;
-
- skb = hfi1_vnic_get_skb(rxq);
- if (unlikely(!skb))
- break;
-
- rc = hfi1_vnic_decap_skb(rxq, skb);
- /* update rx counters */
- hfi1_vnic_update_rx_counters(vinfo, rxq->idx, skb, rc);
- if (unlikely(rc)) {
- dev_kfree_skb_any(skb);
- continue;
- }
-
- skb_checksum_none_assert(skb);
- skb->protocol = eth_type_trans(skb, rxq->netdev);
-
- napi_gro_receive(&rxq->napi, skb);
- (*work_done)++;
- }
-}
-
-/* hfi1_vnic_napi - napi receive polling callback function */
-static int hfi1_vnic_napi(struct napi_struct *napi, int budget)
-{
- struct hfi1_vnic_rx_queue *rxq = container_of(napi,
- struct hfi1_vnic_rx_queue, napi);
- struct hfi1_vnic_vport_info *vinfo = rxq->vinfo;
- int work_done = 0;
+ struct hfi1_vnic_vport_info *vinfo;
+ int next_id = VNIC_ID(0);
- v_dbg("napi %d budget %d\n", rxq->idx, budget);
- hfi1_vnic_handle_rx(rxq, &work_done, budget);
+ vinfo = hfi1_netdev_get_first_data(dd, &next_id);
- v_dbg("napi %d work_done %d\n", rxq->idx, work_done);
- if (work_done < budget)
- napi_complete(napi);
+ if (next_id > VNIC_ID(VNIC_MASK))
+ return NULL;
- return work_done;
+ return vinfo;
}
void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet)
@@ -527,13 +377,14 @@ void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet)
struct hfi1_vnic_vport_info *vinfo = NULL;
struct hfi1_vnic_rx_queue *rxq;
struct sk_buff *skb;
- int l4_type, vesw_id = -1;
+ int l4_type, vesw_id = -1, rc;
u8 q_idx;
+ unsigned char *pad_info;
l4_type = hfi1_16B_get_l4(packet->ebuf);
if (likely(l4_type == OPA_16B_L4_ETHR)) {
vesw_id = HFI1_VNIC_GET_VESWID(packet->ebuf);
- vinfo = xa_load(&dd->vnic.vesws, vesw_id);
+ vinfo = get_vnic_port(dd, vesw_id);
/*
* In case of invalid vesw id, count the error on
@@ -541,10 +392,8 @@ void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet)
*/
if (unlikely(!vinfo)) {
struct hfi1_vnic_vport_info *vinfo_tmp;
- unsigned long index = 0;
- vinfo_tmp = xa_find(&dd->vnic.vesws, &index, ULONG_MAX,
- XA_PRESENT);
+ vinfo_tmp = get_first_vnic_port(dd);
if (vinfo_tmp) {
spin_lock(&vport_cntr_lock);
vinfo_tmp->stats[0].netstats.rx_nohandler++;
@@ -563,12 +412,6 @@ void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet)
rxq = &vinfo->rxq[q_idx];
if (unlikely(!netif_oper_up(vinfo->netdev))) {
vinfo->stats[q_idx].rx_drop_state++;
- skb_queue_purge(&rxq->skbq);
- return;
- }
-
- if (unlikely(skb_queue_len(&rxq->skbq) > HFI1_VNIC_RCV_Q_SIZE)) {
- vinfo->stats[q_idx].netstats.rx_fifo_errors++;
return;
}
@@ -580,62 +423,65 @@ void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet)
memcpy(skb->data, packet->ebuf, packet->tlen);
skb_put(skb, packet->tlen);
- skb_queue_tail(&rxq->skbq, skb);
- if (napi_schedule_prep(&rxq->napi)) {
- v_dbg("napi %d scheduling\n", q_idx);
- __napi_schedule(&rxq->napi);
+ pad_info = skb->data + skb->len - 1;
+ skb_trim(skb, (skb->len - OPA_VNIC_ICRC_TAIL_LEN -
+ ((*pad_info) & 0x7)));
+
+ rc = hfi1_vnic_decap_skb(rxq, skb);
+
+ /* update rx counters */
+ hfi1_vnic_update_rx_counters(vinfo, rxq->idx, skb, rc);
+ if (unlikely(rc)) {
+ dev_kfree_skb_any(skb);
+ return;
}
+
+ skb_checksum_none_assert(skb);
+ skb->protocol = eth_type_trans(skb, rxq->netdev);
+
+ napi_gro_receive(&rxq->napi, skb);
}
static int hfi1_vnic_up(struct hfi1_vnic_vport_info *vinfo)
{
struct hfi1_devdata *dd = vinfo->dd;
struct net_device *netdev = vinfo->netdev;
- int i, rc;
+ int rc;
/* ensure virtual eth switch id is valid */
if (!vinfo->vesw_id)
return -EINVAL;
- rc = xa_insert(&dd->vnic.vesws, vinfo->vesw_id, vinfo, GFP_KERNEL);
+ rc = hfi1_netdev_add_data(dd, VNIC_ID(vinfo->vesw_id), vinfo);
if (rc < 0)
return rc;
- for (i = 0; i < vinfo->num_rx_q; i++) {
- struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i];
-
- skb_queue_head_init(&rxq->skbq);
- napi_enable(&rxq->napi);
- }
+ rc = hfi1_netdev_rx_init(dd);
+ if (rc)
+ goto err_remove;
netif_carrier_on(netdev);
netif_tx_start_all_queues(netdev);
set_bit(HFI1_VNIC_UP, &vinfo->flags);
return 0;
+
+err_remove:
+ hfi1_netdev_remove_data(dd, VNIC_ID(vinfo->vesw_id));
+ return rc;
}
static void hfi1_vnic_down(struct hfi1_vnic_vport_info *vinfo)
{
struct hfi1_devdata *dd = vinfo->dd;
- u8 i;
clear_bit(HFI1_VNIC_UP, &vinfo->flags);
netif_carrier_off(vinfo->netdev);
netif_tx_disable(vinfo->netdev);
- xa_erase(&dd->vnic.vesws, vinfo->vesw_id);
-
- /* ensure irqs see the change */
- msix_vnic_synchronize_irq(dd);
+ hfi1_netdev_remove_data(dd, VNIC_ID(vinfo->vesw_id));
- /* remove unread skbs */
- for (i = 0; i < vinfo->num_rx_q; i++) {
- struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i];
-
- napi_disable(&rxq->napi);
- skb_queue_purge(&rxq->skbq);
- }
+ hfi1_netdev_rx_destroy(dd);
}
static int hfi1_netdev_open(struct net_device *netdev)
@@ -660,70 +506,31 @@ static int hfi1_netdev_close(struct net_device *netdev)
return 0;
}
-static int hfi1_vnic_allot_ctxt(struct hfi1_devdata *dd,
- struct hfi1_ctxtdata **vnic_ctxt)
-{
- int rc;
-
- rc = allocate_vnic_ctxt(dd, vnic_ctxt);
- if (rc) {
- dd_dev_err(dd, "vnic ctxt alloc failed %d\n", rc);
- return rc;
- }
-
- rc = setup_vnic_ctxt(dd, *vnic_ctxt);
- if (rc) {
- dd_dev_err(dd, "vnic ctxt setup failed %d\n", rc);
- deallocate_vnic_ctxt(dd, *vnic_ctxt);
- *vnic_ctxt = NULL;
- }
-
- return rc;
-}
-
static int hfi1_vnic_init(struct hfi1_vnic_vport_info *vinfo)
{
struct hfi1_devdata *dd = vinfo->dd;
- int i, rc = 0;
+ int rc = 0;
mutex_lock(&hfi1_mutex);
- if (!dd->vnic.num_vports) {
+ if (!dd->vnic_num_vports) {
rc = hfi1_vnic_txreq_init(dd);
if (rc)
goto txreq_fail;
}
- for (i = dd->vnic.num_ctxt; i < vinfo->num_rx_q; i++) {
- rc = hfi1_vnic_allot_ctxt(dd, &dd->vnic.ctxt[i]);
- if (rc)
- break;
- hfi1_rcd_get(dd->vnic.ctxt[i]);
- dd->vnic.ctxt[i]->vnic_q_idx = i;
- }
-
- if (i < vinfo->num_rx_q) {
- /*
- * If required amount of contexts is not
- * allocated successfully then remaining contexts
- * are released.
- */
- while (i-- > dd->vnic.num_ctxt) {
- deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]);
- hfi1_rcd_put(dd->vnic.ctxt[i]);
- dd->vnic.ctxt[i] = NULL;
- }
+ rc = hfi1_netdev_rx_init(dd);
+ if (rc) {
+ dd_dev_err(dd, "Unable to initialize netdev contexts\n");
goto alloc_fail;
}
- if (dd->vnic.num_ctxt != i) {
- dd->vnic.num_ctxt = i;
- hfi1_init_vnic_rsm(dd);
- }
+ hfi1_init_vnic_rsm(dd);
- dd->vnic.num_vports++;
+ dd->vnic_num_vports++;
hfi1_vnic_sdma_init(vinfo);
+
alloc_fail:
- if (!dd->vnic.num_vports)
+ if (!dd->vnic_num_vports)
hfi1_vnic_txreq_deinit(dd);
txreq_fail:
mutex_unlock(&hfi1_mutex);
@@ -733,20 +540,14 @@ txreq_fail:
static void hfi1_vnic_deinit(struct hfi1_vnic_vport_info *vinfo)
{
struct hfi1_devdata *dd = vinfo->dd;
- int i;
mutex_lock(&hfi1_mutex);
- if (--dd->vnic.num_vports == 0) {
- for (i = 0; i < dd->vnic.num_ctxt; i++) {
- deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]);
- hfi1_rcd_put(dd->vnic.ctxt[i]);
- dd->vnic.ctxt[i] = NULL;
- }
+ if (--dd->vnic_num_vports == 0) {
hfi1_deinit_vnic_rsm(dd);
- dd->vnic.num_ctxt = 0;
hfi1_vnic_txreq_deinit(dd);
}
mutex_unlock(&hfi1_mutex);
+ hfi1_netdev_rx_destroy(dd);
}
static void hfi1_vnic_set_vesw_id(struct net_device *netdev, int id)
@@ -804,7 +605,7 @@ struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device,
struct rdma_netdev *rn;
int i, size, rc;
- if (!dd->num_vnic_contexts)
+ if (!dd->num_netdev_contexts)
return ERR_PTR(-ENOMEM);
if (!port_num || (port_num > dd->num_pports))
@@ -815,15 +616,16 @@ struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device,
size = sizeof(struct opa_vnic_rdma_netdev) + sizeof(*vinfo);
netdev = alloc_netdev_mqs(size, name, name_assign_type, setup,
- dd->num_sdma, dd->num_vnic_contexts);
+ chip_sdma_engines(dd),
+ dd->num_netdev_contexts);
if (!netdev)
return ERR_PTR(-ENOMEM);
rn = netdev_priv(netdev);
vinfo = opa_vnic_dev_priv(netdev);
vinfo->dd = dd;
- vinfo->num_tx_q = dd->num_sdma;
- vinfo->num_rx_q = dd->num_vnic_contexts;
+ vinfo->num_tx_q = chip_sdma_engines(dd);
+ vinfo->num_rx_q = dd->num_netdev_contexts;
vinfo->netdev = netdev;
rn->free_rdma_netdev = hfi1_vnic_free_rn;
rn->set_id = hfi1_vnic_set_vesw_id;
@@ -841,7 +643,6 @@ struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device,
rxq->idx = i;
rxq->vinfo = vinfo;
rxq->netdev = netdev;
- netif_napi_add(netdev, &rxq->napi, hfi1_vnic_napi, 64);
}
rc = hfi1_vnic_init(vinfo);
diff --git a/drivers/infiniband/hw/hns/hns_roce_ah.c b/drivers/infiniband/hw/hns/hns_roce_ah.c
index 8a522e14ef62..5b2f9314edd3 100644
--- a/drivers/infiniband/hw/hns/hns_roce_ah.c
+++ b/drivers/infiniband/hw/hns/hns_roce_ah.c
@@ -39,13 +39,14 @@
#define HNS_ROCE_VLAN_SL_BIT_MASK 7
#define HNS_ROCE_VLAN_SL_SHIFT 13
-int hns_roce_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
+int hns_roce_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
struct hns_roce_dev *hr_dev = to_hr_dev(ibah->device);
const struct ib_gid_attr *gid_attr;
struct device *dev = hr_dev->dev;
struct hns_roce_ah *ah = to_hr_ah(ibah);
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
u16 vlan_id = 0xffff;
bool vlan_en = false;
diff --git a/drivers/infiniband/hw/hns/hns_roce_alloc.c b/drivers/infiniband/hw/hns/hns_roce_alloc.c
index da574c26e063..a522cb2d29ea 100644
--- a/drivers/infiniband/hw/hns/hns_roce_alloc.c
+++ b/drivers/infiniband/hw/hns/hns_roce_alloc.c
@@ -157,84 +157,78 @@ void hns_roce_bitmap_cleanup(struct hns_roce_bitmap *bitmap)
kfree(bitmap->table);
}
-void hns_roce_buf_free(struct hns_roce_dev *hr_dev, u32 size,
- struct hns_roce_buf *buf)
+void hns_roce_buf_free(struct hns_roce_dev *hr_dev, struct hns_roce_buf *buf)
{
- int i;
struct device *dev = hr_dev->dev;
+ u32 size = buf->size;
+ int i;
+
+ if (size == 0)
+ return;
- if (buf->nbufs == 1) {
+ buf->size = 0;
+
+ if (hns_roce_buf_is_direct(buf)) {
dma_free_coherent(dev, size, buf->direct.buf, buf->direct.map);
} else {
- for (i = 0; i < buf->nbufs; ++i)
+ for (i = 0; i < buf->npages; ++i)
if (buf->page_list[i].buf)
dma_free_coherent(dev, 1 << buf->page_shift,
buf->page_list[i].buf,
buf->page_list[i].map);
kfree(buf->page_list);
+ buf->page_list = NULL;
}
}
int hns_roce_buf_alloc(struct hns_roce_dev *hr_dev, u32 size, u32 max_direct,
struct hns_roce_buf *buf, u32 page_shift)
{
- int i = 0;
- dma_addr_t t;
+ struct hns_roce_buf_list *buf_list;
struct device *dev = hr_dev->dev;
- u32 page_size = 1 << page_shift;
- u32 order;
+ u32 page_size;
+ int i;
- /* SQ/RQ buf lease than one page, SQ + RQ = 8K */
+ /* The minimum shift of the page accessed by hw is HNS_HW_PAGE_SHIFT */
+ buf->page_shift = max_t(int, HNS_HW_PAGE_SHIFT, page_shift);
+
+ page_size = 1 << buf->page_shift;
+ buf->npages = DIV_ROUND_UP(size, page_size);
+
+ /* required size is not bigger than one trunk size */
if (size <= max_direct) {
- buf->nbufs = 1;
- /* Npages calculated by page_size */
- order = get_order(size);
- if (order <= page_shift - PAGE_SHIFT)
- order = 0;
- else
- order -= page_shift - PAGE_SHIFT;
- buf->npages = 1 << order;
- buf->page_shift = page_shift;
- /* MTT PA must be recorded in 4k alignment, t is 4k aligned */
- buf->direct.buf = dma_alloc_coherent(dev, size, &t,
+ buf->page_list = NULL;
+ buf->direct.buf = dma_alloc_coherent(dev, size,
+ &buf->direct.map,
GFP_KERNEL);
if (!buf->direct.buf)
return -ENOMEM;
-
- buf->direct.map = t;
-
- while (t & ((1 << buf->page_shift) - 1)) {
- --buf->page_shift;
- buf->npages *= 2;
- }
} else {
- buf->nbufs = (size + page_size - 1) / page_size;
- buf->npages = buf->nbufs;
- buf->page_shift = page_shift;
- buf->page_list = kcalloc(buf->nbufs, sizeof(*buf->page_list),
- GFP_KERNEL);
-
- if (!buf->page_list)
+ buf_list = kcalloc(buf->npages, sizeof(*buf_list), GFP_KERNEL);
+ if (!buf_list)
return -ENOMEM;
- for (i = 0; i < buf->nbufs; ++i) {
- buf->page_list[i].buf = dma_alloc_coherent(dev,
- page_size,
- &t,
- GFP_KERNEL);
-
- if (!buf->page_list[i].buf)
- goto err_free;
+ for (i = 0; i < buf->npages; i++) {
+ buf_list[i].buf = dma_alloc_coherent(dev, page_size,
+ &buf_list[i].map,
+ GFP_KERNEL);
+ if (!buf_list[i].buf)
+ break;
+ }
- buf->page_list[i].map = t;
+ if (i != buf->npages && i > 0) {
+ while (i-- > 0)
+ dma_free_coherent(dev, page_size,
+ buf_list[i].buf,
+ buf_list[i].map);
+ kfree(buf_list);
+ return -ENOMEM;
}
+ buf->page_list = buf_list;
}
+ buf->size = size;
return 0;
-
-err_free:
- hns_roce_buf_free(hr_dev, size, buf);
- return -ENOMEM;
}
int hns_roce_get_kmem_bufs(struct hns_roce_dev *hr_dev, dma_addr_t *bufs,
@@ -246,33 +240,30 @@ int hns_roce_get_kmem_bufs(struct hns_roce_dev *hr_dev, dma_addr_t *bufs,
end = start + buf_cnt;
if (end > buf->npages) {
dev_err(hr_dev->dev,
- "invalid kmem region,offset %d,buf_cnt %d,total %d!\n",
+ "Failed to check kmem bufs, end %d + %d total %d!\n",
start, buf_cnt, buf->npages);
return -EINVAL;
}
total = 0;
for (i = start; i < end; i++)
- if (buf->nbufs == 1)
- bufs[total++] = buf->direct.map +
- ((dma_addr_t)i << buf->page_shift);
- else
- bufs[total++] = buf->page_list[i].map;
+ bufs[total++] = hns_roce_buf_page(buf, i);
return total;
}
int hns_roce_get_umem_bufs(struct hns_roce_dev *hr_dev, dma_addr_t *bufs,
int buf_cnt, int start, struct ib_umem *umem,
- int page_shift)
+ unsigned int page_shift)
{
struct ib_block_iter biter;
int total = 0;
int idx = 0;
u64 addr;
- if (page_shift < PAGE_SHIFT) {
- dev_err(hr_dev->dev, "invalid page shift %d!\n", page_shift);
+ if (page_shift < HNS_HW_PAGE_SHIFT) {
+ dev_err(hr_dev->dev, "Failed to check umem page shift %d!\n",
+ page_shift);
return -EINVAL;
}
@@ -292,49 +283,6 @@ done:
return total;
}
-void hns_roce_init_buf_region(struct hns_roce_buf_region *region, int hopnum,
- int offset, int buf_cnt)
-{
- if (hopnum == HNS_ROCE_HOP_NUM_0)
- region->hopnum = 0;
- else
- region->hopnum = hopnum;
-
- region->offset = offset;
- region->count = buf_cnt;
-}
-
-void hns_roce_free_buf_list(dma_addr_t **bufs, int region_cnt)
-{
- int i;
-
- for (i = 0; i < region_cnt; i++) {
- kfree(bufs[i]);
- bufs[i] = NULL;
- }
-}
-
-int hns_roce_alloc_buf_list(struct hns_roce_buf_region *regions,
- dma_addr_t **bufs, int region_cnt)
-{
- struct hns_roce_buf_region *r;
- int i;
-
- for (i = 0; i < region_cnt; i++) {
- r = &regions[i];
- bufs[i] = kcalloc(r->count, sizeof(dma_addr_t), GFP_KERNEL);
- if (!bufs[i])
- goto err_alloc;
- }
-
- return 0;
-
-err_alloc:
- hns_roce_free_buf_list(bufs, i);
-
- return -ENOMEM;
-}
-
void hns_roce_cleanup_bitmap(struct hns_roce_dev *hr_dev)
{
if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_SRQ)
diff --git a/drivers/infiniband/hw/hns/hns_roce_common.h b/drivers/infiniband/hw/hns/hns_roce_common.h
index 8e95a1aa1b4f..f5669ff8cfeb 100644
--- a/drivers/infiniband/hw/hns/hns_roce_common.h
+++ b/drivers/infiniband/hw/hns/hns_roce_common.h
@@ -33,10 +33,6 @@
#ifndef _HNS_ROCE_COMMON_H
#define _HNS_ROCE_COMMON_H
-#ifndef assert
-#define assert(cond)
-#endif
-
#define roce_write(dev, reg, val) writel((val), (dev)->reg_base + (reg))
#define roce_read(dev, reg) readl((dev)->reg_base + (reg))
#define roce_raw_write(value, addr) \
diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c
index 5bfb52ffd590..e87d616f7988 100644
--- a/drivers/infiniband/hw/hns/hns_roce_cq.c
+++ b/drivers/infiniband/hw/hns/hns_roce_cq.c
@@ -39,51 +39,40 @@
#include <rdma/hns-abi.h>
#include "hns_roce_common.h"
-static int hns_roce_alloc_cqc(struct hns_roce_dev *hr_dev,
- struct hns_roce_cq *hr_cq)
+static int alloc_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
{
struct hns_roce_cmd_mailbox *mailbox;
- struct hns_roce_hem_table *mtt_table;
struct hns_roce_cq_table *cq_table;
- struct device *dev = hr_dev->dev;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ u64 mtts[MTT_MIN_COUNT] = { 0 };
dma_addr_t dma_handle;
- u64 *mtts;
int ret;
- cq_table = &hr_dev->cq_table;
-
- /* Get the physical address of cq buf */
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- mtt_table = &hr_dev->mr_table.mtt_cqe_table;
- else
- mtt_table = &hr_dev->mr_table.mtt_table;
-
- mtts = hns_roce_table_find(hr_dev, mtt_table, hr_cq->mtt.first_seg,
- &dma_handle);
-
- if (!mtts) {
- dev_err(dev, "Failed to find mtt for CQ buf.\n");
+ ret = hns_roce_mtr_find(hr_dev, &hr_cq->mtr, 0, mtts, ARRAY_SIZE(mtts),
+ &dma_handle);
+ if (ret < 1) {
+ ibdev_err(ibdev, "Failed to find CQ mtr\n");
return -EINVAL;
}
+ cq_table = &hr_dev->cq_table;
ret = hns_roce_bitmap_alloc(&cq_table->bitmap, &hr_cq->cqn);
if (ret) {
- dev_err(dev, "Num of CQ out of range.\n");
+ ibdev_err(ibdev, "Failed to alloc CQ bitmap, err %d\n", ret);
return ret;
}
/* Get CQC memory HEM(Hardware Entry Memory) table */
ret = hns_roce_table_get(hr_dev, &cq_table->table, hr_cq->cqn);
if (ret) {
- dev_err(dev,
- "Get context mem failed(%d) when CQ(0x%lx) alloc.\n",
- ret, hr_cq->cqn);
+ ibdev_err(ibdev, "Failed to get CQ(0x%lx) context, err %d\n",
+ hr_cq->cqn, ret);
goto err_out;
}
ret = xa_err(xa_store(&cq_table->array, hr_cq->cqn, hr_cq, GFP_KERNEL));
if (ret) {
- dev_err(dev, "Failed to xa_store CQ.\n");
+ ibdev_err(ibdev, "Failed to xa_store CQ\n");
goto err_put;
}
@@ -101,9 +90,9 @@ static int hns_roce_alloc_cqc(struct hns_roce_dev *hr_dev,
HNS_ROCE_CMD_CREATE_CQC, HNS_ROCE_CMD_TIMEOUT_MSECS);
hns_roce_free_cmd_mailbox(hr_dev, mailbox);
if (ret) {
- dev_err(dev,
- "Send cmd mailbox failed(%d) when CQ(0x%lx) alloc.\n",
- ret, hr_cq->cqn);
+ ibdev_err(ibdev,
+ "Failed to send create cmd for CQ(0x%lx), err %d\n",
+ hr_cq->cqn, ret);
goto err_xa;
}
@@ -126,7 +115,7 @@ err_out:
return ret;
}
-void hns_roce_free_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
+static void free_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
{
struct hns_roce_cq_table *cq_table = &hr_dev->cq_table;
struct device *dev = hr_dev->dev;
@@ -153,190 +142,86 @@ void hns_roce_free_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
hns_roce_bitmap_free(&cq_table->bitmap, hr_cq->cqn, BITMAP_NO_RR);
}
-static int get_cq_umem(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq,
- struct hns_roce_ib_create_cq ucmd,
- struct ib_udata *udata)
-{
- struct hns_roce_buf *buf = &hr_cq->buf;
- struct hns_roce_mtt *mtt = &hr_cq->mtt;
- struct ib_umem **umem = &hr_cq->umem;
- u32 npages;
- int ret;
-
- *umem = ib_umem_get(&hr_dev->ib_dev, ucmd.buf_addr, buf->size,
- IB_ACCESS_LOCAL_WRITE);
- if (IS_ERR(*umem))
- return PTR_ERR(*umem);
-
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- mtt->mtt_type = MTT_TYPE_CQE;
- else
- mtt->mtt_type = MTT_TYPE_WQE;
-
- npages = DIV_ROUND_UP(ib_umem_page_count(*umem),
- 1 << hr_dev->caps.cqe_buf_pg_sz);
- ret = hns_roce_mtt_init(hr_dev, npages, buf->page_shift, mtt);
- if (ret)
- goto err_buf;
-
- ret = hns_roce_ib_umem_write_mtt(hr_dev, mtt, *umem);
- if (ret)
- goto err_mtt;
-
- return 0;
-
-err_mtt:
- hns_roce_mtt_cleanup(hr_dev, mtt);
-
-err_buf:
- ib_umem_release(*umem);
- return ret;
-}
-
-static int alloc_cq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
+static int alloc_cq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq,
+ struct ib_udata *udata, unsigned long addr)
{
- struct hns_roce_buf *buf = &hr_cq->buf;
- struct hns_roce_mtt *mtt = &hr_cq->mtt;
- int ret;
-
- ret = hns_roce_buf_alloc(hr_dev, buf->size, (1 << buf->page_shift) * 2,
- buf, buf->page_shift);
- if (ret)
- goto out;
-
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- mtt->mtt_type = MTT_TYPE_CQE;
- else
- mtt->mtt_type = MTT_TYPE_WQE;
-
- ret = hns_roce_mtt_init(hr_dev, buf->npages, buf->page_shift, mtt);
- if (ret)
- goto err_buf;
-
- ret = hns_roce_buf_write_mtt(hr_dev, mtt, buf);
- if (ret)
- goto err_mtt;
-
- return 0;
-
-err_mtt:
- hns_roce_mtt_cleanup(hr_dev, mtt);
-
-err_buf:
- hns_roce_buf_free(hr_dev, buf->size, buf);
-
-out:
- return ret;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_buf_attr buf_attr = {};
+ int err;
+
+ buf_attr.page_shift = hr_dev->caps.cqe_buf_pg_sz + HNS_HW_PAGE_SHIFT;
+ buf_attr.region[0].size = hr_cq->cq_depth * hr_dev->caps.cq_entry_sz;
+ buf_attr.region[0].hopnum = hr_dev->caps.cqe_hop_num;
+ buf_attr.region_count = 1;
+ buf_attr.fixed_page = true;
+
+ err = hns_roce_mtr_create(hr_dev, &hr_cq->mtr, &buf_attr,
+ hr_dev->caps.cqe_ba_pg_sz + HNS_HW_PAGE_SHIFT,
+ udata, addr);
+ if (err)
+ ibdev_err(ibdev, "Failed to alloc CQ mtr, err %d\n", err);
+
+ return err;
}
static void free_cq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq)
{
- hns_roce_buf_free(hr_dev, hr_cq->buf.size, &hr_cq->buf);
+ hns_roce_mtr_destroy(hr_dev, &hr_cq->mtr);
}
-static int create_user_cq(struct hns_roce_dev *hr_dev,
- struct hns_roce_cq *hr_cq,
- struct ib_udata *udata,
- struct hns_roce_ib_create_cq_resp *resp)
+static int alloc_cq_db(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq,
+ struct ib_udata *udata, unsigned long addr,
+ struct hns_roce_ib_create_cq_resp *resp)
{
- struct hns_roce_ib_create_cq ucmd;
- struct device *dev = hr_dev->dev;
- int ret;
- struct hns_roce_ucontext *context = rdma_udata_to_drv_context(
- udata, struct hns_roce_ucontext, ibucontext);
-
- if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) {
- dev_err(dev, "Failed to copy_from_udata.\n");
- return -EFAULT;
- }
+ bool has_db = hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB;
+ struct hns_roce_ucontext *uctx;
+ int err;
- /* Get user space address, write it into mtt table */
- ret = get_cq_umem(hr_dev, hr_cq, ucmd, udata);
- if (ret) {
- dev_err(dev, "Failed to get_cq_umem.\n");
- return ret;
- }
-
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB &&
- udata->outlen >= offsetofend(typeof(*resp), cap_flags)) {
- ret = hns_roce_db_map_user(context, udata, ucmd.db_addr,
- &hr_cq->db);
- if (ret) {
- dev_err(dev, "cq record doorbell map failed!\n");
- goto err_mtt;
+ if (udata) {
+ if (has_db &&
+ udata->outlen >= offsetofend(typeof(*resp), cap_flags)) {
+ uctx = rdma_udata_to_drv_context(udata,
+ struct hns_roce_ucontext, ibucontext);
+ err = hns_roce_db_map_user(uctx, udata, addr,
+ &hr_cq->db);
+ if (err)
+ return err;
+ hr_cq->flags |= HNS_ROCE_CQ_FLAG_RECORD_DB;
+ resp->cap_flags |= HNS_ROCE_CQ_FLAG_RECORD_DB;
}
- hr_cq->db_en = 1;
- resp->cap_flags |= HNS_ROCE_SUPPORT_CQ_RECORD_DB;
- }
-
- return 0;
-
-err_mtt:
- hns_roce_mtt_cleanup(hr_dev, &hr_cq->mtt);
- ib_umem_release(hr_cq->umem);
-
- return ret;
-}
-
-static int create_kernel_cq(struct hns_roce_dev *hr_dev,
- struct hns_roce_cq *hr_cq)
-{
- struct device *dev = hr_dev->dev;
- int ret;
-
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB) {
- ret = hns_roce_alloc_db(hr_dev, &hr_cq->db, 1);
- if (ret)
- return ret;
-
- hr_cq->set_ci_db = hr_cq->db.db_record;
- *hr_cq->set_ci_db = 0;
- hr_cq->db_en = 1;
- }
-
- /* Init mtt table and write buff address to mtt table */
- ret = alloc_cq_buf(hr_dev, hr_cq);
- if (ret) {
- dev_err(dev, "Failed to alloc_cq_buf.\n");
- goto err_db;
+ } else {
+ if (has_db) {
+ err = hns_roce_alloc_db(hr_dev, &hr_cq->db, 1);
+ if (err)
+ return err;
+ hr_cq->set_ci_db = hr_cq->db.db_record;
+ *hr_cq->set_ci_db = 0;
+ hr_cq->flags |= HNS_ROCE_CQ_FLAG_RECORD_DB;
+ }
+ hr_cq->cq_db_l = hr_dev->reg_base + hr_dev->odb_offset +
+ DB_REG_OFFSET * hr_dev->priv_uar.index;
}
- hr_cq->cq_db_l = hr_dev->reg_base + hr_dev->odb_offset +
- DB_REG_OFFSET * hr_dev->priv_uar.index;
-
return 0;
-
-err_db:
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB)
- hns_roce_free_db(hr_dev, &hr_cq->db);
-
- return ret;
}
-static void destroy_user_cq(struct hns_roce_dev *hr_dev,
- struct hns_roce_cq *hr_cq,
- struct ib_udata *udata,
- struct hns_roce_ib_create_cq_resp *resp)
+static void free_cq_db(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq,
+ struct ib_udata *udata)
{
- struct hns_roce_ucontext *context = rdma_udata_to_drv_context(
- udata, struct hns_roce_ucontext, ibucontext);
+ struct hns_roce_ucontext *uctx;
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB &&
- udata->outlen >= offsetofend(typeof(*resp), cap_flags))
- hns_roce_db_unmap_user(context, &hr_cq->db);
-
- hns_roce_mtt_cleanup(hr_dev, &hr_cq->mtt);
- ib_umem_release(hr_cq->umem);
-}
-
-static void destroy_kernel_cq(struct hns_roce_dev *hr_dev,
- struct hns_roce_cq *hr_cq)
-{
- hns_roce_mtt_cleanup(hr_dev, &hr_cq->mtt);
- free_cq_buf(hr_dev, hr_cq);
+ if (!(hr_cq->flags & HNS_ROCE_CQ_FLAG_RECORD_DB))
+ return;
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB)
+ hr_cq->flags &= ~HNS_ROCE_CQ_FLAG_RECORD_DB;
+ if (udata) {
+ uctx = rdma_udata_to_drv_context(udata,
+ struct hns_roce_ucontext,
+ ibucontext);
+ hns_roce_db_unmap_user(uctx, &hr_cq->db);
+ } else {
hns_roce_free_db(hr_dev, &hr_cq->db);
+ }
}
int hns_roce_create_cq(struct ib_cq *ib_cq, const struct ib_cq_init_attr *attr,
@@ -345,20 +230,21 @@ int hns_roce_create_cq(struct ib_cq *ib_cq, const struct ib_cq_init_attr *attr,
struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device);
struct hns_roce_ib_create_cq_resp resp = {};
struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq);
- struct device *dev = hr_dev->dev;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_ib_create_cq ucmd = {};
int vector = attr->comp_vector;
u32 cq_entries = attr->cqe;
int ret;
if (cq_entries < 1 || cq_entries > hr_dev->caps.max_cqes) {
- dev_err(dev, "Create CQ failed. entries=%d, max=%d\n",
- cq_entries, hr_dev->caps.max_cqes);
+ ibdev_err(ibdev, "Failed to check CQ count %d max=%d\n",
+ cq_entries, hr_dev->caps.max_cqes);
return -EINVAL;
}
if (vector >= hr_dev->caps.num_comp_vectors) {
- dev_err(dev, "Create CQ failed, vector=%d, max=%d\n",
- vector, hr_dev->caps.num_comp_vectors);
+ ibdev_err(ibdev, "Failed to check CQ vector=%d max=%d\n",
+ vector, hr_dev->caps.num_comp_vectors);
return -EINVAL;
}
@@ -367,30 +253,35 @@ int hns_roce_create_cq(struct ib_cq *ib_cq, const struct ib_cq_init_attr *attr,
hr_cq->ib_cq.cqe = cq_entries - 1; /* used as cqe index */
hr_cq->cq_depth = cq_entries;
hr_cq->vector = vector;
- hr_cq->buf.size = hr_cq->cq_depth * hr_dev->caps.cq_entry_sz;
- hr_cq->buf.page_shift = PAGE_SHIFT + hr_dev->caps.cqe_buf_pg_sz;
spin_lock_init(&hr_cq->lock);
INIT_LIST_HEAD(&hr_cq->sq_list);
INIT_LIST_HEAD(&hr_cq->rq_list);
if (udata) {
- ret = create_user_cq(hr_dev, hr_cq, udata, &resp);
- if (ret) {
- dev_err(dev, "Create cq failed in user mode!\n");
- goto err_cq;
- }
- } else {
- ret = create_kernel_cq(hr_dev, hr_cq);
+ ret = ib_copy_from_udata(&ucmd, udata, sizeof(ucmd));
if (ret) {
- dev_err(dev, "Create cq failed in kernel mode!\n");
- goto err_cq;
+ ibdev_err(ibdev, "Failed to copy CQ udata, err %d\n",
+ ret);
+ return ret;
}
}
- ret = hns_roce_alloc_cqc(hr_dev, hr_cq);
+ ret = alloc_cq_buf(hr_dev, hr_cq, udata, ucmd.buf_addr);
+ if (ret) {
+ ibdev_err(ibdev, "Failed to alloc CQ buf, err %d\n", ret);
+ return ret;
+ }
+
+ ret = alloc_cq_db(hr_dev, hr_cq, udata, ucmd.db_addr, &resp);
if (ret) {
- dev_err(dev, "Alloc CQ failed(%d).\n", ret);
- goto err_dbmap;
+ ibdev_err(ibdev, "Failed to alloc CQ db, err %d\n", ret);
+ goto err_cq_buf;
+ }
+
+ ret = alloc_cqc(hr_dev, hr_cq);
+ if (ret) {
+ ibdev_err(ibdev, "Failed to alloc CQ context, err %d\n", ret);
+ goto err_cq_db;
}
/*
@@ -412,15 +303,11 @@ int hns_roce_create_cq(struct ib_cq *ib_cq, const struct ib_cq_init_attr *attr,
return 0;
err_cqc:
- hns_roce_free_cqc(hr_dev, hr_cq);
-
-err_dbmap:
- if (udata)
- destroy_user_cq(hr_dev, hr_cq, udata, &resp);
- else
- destroy_kernel_cq(hr_dev, hr_cq);
-
-err_cq:
+ free_cqc(hr_dev, hr_cq);
+err_cq_db:
+ free_cq_db(hr_dev, hr_cq, udata);
+err_cq_buf:
+ free_cq_buf(hr_dev, hr_cq);
return ret;
}
@@ -429,28 +316,12 @@ void hns_roce_destroy_cq(struct ib_cq *ib_cq, struct ib_udata *udata)
struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device);
struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq);
- if (hr_dev->hw->destroy_cq) {
+ if (hr_dev->hw->destroy_cq)
hr_dev->hw->destroy_cq(ib_cq, udata);
- return;
- }
-
- hns_roce_free_cqc(hr_dev, hr_cq);
- hns_roce_mtt_cleanup(hr_dev, &hr_cq->mtt);
- ib_umem_release(hr_cq->umem);
- if (udata) {
- if (hr_cq->db_en == 1)
- hns_roce_db_unmap_user(rdma_udata_to_drv_context(
- udata,
- struct hns_roce_ucontext,
- ibucontext),
- &hr_cq->db);
- } else {
- /* Free the buff of stored cq */
- free_cq_buf(hr_dev, hr_cq);
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB)
- hns_roce_free_db(hr_dev, &hr_cq->db);
- }
+ free_cq_buf(hr_dev, hr_cq);
+ free_cq_db(hr_dev, hr_cq, udata);
+ free_cqc(hr_dev, hr_cq);
}
void hns_roce_cq_completion(struct hns_roce_dev *hr_dev, u32 cqn)
diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h
index f6b3cf6b95d6..a77fa6730b2d 100644
--- a/drivers/infiniband/hw/hns/hns_roce_device.h
+++ b/drivers/infiniband/hw/hns/hns_roce_device.h
@@ -66,6 +66,8 @@
#define HNS_ROCE_CQE_WCMD_EMPTY_BIT 0x2
#define HNS_ROCE_MIN_CQE_CNT 16
+#define HNS_ROCE_RESERVED_SGE 1
+
#define HNS_ROCE_MAX_IRQ_NUM 128
#define HNS_ROCE_SGE_IN_WQE 2
@@ -131,12 +133,12 @@ enum {
};
enum {
- HNS_ROCE_SUPPORT_RQ_RECORD_DB = 1 << 0,
- HNS_ROCE_SUPPORT_SQ_RECORD_DB = 1 << 1,
+ HNS_ROCE_QP_CAP_RQ_RECORD_DB = BIT(0),
+ HNS_ROCE_QP_CAP_SQ_RECORD_DB = BIT(1),
};
-enum {
- HNS_ROCE_SUPPORT_CQ_RECORD_DB = 1 << 0,
+enum hns_roce_cq_flags {
+ HNS_ROCE_CQ_FLAG_RECORD_DB = BIT(0),
};
enum hns_roce_qp_state {
@@ -209,6 +211,8 @@ enum {
HNS_ROCE_OPCODE_RDMA_WITH_IMM_RECEIVE = 0x07,
};
+#define HNS_ROCE_CAP_FLAGS_EX_SHIFT 12
+
enum {
HNS_ROCE_CAP_FLAG_REREG_MR = BIT(0),
HNS_ROCE_CAP_FLAG_ROCE_V1_V2 = BIT(1),
@@ -222,13 +226,6 @@ enum {
HNS_ROCE_CAP_FLAG_ATOMIC = BIT(10),
};
-enum hns_roce_mtt_type {
- MTT_TYPE_WQE,
- MTT_TYPE_CQE,
- MTT_TYPE_SRQWQE,
- MTT_TYPE_IDX
-};
-
#define HNS_ROCE_DB_TYPE_COUNT 2
#define HNS_ROCE_DB_UNIT_SIZE 4
@@ -267,9 +264,12 @@ enum {
#define HNS_ROCE_PORT_DOWN 0
#define HNS_ROCE_PORT_UP 1
-#define HNS_ROCE_MTT_ENTRY_PER_SEG 8
+/* The minimum page size is 4K for hardware */
+#define HNS_HW_PAGE_SHIFT 12
+#define HNS_HW_PAGE_SIZE (1 << HNS_HW_PAGE_SHIFT)
-#define PAGE_ADDR_SHIFT 12
+/* The minimum page count for hardware access page directly. */
+#define HNS_HW_DIRECT_PAGE_COUNT 2
struct hns_roce_uar {
u64 pfn;
@@ -300,22 +300,6 @@ struct hns_roce_bitmap {
unsigned long *table;
};
-/* Order bitmap length -- bit num compute formula: 1 << (max_order - order) */
-/* Order = 0: bitmap is biggest, order = max bitmap is least (only a bit) */
-/* Every bit repesent to a partner free/used status in bitmap */
-/*
- * Initial, bits of other bitmap are all 0 except that a bit of max_order is 1
- * Bit = 1 represent to idle and available; bit = 0: not available
- */
-struct hns_roce_buddy {
- /* Members point to every order level bitmap */
- unsigned long **bits;
- /* Represent to avail bits of the order level bitmap */
- u32 *num_free;
- int max_order;
- spinlock_t lock;
-};
-
/* For Hardware Entry Memory */
struct hns_roce_hem_table {
/* HEM type: 0 = qpc, 1 = mtt, 2 = cqc, 3 = srq, 4 = other */
@@ -336,13 +320,6 @@ struct hns_roce_hem_table {
dma_addr_t *bt_l0_dma_addr;
};
-struct hns_roce_mtt {
- unsigned long first_seg;
- int order;
- int page_shift;
- enum hns_roce_mtt_type mtt_type;
-};
-
struct hns_roce_buf_region {
int offset; /* page offset */
u32 count; /* page count */
@@ -357,13 +334,34 @@ struct hns_roce_hem_list {
struct list_head mid_bt[HNS_ROCE_MAX_BT_REGION][HNS_ROCE_MAX_BT_LEVEL];
struct list_head btm_bt; /* link all bottom bt in @mid_bt */
dma_addr_t root_ba; /* pointer to the root ba table */
- int bt_pg_shift;
+};
+
+struct hns_roce_buf_attr {
+ struct {
+ size_t size; /* region size */
+ int hopnum; /* multi-hop addressing hop num */
+ } region[HNS_ROCE_MAX_BT_REGION];
+ int region_count; /* valid region count */
+ unsigned int page_shift; /* buffer page shift */
+ bool fixed_page; /* decide page shift is fixed-size or maximum size */
+ int user_access; /* umem access flag */
+ bool mtt_only; /* only alloc buffer-required MTT memory */
};
/* memory translate region */
struct hns_roce_mtr {
- struct hns_roce_hem_list hem_list;
- int buf_pg_shift;
+ struct hns_roce_hem_list hem_list; /* multi-hop addressing resource */
+ struct ib_umem *umem; /* user space buffer */
+ struct hns_roce_buf *kmem; /* kernel space buffer */
+ struct {
+ dma_addr_t root_ba; /* root BA table's address */
+ bool is_direct; /* addressing without BA table */
+ unsigned int ba_pg_shift; /* BA table page shift */
+ unsigned int buf_pg_shift; /* buffer page shift */
+ int buf_pg_count; /* buffer page count */
+ struct hns_roce_buf_region region[HNS_ROCE_MAX_BT_REGION];
+ unsigned int region_count;
+ } hem_cfg; /* config for hardware addressing */
};
struct hns_roce_mw {
@@ -381,43 +379,22 @@ struct hns_roce_mw {
struct hns_roce_mr {
struct ib_mr ibmr;
- struct ib_umem *umem;
u64 iova; /* MR's virtual orignal addr */
u64 size; /* Address range of MR */
u32 key; /* Key of MR */
u32 pd; /* PD num of MR */
u32 access; /* Access permission of MR */
- u32 npages;
int enabled; /* MR's active status */
int type; /* MR's register type */
- u64 *pbl_buf; /* MR's PBL space */
- dma_addr_t pbl_dma_addr; /* MR's PBL space PA */
- u32 pbl_size; /* PA number in the PBL */
- u64 pbl_ba; /* page table address */
- u32 l0_chunk_last_num; /* L0 last number */
- u32 l1_chunk_last_num; /* L1 last number */
- u64 **pbl_bt_l2; /* PBL BT L2 */
- u64 **pbl_bt_l1; /* PBL BT L1 */
- u64 *pbl_bt_l0; /* PBL BT L0 */
- dma_addr_t *pbl_l2_dma_addr; /* PBL BT L2 dma addr */
- dma_addr_t *pbl_l1_dma_addr; /* PBL BT L1 dma addr */
- dma_addr_t pbl_l0_dma_addr; /* PBL BT L0 dma addr */
- u32 pbl_ba_pg_sz; /* BT chunk page size */
- u32 pbl_buf_pg_sz; /* buf chunk page size */
u32 pbl_hop_num; /* multi-hop number */
+ struct hns_roce_mtr pbl_mtr;
+ u32 npages;
+ dma_addr_t *page_list;
};
struct hns_roce_mr_table {
struct hns_roce_bitmap mtpt_bitmap;
- struct hns_roce_buddy mtt_buddy;
- struct hns_roce_hem_table mtt_table;
struct hns_roce_hem_table mtpt_table;
- struct hns_roce_buddy mtt_cqe_buddy;
- struct hns_roce_hem_table mtt_cqe_table;
- struct hns_roce_buddy mtt_srqwqe_buddy;
- struct hns_roce_hem_table mtt_srqwqe_table;
- struct hns_roce_buddy mtt_idx_buddy;
- struct hns_roce_hem_table mtt_idx_table;
};
struct hns_roce_wq {
@@ -433,7 +410,7 @@ struct hns_roce_wq {
};
struct hns_roce_sge {
- int sge_cnt; /* SGE num */
+ unsigned int sge_cnt; /* SGE num */
int offset;
int sge_shift; /* SGE size */
};
@@ -446,10 +423,9 @@ struct hns_roce_buf_list {
struct hns_roce_buf {
struct hns_roce_buf_list direct;
struct hns_roce_buf_list *page_list;
- int nbufs;
u32 npages;
u32 size;
- int page_shift;
+ unsigned int page_shift;
};
struct hns_roce_db_pgdir {
@@ -482,12 +458,10 @@ struct hns_roce_db {
struct hns_roce_cq {
struct ib_cq ib_cq;
- struct hns_roce_buf buf;
- struct hns_roce_mtt mtt;
+ struct hns_roce_mtr mtr;
struct hns_roce_db db;
- u8 db_en;
+ u32 flags;
spinlock_t lock;
- struct ib_umem *umem;
u32 cq_depth;
u32 cons_index;
u32 *set_ci_db;
@@ -505,11 +479,8 @@ struct hns_roce_cq {
};
struct hns_roce_idx_que {
- struct hns_roce_buf idx_buf;
- int entry_sz;
- u32 buf_size;
- struct ib_umem *umem;
- struct hns_roce_mtt mtt;
+ struct hns_roce_mtr mtr;
+ int entry_shift;
unsigned long *bitmap;
};
@@ -524,10 +495,9 @@ struct hns_roce_srq {
atomic_t refcount;
struct completion free;
- struct hns_roce_buf buf;
+ struct hns_roce_mtr buf_mtr;
+
u64 *wrid;
- struct ib_umem *umem;
- struct hns_roce_mtt mtt;
struct hns_roce_idx_que idx_que;
spinlock_t lock;
int head;
@@ -656,20 +626,15 @@ struct hns_roce_work {
struct hns_roce_qp {
struct ib_qp ibqp;
- struct hns_roce_buf hr_buf;
struct hns_roce_wq rq;
struct hns_roce_db rdb;
struct hns_roce_db sdb;
- u8 rdb_en;
- u8 sdb_en;
+ unsigned long en_flags;
u32 doorbell_qpn;
u32 sq_signal_bits;
struct hns_roce_wq sq;
- struct ib_umem *umem;
- struct hns_roce_mtt mtt;
struct hns_roce_mtr mtr;
- int wqe_bt_pg_shift;
u32 buff_size;
struct mutex mutex;
@@ -769,17 +734,11 @@ struct hns_roce_eq {
int over_ignore;
int coalesce;
int arm_st;
- u64 eqe_ba;
- int eqe_ba_pg_sz;
- int eqe_buf_pg_sz;
int hop_num;
struct hns_roce_mtr mtr;
- struct hns_roce_buf buf;
- int eq_max_cnt;
+ u16 eq_max_cnt;
int eq_period;
int shift;
- dma_addr_t cur_eqe_ba;
- dma_addr_t nxt_eqe_ba;
int event_type;
int sub_type;
};
@@ -1102,15 +1061,67 @@ static inline struct hns_roce_qp
return xa_load(&hr_dev->qp_table_xa, qpn & (hr_dev->caps.num_qps - 1));
}
+static inline bool hns_roce_buf_is_direct(struct hns_roce_buf *buf)
+{
+ if (buf->page_list)
+ return false;
+
+ return true;
+}
+
static inline void *hns_roce_buf_offset(struct hns_roce_buf *buf, int offset)
{
- u32 page_size = 1 << buf->page_shift;
+ if (hns_roce_buf_is_direct(buf))
+ return (char *)(buf->direct.buf) + (offset & (buf->size - 1));
- if (buf->nbufs == 1)
- return (char *)(buf->direct.buf) + offset;
+ return (char *)(buf->page_list[offset >> buf->page_shift].buf) +
+ (offset & ((1 << buf->page_shift) - 1));
+}
+
+static inline dma_addr_t hns_roce_buf_page(struct hns_roce_buf *buf, int idx)
+{
+ if (hns_roce_buf_is_direct(buf))
+ return buf->direct.map + ((dma_addr_t)idx << buf->page_shift);
else
- return (char *)(buf->page_list[offset >> buf->page_shift].buf) +
- (offset & (page_size - 1));
+ return buf->page_list[idx].map;
+}
+
+#define hr_hw_page_align(x) ALIGN(x, 1 << HNS_HW_PAGE_SHIFT)
+
+static inline u64 to_hr_hw_page_addr(u64 addr)
+{
+ return addr >> HNS_HW_PAGE_SHIFT;
+}
+
+static inline u32 to_hr_hw_page_shift(u32 page_shift)
+{
+ return page_shift - HNS_HW_PAGE_SHIFT;
+}
+
+static inline u32 to_hr_hem_hopnum(u32 hopnum, u32 count)
+{
+ if (count > 0)
+ return hopnum == HNS_ROCE_HOP_NUM_0 ? 0 : hopnum;
+
+ return 0;
+}
+
+static inline u32 to_hr_hem_entries_size(u32 count, u32 buf_shift)
+{
+ return hr_hw_page_align(count << buf_shift);
+}
+
+static inline u32 to_hr_hem_entries_count(u32 count, u32 buf_shift)
+{
+ return hr_hw_page_align(count << buf_shift) >> buf_shift;
+}
+
+static inline u32 to_hr_hem_entries_shift(u32 count, u32 buf_shift)
+{
+ if (!count)
+ return 0;
+
+ return ilog2(to_hr_hem_entries_count(count, buf_shift));
}
int hns_roce_init_uar_table(struct hns_roce_dev *dev);
@@ -1125,25 +1136,18 @@ void hns_roce_cmd_event(struct hns_roce_dev *hr_dev, u16 token, u8 status,
int hns_roce_cmd_use_events(struct hns_roce_dev *hr_dev);
void hns_roce_cmd_use_polling(struct hns_roce_dev *hr_dev);
-int hns_roce_mtt_init(struct hns_roce_dev *hr_dev, int npages, int page_shift,
- struct hns_roce_mtt *mtt);
-void hns_roce_mtt_cleanup(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt);
-int hns_roce_buf_write_mtt(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, struct hns_roce_buf *buf);
-
-void hns_roce_mtr_init(struct hns_roce_mtr *mtr, int bt_pg_shift,
- int buf_pg_shift);
-int hns_roce_mtr_attach(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
- dma_addr_t **bufs, struct hns_roce_buf_region *regions,
- int region_cnt);
-void hns_roce_mtr_cleanup(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtr *mtr);
-
/* hns roce hw need current block and next block addr from mtt */
#define MTT_MIN_COUNT 2
int hns_roce_mtr_find(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
int offset, u64 *mtt_buf, int mtt_max, u64 *base_addr);
+int hns_roce_mtr_create(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ struct hns_roce_buf_attr *buf_attr,
+ unsigned int page_shift, struct ib_udata *udata,
+ unsigned long user_addr);
+void hns_roce_mtr_destroy(struct hns_roce_dev *hr_dev,
+ struct hns_roce_mtr *mtr);
+int hns_roce_mtr_map(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ dma_addr_t *pages, int page_cnt);
int hns_roce_init_pd_table(struct hns_roce_dev *hr_dev);
int hns_roce_init_mr_table(struct hns_roce_dev *hr_dev);
@@ -1171,8 +1175,8 @@ void hns_roce_bitmap_free_range(struct hns_roce_bitmap *bitmap,
unsigned long obj, int cnt,
int rr);
-int hns_roce_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata);
+int hns_roce_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata);
int hns_roce_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr);
void hns_roce_destroy_ah(struct ib_ah *ah, u32 flags);
@@ -1200,25 +1204,15 @@ struct ib_mw *hns_roce_alloc_mw(struct ib_pd *pd, enum ib_mw_type,
struct ib_udata *udata);
int hns_roce_dealloc_mw(struct ib_mw *ibmw);
-void hns_roce_buf_free(struct hns_roce_dev *hr_dev, u32 size,
- struct hns_roce_buf *buf);
+void hns_roce_buf_free(struct hns_roce_dev *hr_dev, struct hns_roce_buf *buf);
int hns_roce_buf_alloc(struct hns_roce_dev *hr_dev, u32 size, u32 max_direct,
struct hns_roce_buf *buf, u32 page_shift);
-int hns_roce_ib_umem_write_mtt(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, struct ib_umem *umem);
-
-void hns_roce_init_buf_region(struct hns_roce_buf_region *region, int hopnum,
- int offset, int buf_cnt);
-int hns_roce_alloc_buf_list(struct hns_roce_buf_region *regions,
- dma_addr_t **bufs, int count);
-void hns_roce_free_buf_list(dma_addr_t **bufs, int count);
-
int hns_roce_get_kmem_bufs(struct hns_roce_dev *hr_dev, dma_addr_t *bufs,
int buf_cnt, int start, struct hns_roce_buf *buf);
int hns_roce_get_umem_bufs(struct hns_roce_dev *hr_dev, dma_addr_t *bufs,
int buf_cnt, int start, struct ib_umem *umem,
- int page_shift);
+ unsigned int page_shift);
int hns_roce_create_srq(struct ib_srq *srq,
struct ib_srq_init_attr *srq_init_attr,
@@ -1254,8 +1248,6 @@ int hns_roce_create_cq(struct ib_cq *ib_cq, const struct ib_cq_init_attr *attr,
struct ib_udata *udata);
void hns_roce_destroy_cq(struct ib_cq *ib_cq, struct ib_udata *udata);
-void hns_roce_free_cqc(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq);
-
int hns_roce_db_map_user(struct hns_roce_ucontext *context,
struct ib_udata *udata, unsigned long virt,
struct hns_roce_db *db);
diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c
index 263338b90d7a..c8db6f8ae018 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hem.c
+++ b/drivers/infiniband/hw/hns/hns_roce_hem.c
@@ -75,18 +75,6 @@ bool hns_roce_check_whether_mhop(struct hns_roce_dev *hr_dev, u32 type)
case HEM_TYPE_CQC_TIMER:
hop_num = hr_dev->caps.cqc_timer_hop_num;
break;
- case HEM_TYPE_CQE:
- hop_num = hr_dev->caps.cqe_hop_num;
- break;
- case HEM_TYPE_MTT:
- hop_num = hr_dev->caps.mtt_hop_num;
- break;
- case HEM_TYPE_SRQWQE:
- hop_num = hr_dev->caps.srqwqe_hop_num;
- break;
- case HEM_TYPE_IDX:
- hop_num = hr_dev->caps.idx_hop_num;
- break;
default:
return false;
}
@@ -195,38 +183,6 @@ static int get_hem_table_config(struct hns_roce_dev *hr_dev,
mhop->ba_l0_num = hr_dev->caps.srqc_bt_num;
mhop->hop_num = hr_dev->caps.srqc_hop_num;
break;
- case HEM_TYPE_MTT:
- mhop->buf_chunk_size = 1 << (hr_dev->caps.mtt_buf_pg_sz
- + PAGE_SHIFT);
- mhop->bt_chunk_size = 1 << (hr_dev->caps.mtt_ba_pg_sz
- + PAGE_SHIFT);
- mhop->ba_l0_num = mhop->bt_chunk_size / BA_BYTE_LEN;
- mhop->hop_num = hr_dev->caps.mtt_hop_num;
- break;
- case HEM_TYPE_CQE:
- mhop->buf_chunk_size = 1 << (hr_dev->caps.cqe_buf_pg_sz
- + PAGE_SHIFT);
- mhop->bt_chunk_size = 1 << (hr_dev->caps.cqe_ba_pg_sz
- + PAGE_SHIFT);
- mhop->ba_l0_num = mhop->bt_chunk_size / BA_BYTE_LEN;
- mhop->hop_num = hr_dev->caps.cqe_hop_num;
- break;
- case HEM_TYPE_SRQWQE:
- mhop->buf_chunk_size = 1 << (hr_dev->caps.srqwqe_buf_pg_sz
- + PAGE_SHIFT);
- mhop->bt_chunk_size = 1 << (hr_dev->caps.srqwqe_ba_pg_sz
- + PAGE_SHIFT);
- mhop->ba_l0_num = mhop->bt_chunk_size / BA_BYTE_LEN;
- mhop->hop_num = hr_dev->caps.srqwqe_hop_num;
- break;
- case HEM_TYPE_IDX:
- mhop->buf_chunk_size = 1 << (hr_dev->caps.idx_buf_pg_sz
- + PAGE_SHIFT);
- mhop->bt_chunk_size = 1 << (hr_dev->caps.idx_ba_pg_sz
- + PAGE_SHIFT);
- mhop->ba_l0_num = mhop->bt_chunk_size / BA_BYTE_LEN;
- mhop->hop_num = hr_dev->caps.idx_hop_num;
- break;
default:
dev_err(dev, "Table %d not support multi-hop addressing!\n",
type);
@@ -899,57 +855,6 @@ out:
return addr;
}
-int hns_roce_table_get_range(struct hns_roce_dev *hr_dev,
- struct hns_roce_hem_table *table,
- unsigned long start, unsigned long end)
-{
- struct hns_roce_hem_mhop mhop;
- unsigned long inc = table->table_chunk_size / table->obj_size;
- unsigned long i = 0;
- int ret;
-
- if (hns_roce_check_whether_mhop(hr_dev, table->type)) {
- ret = hns_roce_calc_hem_mhop(hr_dev, table, NULL, &mhop);
- if (ret)
- goto fail;
- inc = mhop.bt_chunk_size / table->obj_size;
- }
-
- /* Allocate MTT entry memory according to chunk(128K) */
- for (i = start; i <= end; i += inc) {
- ret = hns_roce_table_get(hr_dev, table, i);
- if (ret)
- goto fail;
- }
-
- return 0;
-
-fail:
- while (i > start) {
- i -= inc;
- hns_roce_table_put(hr_dev, table, i);
- }
- return ret;
-}
-
-void hns_roce_table_put_range(struct hns_roce_dev *hr_dev,
- struct hns_roce_hem_table *table,
- unsigned long start, unsigned long end)
-{
- struct hns_roce_hem_mhop mhop;
- unsigned long inc = table->table_chunk_size / table->obj_size;
- unsigned long i;
-
- if (hns_roce_check_whether_mhop(hr_dev, table->type)) {
- if (hns_roce_calc_hem_mhop(hr_dev, table, NULL, &mhop))
- return;
- inc = mhop.bt_chunk_size / table->obj_size;
- }
-
- for (i = start; i <= end; i += inc)
- hns_roce_table_put(hr_dev, table, i);
-}
-
int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_table *table, u32 type,
unsigned long obj_size, unsigned long nobj,
@@ -1112,12 +1017,6 @@ void hns_roce_cleanup_hem_table(struct hns_roce_dev *hr_dev,
void hns_roce_cleanup_hem(struct hns_roce_dev *hr_dev)
{
- if ((hr_dev->caps.num_idx_segs))
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_idx_table);
- if (hr_dev->caps.num_srqwqe_segs)
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_srqwqe_table);
if (hr_dev->caps.srqc_entry_sz)
hns_roce_cleanup_hem_table(hr_dev,
&hr_dev->srq_table.table);
@@ -1137,10 +1036,6 @@ void hns_roce_cleanup_hem(struct hns_roce_dev *hr_dev)
hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.irrl_table);
hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.qp_table);
hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtpt_table);
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_cqe_table);
- hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table);
}
struct roce_hem_item {
@@ -1505,7 +1400,7 @@ err_exit:
int hns_roce_hem_list_request(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_list *hem_list,
const struct hns_roce_buf_region *regions,
- int region_cnt)
+ int region_cnt, unsigned int bt_pg_shift)
{
const struct hns_roce_buf_region *r;
int ofs, end;
@@ -1519,7 +1414,7 @@ int hns_roce_hem_list_request(struct hns_roce_dev *hr_dev,
return -EINVAL;
}
- unit = (1 << hem_list->bt_pg_shift) / BA_BYTE_LEN;
+ unit = (1 << bt_pg_shift) / BA_BYTE_LEN;
for (i = 0; i < region_cnt; i++) {
r = &regions[i];
if (!r->count)
@@ -1566,8 +1461,7 @@ void hns_roce_hem_list_release(struct hns_roce_dev *hr_dev,
hem_list->root_ba = 0;
}
-void hns_roce_hem_list_init(struct hns_roce_hem_list *hem_list,
- int bt_page_order)
+void hns_roce_hem_list_init(struct hns_roce_hem_list *hem_list)
{
int i, j;
@@ -1576,8 +1470,6 @@ void hns_roce_hem_list_init(struct hns_roce_hem_list *hem_list,
for (i = 0; i < HNS_ROCE_MAX_BT_REGION; i++)
for (j = 0; j < HNS_ROCE_MAX_BT_LEVEL; j++)
INIT_LIST_HEAD(&hem_list->mid_bt[i][j]);
-
- hem_list->bt_pg_shift = bt_page_order;
}
void *hns_roce_hem_list_find_mtt(struct hns_roce_dev *hr_dev,
diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.h b/drivers/infiniband/hw/hns/hns_roce_hem.h
index 3bb8f78fb7b0..b34c940077bb 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hem.h
+++ b/drivers/infiniband/hw/hns/hns_roce_hem.h
@@ -115,12 +115,6 @@ void hns_roce_table_put(struct hns_roce_dev *hr_dev,
void *hns_roce_table_find(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_table *table, unsigned long obj,
dma_addr_t *dma_handle);
-int hns_roce_table_get_range(struct hns_roce_dev *hr_dev,
- struct hns_roce_hem_table *table,
- unsigned long start, unsigned long end);
-void hns_roce_table_put_range(struct hns_roce_dev *hr_dev,
- struct hns_roce_hem_table *table,
- unsigned long start, unsigned long end);
int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_table *table, u32 type,
unsigned long obj_size, unsigned long nobj,
@@ -133,14 +127,13 @@ int hns_roce_calc_hem_mhop(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_mhop *mhop);
bool hns_roce_check_whether_mhop(struct hns_roce_dev *hr_dev, u32 type);
-void hns_roce_hem_list_init(struct hns_roce_hem_list *hem_list,
- int bt_page_order);
+void hns_roce_hem_list_init(struct hns_roce_hem_list *hem_list);
int hns_roce_hem_list_calc_root_ba(const struct hns_roce_buf_region *regions,
int region_cnt, int unit);
int hns_roce_hem_list_request(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_list *hem_list,
const struct hns_roce_buf_region *regions,
- int region_cnt);
+ int region_cnt, unsigned int bt_pg_shift);
void hns_roce_hem_list_release(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_list *hem_list);
void *hns_roce_hem_list_find_mtt(struct hns_roce_dev *hr_dev,
diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c
index 5ff028d77be3..d02207cd30df 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c
+++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c
@@ -503,16 +503,13 @@ static void hns_roce_set_odb(struct hns_roce_dev *hr_dev, u32 odb_alept,
static void hns_roce_set_sdb_ext(struct hns_roce_dev *hr_dev, u32 ext_sdb_alept,
u32 ext_sdb_alful)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_db_table *db = &priv->db_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_db_table *db;
dma_addr_t sdb_dma_addr;
__le32 tmp;
u32 val;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- db = &priv->db_table;
-
/* Configure extend SDB threshold */
roce_write(hr_dev, ROCEE_EXT_DB_SQ_WL_EMPTY_REG, ext_sdb_alept);
roce_write(hr_dev, ROCEE_EXT_DB_SQ_WL_REG, ext_sdb_alful);
@@ -545,16 +542,13 @@ static void hns_roce_set_sdb_ext(struct hns_roce_dev *hr_dev, u32 ext_sdb_alept,
static void hns_roce_set_odb_ext(struct hns_roce_dev *hr_dev, u32 ext_odb_alept,
u32 ext_odb_alful)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_db_table *db = &priv->db_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_db_table *db;
dma_addr_t odb_dma_addr;
__le32 tmp;
u32 val;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- db = &priv->db_table;
-
/* Configure extend ODB threshold */
roce_write(hr_dev, ROCEE_EXT_DB_OTHERS_WL_EMPTY_REG, ext_odb_alept);
roce_write(hr_dev, ROCEE_EXT_DB_OTHERS_WL_REG, ext_odb_alful);
@@ -583,16 +577,13 @@ static void hns_roce_set_odb_ext(struct hns_roce_dev *hr_dev, u32 ext_odb_alept,
static int hns_roce_db_ext_init(struct hns_roce_dev *hr_dev, u32 sdb_ext_mod,
u32 odb_ext_mod)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_db_table *db = &priv->db_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_db_table *db;
dma_addr_t sdb_dma_addr;
dma_addr_t odb_dma_addr;
int ret = 0;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- db = &priv->db_table;
-
db->ext_db = kmalloc(sizeof(*db->ext_db), GFP_KERNEL);
if (!db->ext_db)
return -ENOMEM;
@@ -692,14 +683,14 @@ static struct hns_roce_qp *hns_roce_v1_create_lp_qp(struct hns_roce_dev *hr_dev,
static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
struct hns_roce_caps *caps = &hr_dev->caps;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct device *dev = &hr_dev->pdev->dev;
struct ib_cq_init_attr cq_init_attr;
- struct hns_roce_free_mr *free_mr;
struct ib_qp_attr attr = { 0 };
- struct hns_roce_v1_priv *priv;
struct hns_roce_qp *hr_qp;
- struct ib_device *ibdev;
struct ib_cq *cq;
struct ib_pd *pd;
union ib_gid dgid;
@@ -712,14 +703,10 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev)
u8 port = 0;
u8 sl;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
-
/* Reserved cq for loop qp */
cq_init_attr.cqe = HNS_ROCE_MIN_WQE_NUM * 2;
cq_init_attr.comp_vector = 0;
- ibdev = &hr_dev->ib_dev;
cq = rdma_zalloc_drv_obj(ibdev, ib_cq);
if (!cq)
return -ENOMEM;
@@ -868,16 +855,13 @@ alloc_cq_failed:
static void hns_roce_v1_release_lp_qp(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_v1_priv *priv;
struct hns_roce_qp *hr_qp;
int ret;
int i;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
-
for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) {
hr_qp = free_mr->mr_free_qp[i];
if (!hr_qp)
@@ -897,18 +881,15 @@ static void hns_roce_v1_release_lp_qp(struct hns_roce_dev *hr_dev)
static int hns_roce_db_init(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_db_table *db = &priv->db_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_db_table *db;
u32 sdb_ext_mod;
u32 odb_ext_mod;
u32 sdb_evt_mod;
u32 odb_evt_mod;
int ret = 0;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- db = &priv->db_table;
-
memset(db, 0, sizeof(*db));
/* Default DB mode */
@@ -954,15 +935,12 @@ static void hns_roce_v1_recreate_lp_qp_work_fn(struct work_struct *work)
static int hns_roce_v1_recreate_lp_qp(struct hns_roce_dev *hr_dev)
{
- struct device *dev = &hr_dev->pdev->dev;
+ long end = HNS_ROCE_V1_RECREATE_LP_QP_TIMEOUT_MSECS;
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
struct hns_roce_recreate_lp_qp_work *lp_qp_work;
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_v1_priv *priv;
+ struct device *dev = &hr_dev->pdev->dev;
struct completion comp;
- long end = HNS_ROCE_V1_RECREATE_LP_QP_TIMEOUT_MSECS;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
lp_qp_work = kzalloc(sizeof(struct hns_roce_recreate_lp_qp_work),
GFP_KERNEL);
@@ -1021,29 +999,21 @@ static int hns_roce_v1_send_lp_wqe(struct hns_roce_qp *hr_qp)
static void hns_roce_v1_mr_free_work_fn(struct work_struct *work)
{
- struct hns_roce_mr_free_work *mr_work;
- struct ib_wc wc[HNS_ROCE_V1_RESV_QP];
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_cq *mr_free_cq;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_dev *hr_dev;
- struct hns_roce_mr *hr_mr;
- struct hns_roce_qp *hr_qp;
- struct device *dev;
unsigned long end =
msecs_to_jiffies(HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS) + jiffies;
- int i;
- int ret;
+ struct hns_roce_mr_free_work *mr_work =
+ container_of(work, struct hns_roce_mr_free_work, work);
+ struct hns_roce_dev *hr_dev = to_hr_dev(mr_work->ib_dev);
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
+ struct hns_roce_cq *mr_free_cq = free_mr->mr_free_cq;
+ struct hns_roce_mr *hr_mr = mr_work->mr;
+ struct device *dev = &hr_dev->pdev->dev;
+ struct ib_wc wc[HNS_ROCE_V1_RESV_QP];
+ struct hns_roce_qp *hr_qp;
int ne = 0;
-
- mr_work = container_of(work, struct hns_roce_mr_free_work, work);
- hr_mr = (struct hns_roce_mr *)mr_work->mr;
- hr_dev = to_hr_dev(mr_work->ib_dev);
- dev = &hr_dev->pdev->dev;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
- mr_free_cq = free_mr->mr_free_cq;
+ int ret;
+ int i;
for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) {
hr_qp = free_mr->mr_free_qp[i];
@@ -1092,19 +1062,15 @@ free_work:
static int hns_roce_v1_dereg_mr(struct hns_roce_dev *hr_dev,
struct hns_roce_mr *mr, struct ib_udata *udata)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
+ long end = HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS;
struct device *dev = &hr_dev->pdev->dev;
struct hns_roce_mr_free_work *mr_work;
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_v1_priv *priv;
- struct completion comp;
- long end = HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS;
unsigned long start = jiffies;
- int npages;
+ struct completion comp;
int ret = 0;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
-
if (mr->enabled) {
if (hns_roce_hw_destroy_mpt(hr_dev, NULL,
key_to_hw_index(mr->key) &
@@ -1146,17 +1112,9 @@ free_mr:
dev_dbg(dev, "Free mr 0x%x use 0x%x us.\n",
mr->key, jiffies_to_usecs(jiffies) - jiffies_to_usecs(start));
- if (mr->size != ~0ULL) {
- npages = ib_umem_page_count(mr->umem);
- dma_free_coherent(dev, npages * 8, mr->pbl_buf,
- mr->pbl_dma_addr);
- }
-
hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap,
key_to_hw_index(mr->key), 0);
-
- ib_umem_release(mr->umem);
-
+ hns_roce_mtr_destroy(hr_dev, &mr->pbl_mtr);
kfree(mr);
return ret;
@@ -1164,12 +1122,9 @@ free_mr:
static void hns_roce_db_free(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_db_table *db = &priv->db_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_db_table *db;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- db = &priv->db_table;
if (db->sdb_ext_mod) {
dma_free_coherent(dev, HNS_ROCE_V1_EXT_SDB_SIZE,
@@ -1190,17 +1145,14 @@ static void hns_roce_db_free(struct hns_roce_dev *hr_dev)
static int hns_roce_raq_init(struct hns_roce_dev *hr_dev)
{
- int ret;
- u32 val;
- __le32 tmp;
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_raq_table *raq = &priv->raq_table;
+ struct device *dev = &hr_dev->pdev->dev;
int raq_shift = 0;
dma_addr_t addr;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_raq_table *raq;
- struct device *dev = &hr_dev->pdev->dev;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- raq = &priv->raq_table;
+ __le32 tmp;
+ u32 val;
+ int ret;
raq->e_raq_buf = kzalloc(sizeof(*(raq->e_raq_buf)), GFP_KERNEL);
if (!raq->e_raq_buf)
@@ -1280,12 +1232,9 @@ err_dma_alloc_raq:
static void hns_roce_raq_free(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_raq_table *raq = &priv->raq_table;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- struct hns_roce_raq_table *raq;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- raq = &priv->raq_table;
dma_free_coherent(dev, HNS_ROCE_V1_RAQ_SIZE, raq->e_raq_buf->buf,
raq->e_raq_buf->map);
@@ -1319,12 +1268,10 @@ static void hns_roce_port_enable(struct hns_roce_dev *hr_dev, int enable_flag)
static int hns_roce_bt_init(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
int ret;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
-
priv->bt_table.qpc_buf.buf = dma_alloc_coherent(dev,
HNS_ROCE_BT_RSV_BUF_SIZE, &priv->bt_table.qpc_buf.map,
GFP_KERNEL);
@@ -1362,10 +1309,8 @@ err_failed_alloc_mtpt_buf:
static void hns_roce_bt_free(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
dma_free_coherent(dev, HNS_ROCE_BT_RSV_BUF_SIZE,
priv->bt_table.cqc_buf.buf, priv->bt_table.cqc_buf.map);
@@ -1379,12 +1324,9 @@ static void hns_roce_bt_free(struct hns_roce_dev *hr_dev)
static int hns_roce_tptr_init(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_buf_list *tptr_buf = &priv->tptr_table.tptr_buf;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_buf_list *tptr_buf;
- struct hns_roce_v1_priv *priv;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- tptr_buf = &priv->tptr_table.tptr_buf;
/*
* This buffer will be used for CQ's tptr(tail pointer), also
@@ -1405,12 +1347,9 @@ static int hns_roce_tptr_init(struct hns_roce_dev *hr_dev)
static void hns_roce_tptr_free(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_buf_list *tptr_buf = &priv->tptr_table.tptr_buf;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_buf_list *tptr_buf;
- struct hns_roce_v1_priv *priv;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- tptr_buf = &priv->tptr_table.tptr_buf;
dma_free_coherent(dev, HNS_ROCE_V1_TPTR_BUF_SIZE,
tptr_buf->buf, tptr_buf->map);
@@ -1418,14 +1357,11 @@ static void hns_roce_tptr_free(struct hns_roce_dev *hr_dev)
static int hns_roce_free_mr_init(struct hns_roce_dev *hr_dev)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_v1_priv *priv;
int ret = 0;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
-
free_mr->free_mr_wq = create_singlethread_workqueue("hns_roce_free_mr");
if (!free_mr->free_mr_wq) {
dev_err(dev, "Create free mr workqueue failed!\n");
@@ -1444,11 +1380,8 @@ static int hns_roce_free_mr_init(struct hns_roce_dev *hr_dev)
static void hns_roce_free_mr_free(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_free_mr *free_mr;
- struct hns_roce_v1_priv *priv;
-
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- free_mr = &priv->free_mr;
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_free_mr *free_mr = &priv->free_mr;
flush_workqueue(free_mr->free_mr_wq);
destroy_workqueue(free_mr->free_mr_wq);
@@ -1826,9 +1759,12 @@ static void hns_roce_v1_set_mtu(struct hns_roce_dev *hr_dev, u8 phy_port,
static int hns_roce_v1_write_mtpt(void *mb_buf, struct hns_roce_mr *mr,
unsigned long mtpt_idx)
{
+ struct hns_roce_dev *hr_dev = to_hr_dev(mr->ibmr.device);
+ u64 pages[HNS_ROCE_MAX_INNER_MTPT_NUM] = { 0 };
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct hns_roce_v1_mpt_entry *mpt_entry;
- struct sg_dma_page_iter sg_iter;
- u64 *pages;
+ dma_addr_t pbl_ba;
+ int count;
int i;
/* MPT filled into mailbox buf */
@@ -1878,22 +1814,15 @@ static int hns_roce_v1_write_mtpt(void *mb_buf, struct hns_roce_mr *mr,
if (mr->type == MR_TYPE_DMA)
return 0;
- pages = (u64 *) __get_free_page(GFP_KERNEL);
- if (!pages)
- return -ENOMEM;
-
- i = 0;
- for_each_sg_dma_page(mr->umem->sg_head.sgl, &sg_iter, mr->umem->nmap, 0) {
- pages[i] = ((u64)sg_page_iter_dma_address(&sg_iter)) >> 12;
-
- /* Directly record to MTPT table firstly 7 entry */
- if (i >= HNS_ROCE_MAX_INNER_MTPT_NUM)
- break;
- i++;
+ count = hns_roce_mtr_find(hr_dev, &mr->pbl_mtr, 0, pages,
+ ARRAY_SIZE(pages), &pbl_ba);
+ if (count < 1) {
+ ibdev_err(ibdev, "failed to find PBL mtr, count = %d.", count);
+ return -ENOBUFS;
}
/* Register user mr */
- for (i = 0; i < HNS_ROCE_MAX_INNER_MTPT_NUM; i++) {
+ for (i = 0; i < count; i++) {
switch (i) {
case 0:
mpt_entry->pa0_l = cpu_to_le32((u32)(pages[i]));
@@ -1959,20 +1888,17 @@ static int hns_roce_v1_write_mtpt(void *mb_buf, struct hns_roce_mr *mr,
}
}
- free_page((unsigned long) pages);
-
- mpt_entry->pbl_addr_l = cpu_to_le32((u32)(mr->pbl_dma_addr));
-
+ mpt_entry->pbl_addr_l = cpu_to_le32(pbl_ba);
roce_set_field(mpt_entry->mpt_byte_12, MPT_BYTE_12_PBL_ADDR_H_M,
- MPT_BYTE_12_PBL_ADDR_H_S,
- ((u32)(mr->pbl_dma_addr >> 32)));
+ MPT_BYTE_12_PBL_ADDR_H_S, upper_32_bits(pbl_ba));
return 0;
}
static void *get_cqe(struct hns_roce_cq *hr_cq, int n)
{
- return hns_roce_buf_offset(&hr_cq->buf, n * HNS_ROCE_V1_CQE_ENTRY_SIZE);
+ return hns_roce_buf_offset(hr_cq->mtr.kmem,
+ n * HNS_ROCE_V1_CQE_ENTRY_SIZE);
}
static void *get_sw_cqe(struct hns_roce_cq *hr_cq, int n)
@@ -2066,16 +1992,12 @@ static void hns_roce_v1_write_cqc(struct hns_roce_dev *hr_dev,
struct hns_roce_cq *hr_cq, void *mb_buf,
u64 *mtts, dma_addr_t dma_handle)
{
- struct hns_roce_cq_context *cq_context = NULL;
- struct hns_roce_buf_list *tptr_buf;
- struct hns_roce_v1_priv *priv;
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
+ struct hns_roce_buf_list *tptr_buf = &priv->tptr_table.tptr_buf;
+ struct hns_roce_cq_context *cq_context = mb_buf;
dma_addr_t tptr_dma_addr;
int offset;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
- tptr_buf = &priv->tptr_table.tptr_buf;
-
- cq_context = mb_buf;
memset(cq_context, 0, sizeof(*cq_context));
/* Get the tptr for this CQ. */
@@ -2416,16 +2338,14 @@ static int hns_roce_v1_clear_hem(struct hns_roce_dev *hr_dev,
struct hns_roce_hem_table *table, int obj,
int step_idx)
{
+ struct hns_roce_v1_priv *priv = hr_dev->priv;
struct device *dev = &hr_dev->pdev->dev;
- struct hns_roce_v1_priv *priv;
- unsigned long flags = 0;
long end = HW_SYNC_TIMEOUT_MSECS;
__le32 bt_cmd_val[2] = {0};
+ unsigned long flags = 0;
void __iomem *bt_cmd;
u64 bt_ba = 0;
- priv = (struct hns_roce_v1_priv *)hr_dev->priv;
-
switch (table->type) {
case HEM_TYPE_QPC:
bt_ba = priv->bt_table.qpc_buf.map >> 12;
@@ -2479,7 +2399,6 @@ static int hns_roce_v1_clear_hem(struct hns_roce_dev *hr_dev,
}
static int hns_roce_v1_qp_modify(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt,
enum hns_roce_qp_state cur_state,
enum hns_roce_qp_state new_state,
struct hns_roce_qp_context *context,
@@ -2560,6 +2479,29 @@ static int hns_roce_v1_qp_modify(struct hns_roce_dev *hr_dev,
return ret;
}
+static int find_wqe_mtt(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
+ u64 *sq_ba, u64 *rq_ba, dma_addr_t *bt_ba)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ int rq_pa_start;
+ int count;
+
+ count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr, 0, sq_ba, 1, bt_ba);
+ if (count < 1) {
+ ibdev_err(ibdev, "Failed to find SQ ba\n");
+ return -ENOBUFS;
+ }
+ rq_pa_start = hr_qp->rq.offset >> hr_qp->mtr.hem_cfg.buf_pg_shift;
+ count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr, rq_pa_start, rq_ba, 1,
+ NULL);
+ if (!count) {
+ ibdev_err(ibdev, "Failed to find RQ ba\n");
+ return -ENOBUFS;
+ }
+
+ return 0;
+}
+
static int hns_roce_v1_m_sqp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
int attr_mask, enum ib_qp_state cur_state,
enum ib_qp_state new_state)
@@ -2567,25 +2509,20 @@ static int hns_roce_v1_m_sqp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
struct hns_roce_sqp_context *context;
- struct device *dev = &hr_dev->pdev->dev;
dma_addr_t dma_handle = 0;
u32 __iomem *addr;
- int rq_pa_start;
+ u64 sq_ba = 0;
+ u64 rq_ba = 0;
__le32 tmp;
u32 reg_val;
- u64 *mtts;
context = kzalloc(sizeof(*context), GFP_KERNEL);
if (!context)
return -ENOMEM;
/* Search QP buf's MTTs */
- mtts = hns_roce_table_find(hr_dev, &hr_dev->mr_table.mtt_table,
- hr_qp->mtt.first_seg, &dma_handle);
- if (!mtts) {
- dev_err(dev, "qp buf pa find failed\n");
+ if (find_wqe_mtt(hr_dev, hr_qp, &sq_ba, &rq_ba, &dma_handle))
goto out;
- }
if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {
roce_set_field(context->qp1c_bytes_4,
@@ -2599,11 +2536,11 @@ static int hns_roce_v1_m_sqp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
roce_set_field(context->qp1c_bytes_4, QP1C_BYTES_4_PD_M,
QP1C_BYTES_4_PD_S, to_hr_pd(ibqp->pd)->pdn);
- context->sq_rq_bt_l = cpu_to_le32((u32)(dma_handle));
+ context->sq_rq_bt_l = cpu_to_le32(dma_handle);
roce_set_field(context->qp1c_bytes_12,
QP1C_BYTES_12_SQ_RQ_BT_H_M,
QP1C_BYTES_12_SQ_RQ_BT_H_S,
- ((u32)(dma_handle >> 32)));
+ upper_32_bits(dma_handle));
roce_set_field(context->qp1c_bytes_16, QP1C_BYTES_16_RQ_HEAD_M,
QP1C_BYTES_16_RQ_HEAD_S, hr_qp->rq.head);
@@ -2624,14 +2561,12 @@ static int hns_roce_v1_m_sqp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
roce_set_field(context->qp1c_bytes_20, QP1C_BYTES_20_PKEY_IDX_M,
QP1C_BYTES_20_PKEY_IDX_S, attr->pkey_index);
- rq_pa_start = (u32)hr_qp->rq.offset / PAGE_SIZE;
- context->cur_rq_wqe_ba_l =
- cpu_to_le32((u32)(mtts[rq_pa_start]));
+ context->cur_rq_wqe_ba_l = cpu_to_le32(rq_ba);
roce_set_field(context->qp1c_bytes_28,
QP1C_BYTES_28_CUR_RQ_WQE_BA_H_M,
QP1C_BYTES_28_CUR_RQ_WQE_BA_H_S,
- (mtts[rq_pa_start]) >> 32);
+ upper_32_bits(rq_ba));
roce_set_field(context->qp1c_bytes_28,
QP1C_BYTES_28_RQ_CUR_IDX_M,
QP1C_BYTES_28_RQ_CUR_IDX_S, 0);
@@ -2645,12 +2580,12 @@ static int hns_roce_v1_m_sqp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
QP1C_BYTES_32_TX_CQ_NUM_S,
to_hr_cq(ibqp->send_cq)->cqn);
- context->cur_sq_wqe_ba_l = cpu_to_le32((u32)mtts[0]);
+ context->cur_sq_wqe_ba_l = cpu_to_le32(sq_ba);
roce_set_field(context->qp1c_bytes_40,
QP1C_BYTES_40_CUR_SQ_WQE_BA_H_M,
QP1C_BYTES_40_CUR_SQ_WQE_BA_H_S,
- (mtts[0]) >> 32);
+ upper_32_bits(sq_ba));
roce_set_field(context->qp1c_bytes_40,
QP1C_BYTES_40_SQ_CUR_IDX_M,
QP1C_BYTES_40_SQ_CUR_IDX_S, 0);
@@ -2704,6 +2639,28 @@ out:
return -EINVAL;
}
+static bool check_qp_state(enum ib_qp_state cur_state,
+ enum ib_qp_state new_state)
+{
+ static const bool sm[][IB_QPS_ERR + 1] = {
+ [IB_QPS_RESET] = { [IB_QPS_RESET] = true,
+ [IB_QPS_INIT] = true },
+ [IB_QPS_INIT] = { [IB_QPS_RESET] = true,
+ [IB_QPS_INIT] = true,
+ [IB_QPS_RTR] = true,
+ [IB_QPS_ERR] = true },
+ [IB_QPS_RTR] = { [IB_QPS_RESET] = true,
+ [IB_QPS_RTS] = true,
+ [IB_QPS_ERR] = true },
+ [IB_QPS_RTS] = { [IB_QPS_RESET] = true, [IB_QPS_ERR] = true },
+ [IB_QPS_SQD] = {},
+ [IB_QPS_SQE] = {},
+ [IB_QPS_ERR] = { [IB_QPS_RESET] = true, [IB_QPS_ERR] = true }
+ };
+
+ return sm[cur_state][new_state];
+}
+
static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
int attr_mask, enum ib_qp_state cur_state,
enum ib_qp_state new_state)
@@ -2716,26 +2673,29 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
dma_addr_t dma_handle_2 = 0;
dma_addr_t dma_handle = 0;
__le32 doorbell[2] = {0};
- int rq_pa_start = 0;
u64 *mtts_2 = NULL;
int ret = -EINVAL;
- u64 *mtts = NULL;
+ u64 sq_ba = 0;
+ u64 rq_ba = 0;
int port;
u8 port_num;
u8 *dmac;
u8 *smac;
+ if (!check_qp_state(cur_state, new_state)) {
+ ibdev_err(ibqp->device,
+ "not support QP(%u) status from %d to %d\n",
+ ibqp->qp_num, cur_state, new_state);
+ return -EINVAL;
+ }
+
context = kzalloc(sizeof(*context), GFP_KERNEL);
if (!context)
return -ENOMEM;
/* Search qp buf's mtts */
- mtts = hns_roce_table_find(hr_dev, &hr_dev->mr_table.mtt_table,
- hr_qp->mtt.first_seg, &dma_handle);
- if (mtts == NULL) {
- dev_err(dev, "qp buf pa find failed\n");
+ if (find_wqe_mtt(hr_dev, hr_qp, &sq_ba, &rq_ba, &dma_handle))
goto out;
- }
/* Search IRRL's mtts */
mtts_2 = hns_roce_table_find(hr_dev, &hr_dev->qp_table.irrl_table,
@@ -2890,11 +2850,11 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
dmac = (u8 *)attr->ah_attr.roce.dmac;
- context->sq_rq_bt_l = cpu_to_le32((u32)(dma_handle));
+ context->sq_rq_bt_l = cpu_to_le32(dma_handle);
roce_set_field(context->qpc_bytes_24,
QP_CONTEXT_QPC_BYTES_24_SQ_RQ_BT_H_M,
QP_CONTEXT_QPC_BYTES_24_SQ_RQ_BT_H_S,
- ((u32)(dma_handle >> 32)));
+ upper_32_bits(dma_handle));
roce_set_bit(context->qpc_bytes_24,
QP_CONTEXT_QPC_BYTE_24_REMOTE_ENABLE_E2E_CREDITS_S,
1);
@@ -2993,14 +2953,12 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
QP_CONTEXT_QPC_BYTES_68_RQ_CUR_INDEX_M,
QP_CONTEXT_QPC_BYTES_68_RQ_CUR_INDEX_S, 0);
- rq_pa_start = (u32)hr_qp->rq.offset / PAGE_SIZE;
- context->cur_rq_wqe_ba_l =
- cpu_to_le32((u32)(mtts[rq_pa_start]));
+ context->cur_rq_wqe_ba_l = cpu_to_le32(rq_ba);
roce_set_field(context->qpc_bytes_76,
QP_CONTEXT_QPC_BYTES_76_CUR_RQ_WQE_BA_H_M,
QP_CONTEXT_QPC_BYTES_76_CUR_RQ_WQE_BA_H_S,
- mtts[rq_pa_start] >> 32);
+ upper_32_bits(rq_ba));
roce_set_field(context->qpc_bytes_76,
QP_CONTEXT_QPC_BYTES_76_RX_REQ_MSN_M,
QP_CONTEXT_QPC_BYTES_76_RX_REQ_MSN_S, 0);
@@ -3062,8 +3020,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
QP_CONTEXT_QPC_BYTES_156_SL_S,
rdma_ah_get_sl(&attr->ah_attr));
hr_qp->sl = rdma_ah_get_sl(&attr->ah_attr);
- } else if (cur_state == IB_QPS_RTR &&
- new_state == IB_QPS_RTS) {
+ } else if (cur_state == IB_QPS_RTR && new_state == IB_QPS_RTS) {
/* If exist optional param, return error */
if ((attr_mask & IB_QP_ALT_PATH) ||
(attr_mask & IB_QP_ACCESS_FLAGS) ||
@@ -3075,12 +3032,12 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
goto out;
}
- context->rx_cur_sq_wqe_ba_l = cpu_to_le32((u32)(mtts[0]));
+ context->rx_cur_sq_wqe_ba_l = cpu_to_le32(sq_ba);
roce_set_field(context->qpc_bytes_120,
QP_CONTEXT_QPC_BYTES_120_RX_CUR_SQ_WQE_BA_H_M,
QP_CONTEXT_QPC_BYTES_120_RX_CUR_SQ_WQE_BA_H_S,
- (mtts[0]) >> 32);
+ upper_32_bits(sq_ba));
roce_set_field(context->qpc_bytes_124,
QP_CONTEXT_QPC_BYTES_124_RX_ACK_MSN_M,
@@ -3223,28 +3180,18 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
QP_CONTEXT_QPC_BYTES_180_SQ_HEAD_M,
QP_CONTEXT_QPC_BYTES_180_SQ_HEAD_S, 0);
- context->tx_cur_sq_wqe_ba_l = cpu_to_le32((u32)(mtts[0]));
+ context->tx_cur_sq_wqe_ba_l = cpu_to_le32(sq_ba);
roce_set_field(context->qpc_bytes_188,
QP_CONTEXT_QPC_BYTES_188_TX_CUR_SQ_WQE_BA_H_M,
QP_CONTEXT_QPC_BYTES_188_TX_CUR_SQ_WQE_BA_H_S,
- (mtts[0]) >> 32);
+ upper_32_bits(sq_ba));
roce_set_bit(context->qpc_bytes_188,
QP_CONTEXT_QPC_BYTES_188_PKT_RETRY_FLG_S, 0);
roce_set_field(context->qpc_bytes_188,
QP_CONTEXT_QPC_BYTES_188_TX_RETRY_CUR_INDEX_M,
QP_CONTEXT_QPC_BYTES_188_TX_RETRY_CUR_INDEX_S,
0);
- } else if (!((cur_state == IB_QPS_INIT && new_state == IB_QPS_RESET) ||
- (cur_state == IB_QPS_INIT && new_state == IB_QPS_ERR) ||
- (cur_state == IB_QPS_RTR && new_state == IB_QPS_RESET) ||
- (cur_state == IB_QPS_RTR && new_state == IB_QPS_ERR) ||
- (cur_state == IB_QPS_RTS && new_state == IB_QPS_RESET) ||
- (cur_state == IB_QPS_RTS && new_state == IB_QPS_ERR) ||
- (cur_state == IB_QPS_ERR && new_state == IB_QPS_RESET) ||
- (cur_state == IB_QPS_ERR && new_state == IB_QPS_ERR))) {
- dev_err(dev, "not support this status migration\n");
- goto out;
}
/* Every status migrate must change state */
@@ -3253,8 +3200,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr,
QP_CONTEXT_QPC_BYTES_144_QP_STATE_S, new_state);
/* SW pass context to HW */
- ret = hns_roce_v1_qp_modify(hr_dev, &hr_qp->mtt,
- to_hns_roce_state(cur_state),
+ ret = hns_roce_v1_qp_modify(hr_dev, to_hns_roce_state(cur_state),
to_hns_roce_state(new_state), context,
hr_qp);
if (ret) {
@@ -3636,8 +3582,6 @@ static void hns_roce_v1_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata)
u32 cqe_cnt_cur;
int wait_time = 0;
- hns_roce_free_cqc(hr_dev, hr_cq);
-
/*
* Before freeing cq buffer, we need to ensure that the outstanding CQE
* have been written by checking the CQE counter.
@@ -3660,14 +3604,6 @@ static void hns_roce_v1_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata)
}
wait_time++;
}
-
- hns_roce_mtt_cleanup(hr_dev, &hr_cq->mtt);
-
- ib_umem_release(hr_cq->umem);
- if (!udata) {
- /* Free the buff of stored cq */
- hns_roce_buf_free(hr_dev, hr_cq->buf.size, &hr_cq->buf);
- }
}
static void set_eq_cons_index_v1(struct hns_roce_eq *eq, int req_not)
diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c
index c3316672b70e..c597d7281629 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c
+++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c
@@ -95,6 +95,7 @@ static void set_frmr_seg(struct hns_roce_v2_rc_send_wqe *rc_sq_wqe,
{
struct hns_roce_mr *mr = to_hr_mr(wr->mr);
struct hns_roce_wqe_frmr_seg *fseg = wqe;
+ u64 pbl_ba;
/* use ib_access_flags */
roce_set_bit(rc_sq_wqe->byte_4, V2_RC_FRMR_WQE_BYTE_4_BIND_EN_S,
@@ -109,26 +110,27 @@ static void set_frmr_seg(struct hns_roce_v2_rc_send_wqe *rc_sq_wqe,
wr->access & IB_ACCESS_LOCAL_WRITE ? 1 : 0);
/* Data structure reuse may lead to confusion */
- rc_sq_wqe->msg_len = cpu_to_le32(mr->pbl_ba & 0xffffffff);
- rc_sq_wqe->inv_key = cpu_to_le32(mr->pbl_ba >> 32);
+ pbl_ba = mr->pbl_mtr.hem_cfg.root_ba;
+ rc_sq_wqe->msg_len = cpu_to_le32(lower_32_bits(pbl_ba));
+ rc_sq_wqe->inv_key = cpu_to_le32(upper_32_bits(pbl_ba));
rc_sq_wqe->byte_16 = cpu_to_le32(wr->mr->length & 0xffffffff);
rc_sq_wqe->byte_20 = cpu_to_le32(wr->mr->length >> 32);
rc_sq_wqe->rkey = cpu_to_le32(wr->key);
rc_sq_wqe->va = cpu_to_le64(wr->mr->iova);
- fseg->pbl_size = cpu_to_le32(mr->pbl_size);
+ fseg->pbl_size = cpu_to_le32(mr->npages);
roce_set_field(fseg->mode_buf_pg_sz,
V2_RC_FRMR_WQE_BYTE_40_PBL_BUF_PG_SZ_M,
V2_RC_FRMR_WQE_BYTE_40_PBL_BUF_PG_SZ_S,
- mr->pbl_buf_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(mr->pbl_mtr.hem_cfg.buf_pg_shift));
roce_set_bit(fseg->mode_buf_pg_sz,
V2_RC_FRMR_WQE_BYTE_40_BLK_MODE_S, 0);
}
static void set_atomic_seg(const struct ib_send_wr *wr, void *wqe,
struct hns_roce_v2_rc_send_wqe *rc_sq_wqe,
- int valid_num_sge)
+ unsigned int valid_num_sge)
{
struct hns_roce_wqe_atomic_seg *aseg;
@@ -149,56 +151,33 @@ static void set_atomic_seg(const struct ib_send_wr *wr, void *wqe,
}
static void set_extend_sge(struct hns_roce_qp *qp, const struct ib_send_wr *wr,
- unsigned int *sge_ind, int valid_num_sge)
+ unsigned int *sge_ind, unsigned int valid_num_sge)
{
struct hns_roce_v2_wqe_data_seg *dseg;
- struct ib_sge *sg;
- int num_in_wqe = 0;
- int extend_sge_num;
- int fi_sge_num;
- int se_sge_num;
- int shift;
- int i;
+ unsigned int cnt = valid_num_sge;
+ struct ib_sge *sge = wr->sg_list;
+ unsigned int idx = *sge_ind;
- if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC)
- num_in_wqe = HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE;
- extend_sge_num = valid_num_sge - num_in_wqe;
- sg = wr->sg_list + num_in_wqe;
- shift = qp->hr_buf.page_shift;
+ if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) {
+ cnt -= HNS_ROCE_SGE_IN_WQE;
+ sge += HNS_ROCE_SGE_IN_WQE;
+ }
- /*
- * Check whether wr->num_sge sges are in the same page. If not, we
- * should calculate how many sges in the first page and the second
- * page.
- */
- dseg = hns_roce_get_extend_sge(qp, (*sge_ind) & (qp->sge.sge_cnt - 1));
- fi_sge_num = (round_up((uintptr_t)dseg, 1 << shift) -
- (uintptr_t)dseg) /
- sizeof(struct hns_roce_v2_wqe_data_seg);
- if (extend_sge_num > fi_sge_num) {
- se_sge_num = extend_sge_num - fi_sge_num;
- for (i = 0; i < fi_sge_num; i++) {
- set_data_seg_v2(dseg++, sg + i);
- (*sge_ind)++;
- }
- dseg = hns_roce_get_extend_sge(qp,
- (*sge_ind) & (qp->sge.sge_cnt - 1));
- for (i = 0; i < se_sge_num; i++) {
- set_data_seg_v2(dseg++, sg + fi_sge_num + i);
- (*sge_ind)++;
- }
- } else {
- for (i = 0; i < extend_sge_num; i++) {
- set_data_seg_v2(dseg++, sg + i);
- (*sge_ind)++;
- }
+ while (cnt > 0) {
+ dseg = hns_roce_get_extend_sge(qp, idx & (qp->sge.sge_cnt - 1));
+ set_data_seg_v2(dseg, sge);
+ idx++;
+ sge++;
+ cnt--;
}
+
+ *sge_ind = idx;
}
static int set_rwqe_data_seg(struct ib_qp *ibqp, const struct ib_send_wr *wr,
struct hns_roce_v2_rc_send_wqe *rc_sq_wqe,
void *wqe, unsigned int *sge_ind,
- int valid_num_sge)
+ unsigned int valid_num_sge)
{
struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
struct hns_roce_v2_wqe_data_seg *dseg = wqe;
@@ -208,15 +187,15 @@ static int set_rwqe_data_seg(struct ib_qp *ibqp, const struct ib_send_wr *wr,
int i;
if (wr->send_flags & IB_SEND_INLINE && valid_num_sge) {
- if (le32_to_cpu(rc_sq_wqe->msg_len) >
- hr_dev->caps.max_sq_inline) {
+ if (unlikely(le32_to_cpu(rc_sq_wqe->msg_len) >
+ hr_dev->caps.max_sq_inline)) {
ibdev_err(ibdev, "inline len(1-%d)=%d, illegal",
rc_sq_wqe->msg_len,
hr_dev->caps.max_sq_inline);
return -EINVAL;
}
- if (wr->opcode == IB_WR_RDMA_READ) {
+ if (unlikely(wr->opcode == IB_WR_RDMA_READ)) {
ibdev_err(ibdev, "Not support inline data!\n");
return -EINVAL;
}
@@ -230,7 +209,7 @@ static int set_rwqe_data_seg(struct ib_qp *ibqp, const struct ib_send_wr *wr,
roce_set_bit(rc_sq_wqe->byte_4, V2_RC_SEND_WQE_BYTE_4_INLINE_S,
1);
} else {
- if (valid_num_sge <= HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE) {
+ if (valid_num_sge <= HNS_ROCE_SGE_IN_WQE) {
for (i = 0; i < wr->num_sge; i++) {
if (likely(wr->sg_list[i].length)) {
set_data_seg_v2(dseg, wr->sg_list + i);
@@ -243,8 +222,8 @@ static int set_rwqe_data_seg(struct ib_qp *ibqp, const struct ib_send_wr *wr,
V2_RC_SEND_WQE_BYTE_20_MSG_START_SGE_IDX_S,
(*sge_ind) & (qp->sge.sge_cnt - 1));
- for (i = 0; i < wr->num_sge &&
- j < HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE; i++) {
+ for (i = 0; i < wr->num_sge && j < HNS_ROCE_SGE_IN_WQE;
+ i++) {
if (likely(wr->sg_list[i].length)) {
set_data_seg_v2(dseg, wr->sg_list + i);
dseg++;
@@ -290,10 +269,11 @@ static int check_send_valid(struct hns_roce_dev *hr_dev,
return 0;
}
-static inline int calc_wr_sge_num(const struct ib_send_wr *wr, u32 *sge_len)
+static unsigned int calc_wr_sge_num(const struct ib_send_wr *wr,
+ unsigned int *sge_len)
{
- int valid_num = 0;
- u32 len = 0;
+ unsigned int valid_num = 0;
+ unsigned int len = 0;
int i;
for (i = 0; i < wr->num_sge; i++) {
@@ -424,7 +404,7 @@ static inline int set_rc_wqe(struct hns_roce_qp *qp,
{
struct hns_roce_v2_rc_send_wqe *rc_sq_wqe = wqe;
unsigned int curr_idx = *sge_idx;
- int valid_num_sge;
+ unsigned int valid_num_sge;
u32 msg_len = 0;
int ret = 0;
@@ -521,8 +501,7 @@ static inline void update_sq_db(struct hns_roce_dev *hr_dev,
roce_set_field(sq_db.byte_4, V2_DB_BYTE_4_CMD_M,
V2_DB_BYTE_4_CMD_S, HNS_ROCE_V2_SQ_DB);
roce_set_field(sq_db.parameter, V2_DB_PARAMETER_IDX_M,
- V2_DB_PARAMETER_IDX_S,
- qp->sq.head & ((qp->sq.wqe_cnt << 1) - 1));
+ V2_DB_PARAMETER_IDX_S, qp->sq.head);
roce_set_field(sq_db.parameter, V2_DB_PARAMETER_SL_M,
V2_DB_PARAMETER_SL_S, qp->sl);
@@ -548,7 +527,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp,
spin_lock_irqsave(&qp->sq.lock, flags);
ret = check_send_valid(hr_dev, qp);
- if (ret) {
+ if (unlikely(ret)) {
*bad_wr = wr;
nreq = 0;
goto out;
@@ -584,7 +563,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp,
else if (ibqp->qp_type == IB_QPT_RC)
ret = set_rc_wqe(qp, wr, wqe, &sge_idx, owner_bit);
- if (ret) {
+ if (unlikely(ret)) {
*bad_wr = wr;
goto out;
}
@@ -634,15 +613,15 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp,
spin_lock_irqsave(&hr_qp->rq.lock, flags);
ret = check_recv_valid(hr_dev, hr_qp);
- if (ret) {
+ if (unlikely(ret)) {
*bad_wr = wr;
nreq = 0;
goto out;
}
for (nreq = 0; wr; ++nreq, wr = wr->next) {
- if (hns_roce_wq_overflow(&hr_qp->rq, nreq,
- hr_qp->ibqp.recv_cq)) {
+ if (unlikely(hns_roce_wq_overflow(&hr_qp->rq, nreq,
+ hr_qp->ibqp.recv_cq))) {
ret = -ENOMEM;
*bad_wr = wr;
goto out;
@@ -650,7 +629,7 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp,
wqe_idx = (hr_qp->rq.head + nreq) & (hr_qp->rq.wqe_cnt - 1);
- if (unlikely(wr->num_sge > hr_qp->rq.max_gs)) {
+ if (unlikely(wr->num_sge >= hr_qp->rq.max_gs)) {
ibdev_err(ibdev, "rq:num_sge=%d >= qp->sq.max_gs=%d\n",
wr->num_sge, hr_qp->rq.max_gs);
ret = -EINVAL;
@@ -667,13 +646,14 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp,
dseg++;
}
- if (i < hr_qp->rq.max_gs) {
+ if (wr->num_sge < hr_qp->rq.max_gs) {
dseg->lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY);
dseg->addr = 0;
+ dseg->len = cpu_to_le32(HNS_ROCE_INVALID_SGE_LENGTH);
}
/* rq support inline data */
- if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) {
+ if (hr_qp->rq_inl_buf.wqe_cnt) {
sge_list = hr_qp->rq_inl_buf.wqe_list[wqe_idx].sg_list;
hr_qp->rq_inl_buf.wqe_list[wqe_idx].sge_cnt =
(u32)wr->num_sge;
@@ -715,6 +695,129 @@ out:
return ret;
}
+static void *get_srq_wqe(struct hns_roce_srq *srq, int n)
+{
+ return hns_roce_buf_offset(srq->buf_mtr.kmem, n << srq->wqe_shift);
+}
+
+static void *get_idx_buf(struct hns_roce_idx_que *idx_que, int n)
+{
+ return hns_roce_buf_offset(idx_que->mtr.kmem,
+ n << idx_que->entry_shift);
+}
+
+static void hns_roce_free_srq_wqe(struct hns_roce_srq *srq, int wqe_index)
+{
+ /* always called with interrupts disabled. */
+ spin_lock(&srq->lock);
+
+ bitmap_clear(srq->idx_que.bitmap, wqe_index, 1);
+ srq->tail++;
+
+ spin_unlock(&srq->lock);
+}
+
+static int find_empty_entry(struct hns_roce_idx_que *idx_que,
+ unsigned long size)
+{
+ int wqe_idx;
+
+ if (unlikely(bitmap_full(idx_que->bitmap, size)))
+ return -ENOSPC;
+
+ wqe_idx = find_first_zero_bit(idx_que->bitmap, size);
+
+ bitmap_set(idx_que->bitmap, wqe_idx, 1);
+
+ return wqe_idx;
+}
+
+static int hns_roce_v2_post_srq_recv(struct ib_srq *ibsrq,
+ const struct ib_recv_wr *wr,
+ const struct ib_recv_wr **bad_wr)
+{
+ struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
+ struct hns_roce_srq *srq = to_hr_srq(ibsrq);
+ struct hns_roce_v2_wqe_data_seg *dseg;
+ struct hns_roce_v2_db srq_db;
+ unsigned long flags;
+ __le32 *srq_idx;
+ int ret = 0;
+ int wqe_idx;
+ void *wqe;
+ int nreq;
+ int ind;
+ int i;
+
+ spin_lock_irqsave(&srq->lock, flags);
+
+ ind = srq->head & (srq->wqe_cnt - 1);
+
+ for (nreq = 0; wr; ++nreq, wr = wr->next) {
+ if (unlikely(wr->num_sge >= srq->max_gs)) {
+ ret = -EINVAL;
+ *bad_wr = wr;
+ break;
+ }
+
+ if (unlikely(srq->head == srq->tail)) {
+ ret = -ENOMEM;
+ *bad_wr = wr;
+ break;
+ }
+
+ wqe_idx = find_empty_entry(&srq->idx_que, srq->wqe_cnt);
+ if (unlikely(wqe_idx < 0)) {
+ ret = -ENOMEM;
+ *bad_wr = wr;
+ break;
+ }
+
+ wqe = get_srq_wqe(srq, wqe_idx);
+ dseg = (struct hns_roce_v2_wqe_data_seg *)wqe;
+
+ for (i = 0; i < wr->num_sge; ++i) {
+ dseg[i].len = cpu_to_le32(wr->sg_list[i].length);
+ dseg[i].lkey = cpu_to_le32(wr->sg_list[i].lkey);
+ dseg[i].addr = cpu_to_le64(wr->sg_list[i].addr);
+ }
+
+ if (wr->num_sge < srq->max_gs) {
+ dseg[i].len = cpu_to_le32(HNS_ROCE_INVALID_SGE_LENGTH);
+ dseg[i].lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY);
+ dseg[i].addr = 0;
+ }
+
+ srq_idx = get_idx_buf(&srq->idx_que, ind);
+ *srq_idx = cpu_to_le32(wqe_idx);
+
+ srq->wrid[wqe_idx] = wr->wr_id;
+ ind = (ind + 1) & (srq->wqe_cnt - 1);
+ }
+
+ if (likely(nreq)) {
+ srq->head += nreq;
+
+ /*
+ * Make sure that descriptors are written before
+ * doorbell record.
+ */
+ wmb();
+
+ srq_db.byte_4 =
+ cpu_to_le32(HNS_ROCE_V2_SRQ_DB << V2_DB_BYTE_4_CMD_S |
+ (srq->srqn & V2_DB_BYTE_4_TAG_M));
+ srq_db.parameter =
+ cpu_to_le32(srq->head & V2_DB_PARAMETER_IDX_M);
+
+ hns_roce_write64(hr_dev, (__le32 *)&srq_db, srq->db_reg_l);
+ }
+
+ spin_unlock_irqrestore(&srq->lock, flags);
+
+ return ret;
+}
+
static int hns_roce_v2_cmd_hw_reseted(struct hns_roce_dev *hr_dev,
unsigned long instance_stage,
unsigned long reset_stage)
@@ -742,7 +845,7 @@ static int hns_roce_v2_cmd_hw_resetting(struct hns_roce_dev *hr_dev,
unsigned long instance_stage,
unsigned long reset_stage)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
@@ -768,7 +871,7 @@ static int hns_roce_v2_cmd_hw_resetting(struct hns_roce_dev *hr_dev,
static int hns_roce_v2_cmd_sw_resetting(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
@@ -785,7 +888,7 @@ static int hns_roce_v2_cmd_sw_resetting(struct hns_roce_dev *hr_dev)
static int hns_roce_v2_rst_process_cmd(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
unsigned long instance_stage; /* the current instance stage */
@@ -865,7 +968,7 @@ static void hns_roce_free_cmq_desc(struct hns_roce_dev *hr_dev,
static int hns_roce_init_cmq_ring(struct hns_roce_dev *hr_dev, bool ring_type)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hns_roce_v2_cmq_ring *ring = (ring_type == TYPE_CSQ) ?
&priv->cmq.csq : &priv->cmq.crq;
@@ -878,7 +981,7 @@ static int hns_roce_init_cmq_ring(struct hns_roce_dev *hr_dev, bool ring_type)
static void hns_roce_cmq_init_regs(struct hns_roce_dev *hr_dev, bool ring_type)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hns_roce_v2_cmq_ring *ring = (ring_type == TYPE_CSQ) ?
&priv->cmq.csq : &priv->cmq.crq;
dma_addr_t dma = ring->desc_dma_addr;
@@ -904,7 +1007,7 @@ static void hns_roce_cmq_init_regs(struct hns_roce_dev *hr_dev, bool ring_type)
static int hns_roce_v2_cmq_init(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
int ret;
/* Setup the queue entries for command queue */
@@ -948,7 +1051,7 @@ err_crq:
static void hns_roce_v2_cmq_exit(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
hns_roce_free_cmq_desc(hr_dev, &priv->cmq.csq);
hns_roce_free_cmq_desc(hr_dev, &priv->cmq.crq);
@@ -970,15 +1073,15 @@ static void hns_roce_cmq_setup_basic_desc(struct hns_roce_cmq_desc *desc,
static int hns_roce_cmq_csq_done(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
u32 head = roce_read(hr_dev, ROCEE_TX_CMQ_HEAD_REG);
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
return head == priv->cmq.csq.next_to_use;
}
static int hns_roce_cmq_csq_clean(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq;
struct hns_roce_cmq_desc *desc;
u16 ntc = csq->next_to_clean;
@@ -1003,7 +1106,7 @@ static int hns_roce_cmq_csq_clean(struct hns_roce_dev *hr_dev)
static int __hns_roce_cmq_send(struct hns_roce_dev *hr_dev,
struct hns_roce_cmq_desc *desc, int num)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hns_roce_v2_cmq_ring *csq = &priv->cmq.csq;
struct hns_roce_cmq_desc *desc_to_use;
bool complete = false;
@@ -1131,7 +1234,7 @@ static int hns_roce_cmq_query_hw_info(struct hns_roce_dev *hr_dev)
static bool hns_roce_func_clr_chk_rst(struct hns_roce_dev *hr_dev)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
unsigned long reset_cnt;
@@ -1151,7 +1254,7 @@ static bool hns_roce_func_clr_chk_rst(struct hns_roce_dev *hr_dev)
static void hns_roce_func_clr_rst_prc(struct hns_roce_dev *hr_dev, int retval,
int flag)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
unsigned long instance_stage;
@@ -1349,34 +1452,26 @@ static int hns_roce_query_pf_resource(struct hns_roce_dev *hr_dev)
static int hns_roce_query_pf_timer_resource(struct hns_roce_dev *hr_dev)
{
struct hns_roce_pf_timer_res_a *req_a;
- struct hns_roce_cmq_desc desc[2];
- int ret, i;
+ struct hns_roce_cmq_desc desc;
+ int ret;
- for (i = 0; i < 2; i++) {
- hns_roce_cmq_setup_basic_desc(&desc[i],
- HNS_ROCE_OPC_QUERY_PF_TIMER_RES,
- true);
+ hns_roce_cmq_setup_basic_desc(&desc, HNS_ROCE_OPC_QUERY_PF_TIMER_RES,
+ true);
- if (i == 0)
- desc[i].flag |= cpu_to_le16(HNS_ROCE_CMD_FLAG_NEXT);
- else
- desc[i].flag &= ~cpu_to_le16(HNS_ROCE_CMD_FLAG_NEXT);
- }
-
- ret = hns_roce_cmq_send(hr_dev, desc, 2);
+ ret = hns_roce_cmq_send(hr_dev, &desc, 1);
if (ret)
return ret;
- req_a = (struct hns_roce_pf_timer_res_a *)desc[0].data;
+ req_a = (struct hns_roce_pf_timer_res_a *)desc.data;
hr_dev->caps.qpc_timer_bt_num =
- roce_get_field(req_a->qpc_timer_bt_idx_num,
- PF_RES_DATA_1_PF_QPC_TIMER_BT_NUM_M,
- PF_RES_DATA_1_PF_QPC_TIMER_BT_NUM_S);
+ roce_get_field(req_a->qpc_timer_bt_idx_num,
+ PF_RES_DATA_1_PF_QPC_TIMER_BT_NUM_M,
+ PF_RES_DATA_1_PF_QPC_TIMER_BT_NUM_S);
hr_dev->caps.cqc_timer_bt_num =
- roce_get_field(req_a->cqc_timer_bt_idx_num,
- PF_RES_DATA_2_PF_CQC_TIMER_BT_NUM_M,
- PF_RES_DATA_2_PF_CQC_TIMER_BT_NUM_S);
+ roce_get_field(req_a->cqc_timer_bt_idx_num,
+ PF_RES_DATA_2_PF_CQC_TIMER_BT_NUM_M,
+ PF_RES_DATA_2_PF_CQC_TIMER_BT_NUM_S);
return 0;
}
@@ -1786,6 +1881,9 @@ static int hns_roce_query_pf_caps(struct hns_roce_dev *hr_dev)
caps->flags = roce_get_field(resp_c->cap_flags_num_pds,
V2_QUERY_PF_CAPS_C_CAP_FLAGS_M,
V2_QUERY_PF_CAPS_C_CAP_FLAGS_S);
+ caps->flags |= le16_to_cpu(resp_d->cap_flags_ex) <<
+ HNS_ROCE_CAP_FLAGS_EX_SHIFT;
+
caps->num_cqs = 1 << roce_get_field(resp_c->max_gid_num_cqs,
V2_QUERY_PF_CAPS_C_NUM_CQS_M,
V2_QUERY_PF_CAPS_C_NUM_CQS_S);
@@ -1978,11 +2076,6 @@ static int hns_roce_v2_profile(struct hns_roce_dev *hr_dev)
hr_dev->vendor_part_id = hr_dev->pci_dev->device;
hr_dev->sys_image_guid = be64_to_cpu(hr_dev->ib_dev.node_guid);
- caps->num_mtt_segs = HNS_ROCE_V2_MAX_MTT_SEGS;
- caps->num_cqe_segs = HNS_ROCE_V2_MAX_CQE_SEGS;
- caps->num_srqwqe_segs = HNS_ROCE_V2_MAX_SRQWQE_SEGS;
- caps->num_idx_segs = HNS_ROCE_V2_MAX_IDX_SEGS;
-
caps->pbl_ba_pg_sz = HNS_ROCE_BA_PG_SZ_SUPPORTED_16K;
caps->pbl_buf_pg_sz = 0;
caps->pbl_hop_num = HNS_ROCE_PBL_HOP_NUM;
@@ -2040,8 +2133,6 @@ static int hns_roce_config_link_table(struct hns_roce_dev *hr_dev,
page_num = link_tbl->npages;
entry = link_tbl->table.buf;
- memset(req_a, 0, sizeof(*req_a));
- memset(req_b, 0, sizeof(*req_b));
for (i = 0; i < 2; i++) {
hns_roce_cmq_setup_basic_desc(&desc[i], opcode, false);
@@ -2050,39 +2141,30 @@ static int hns_roce_config_link_table(struct hns_roce_dev *hr_dev,
desc[i].flag |= cpu_to_le16(HNS_ROCE_CMD_FLAG_NEXT);
else
desc[i].flag &= ~cpu_to_le16(HNS_ROCE_CMD_FLAG_NEXT);
-
- if (i == 0) {
- req_a->base_addr_l =
- cpu_to_le32(link_tbl->table.map & 0xffffffff);
- req_a->base_addr_h =
- cpu_to_le32(link_tbl->table.map >> 32);
- roce_set_field(req_a->depth_pgsz_init_en,
- CFG_LLM_QUE_DEPTH_M, CFG_LLM_QUE_DEPTH_S,
- link_tbl->npages);
- roce_set_field(req_a->depth_pgsz_init_en,
- CFG_LLM_QUE_PGSZ_M, CFG_LLM_QUE_PGSZ_S,
- link_tbl->pg_sz);
- req_a->head_ba_l = cpu_to_le32(entry[0].blk_ba0);
- req_a->head_ba_h_nxtptr =
- cpu_to_le32(entry[0].blk_ba1_nxt_ptr);
- roce_set_field(req_a->head_ptr, CFG_LLM_HEAD_PTR_M,
- CFG_LLM_HEAD_PTR_S, 0);
- } else {
- req_b->tail_ba_l =
- cpu_to_le32(entry[page_num - 1].blk_ba0);
- roce_set_field(req_b->tail_ba_h, CFG_LLM_TAIL_BA_H_M,
- CFG_LLM_TAIL_BA_H_S,
- entry[page_num - 1].blk_ba1_nxt_ptr &
- HNS_ROCE_LINK_TABLE_BA1_M);
- roce_set_field(req_b->tail_ptr, CFG_LLM_TAIL_PTR_M,
- CFG_LLM_TAIL_PTR_S,
- (entry[page_num - 2].blk_ba1_nxt_ptr &
- HNS_ROCE_LINK_TABLE_NXT_PTR_M) >>
- HNS_ROCE_LINK_TABLE_NXT_PTR_S);
- }
}
+
+ req_a->base_addr_l = cpu_to_le32(link_tbl->table.map & 0xffffffff);
+ req_a->base_addr_h = cpu_to_le32(link_tbl->table.map >> 32);
+ roce_set_field(req_a->depth_pgsz_init_en, CFG_LLM_QUE_DEPTH_M,
+ CFG_LLM_QUE_DEPTH_S, link_tbl->npages);
+ roce_set_field(req_a->depth_pgsz_init_en, CFG_LLM_QUE_PGSZ_M,
+ CFG_LLM_QUE_PGSZ_S, link_tbl->pg_sz);
roce_set_field(req_a->depth_pgsz_init_en, CFG_LLM_INIT_EN_M,
CFG_LLM_INIT_EN_S, 1);
+ req_a->head_ba_l = cpu_to_le32(entry[0].blk_ba0);
+ req_a->head_ba_h_nxtptr = cpu_to_le32(entry[0].blk_ba1_nxt_ptr);
+ roce_set_field(req_a->head_ptr, CFG_LLM_HEAD_PTR_M, CFG_LLM_HEAD_PTR_S,
+ 0);
+
+ req_b->tail_ba_l = cpu_to_le32(entry[page_num - 1].blk_ba0);
+ roce_set_field(req_b->tail_ba_h, CFG_LLM_TAIL_BA_H_M,
+ CFG_LLM_TAIL_BA_H_S,
+ entry[page_num - 1].blk_ba1_nxt_ptr &
+ HNS_ROCE_LINK_TABLE_BA1_M);
+ roce_set_field(req_b->tail_ptr, CFG_LLM_TAIL_PTR_M, CFG_LLM_TAIL_PTR_S,
+ (entry[page_num - 2].blk_ba1_nxt_ptr &
+ HNS_ROCE_LINK_TABLE_NXT_PTR_M) >>
+ HNS_ROCE_LINK_TABLE_NXT_PTR_S);
return hns_roce_cmq_send(hr_dev, desc, 2);
}
@@ -2438,12 +2520,9 @@ static int hns_roce_v2_set_mac(struct hns_roce_dev *hr_dev, u8 phy_port,
reg_smac_l = *(u32 *)(&addr[0]);
reg_smac_h = *(u16 *)(&addr[4]);
- memset(smac_tb, 0, sizeof(*smac_tb));
- roce_set_field(smac_tb->tb_idx_rsv,
- CFG_SMAC_TB_IDX_M,
+ roce_set_field(smac_tb->tb_idx_rsv, CFG_SMAC_TB_IDX_M,
CFG_SMAC_TB_IDX_S, phy_port);
- roce_set_field(smac_tb->vf_smac_h_rsv,
- CFG_SMAC_TB_VF_SMAC_H_M,
+ roce_set_field(smac_tb->vf_smac_h_rsv, CFG_SMAC_TB_VF_SMAC_H_M,
CFG_SMAC_TB_VF_SMAC_H_S, reg_smac_h);
smac_tb->vf_smac_l = cpu_to_le32(reg_smac_l);
@@ -2453,32 +2532,30 @@ static int hns_roce_v2_set_mac(struct hns_roce_dev *hr_dev, u8 phy_port,
static int set_mtpt_pbl(struct hns_roce_v2_mpt_entry *mpt_entry,
struct hns_roce_mr *mr)
{
- struct sg_dma_page_iter sg_iter;
- u64 page_addr;
- u64 *pages;
- int i;
+ struct hns_roce_dev *hr_dev = to_hr_dev(mr->ibmr.device);
+ u64 pages[HNS_ROCE_V2_MAX_INNER_MTPT_NUM] = { 0 };
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ dma_addr_t pbl_ba;
+ int i, count;
- mpt_entry->pbl_size = cpu_to_le32(mr->pbl_size);
- mpt_entry->pbl_ba_l = cpu_to_le32(lower_32_bits(mr->pbl_ba >> 3));
- roce_set_field(mpt_entry->byte_48_mode_ba,
- V2_MPT_BYTE_48_PBL_BA_H_M, V2_MPT_BYTE_48_PBL_BA_H_S,
- upper_32_bits(mr->pbl_ba >> 3));
+ count = hns_roce_mtr_find(hr_dev, &mr->pbl_mtr, 0, pages,
+ ARRAY_SIZE(pages), &pbl_ba);
+ if (count < 1) {
+ ibdev_err(ibdev, "failed to find PBL mtr, count = %d.\n",
+ count);
+ return -ENOBUFS;
+ }
- pages = (u64 *)__get_free_page(GFP_KERNEL);
- if (!pages)
- return -ENOMEM;
+ /* Aligned to the hardware address access unit */
+ for (i = 0; i < count; i++)
+ pages[i] >>= 6;
- i = 0;
- for_each_sg_dma_page(mr->umem->sg_head.sgl, &sg_iter, mr->umem->nmap, 0) {
- page_addr = sg_page_iter_dma_address(&sg_iter);
- pages[i] = page_addr >> 6;
+ mpt_entry->pbl_size = cpu_to_le32(mr->npages);
+ mpt_entry->pbl_ba_l = cpu_to_le32(pbl_ba >> 3);
+ roce_set_field(mpt_entry->byte_48_mode_ba,
+ V2_MPT_BYTE_48_PBL_BA_H_M, V2_MPT_BYTE_48_PBL_BA_H_S,
+ upper_32_bits(pbl_ba >> 3));
- /* Record the first 2 entry directly to MTPT table */
- if (i >= HNS_ROCE_V2_MAX_INNER_MTPT_NUM - 1)
- goto found;
- i++;
- }
-found:
mpt_entry->pa0_l = cpu_to_le32(lower_32_bits(pages[0]));
roce_set_field(mpt_entry->byte_56_pa0_h, V2_MPT_BYTE_56_PA0_H_M,
V2_MPT_BYTE_56_PA0_H_S, upper_32_bits(pages[0]));
@@ -2489,9 +2566,7 @@ found:
roce_set_field(mpt_entry->byte_64_buf_pa1,
V2_MPT_BYTE_64_PBL_BUF_PG_SZ_M,
V2_MPT_BYTE_64_PBL_BUF_PG_SZ_S,
- mr->pbl_buf_pg_sz + PG_SHIFT_OFFSET);
-
- free_page((unsigned long)pages);
+ to_hr_hw_page_shift(mr->pbl_mtr.hem_cfg.buf_pg_shift));
return 0;
}
@@ -2513,7 +2588,7 @@ static int hns_roce_v2_write_mtpt(void *mb_buf, struct hns_roce_mr *mr,
roce_set_field(mpt_entry->byte_4_pd_hop_st,
V2_MPT_BYTE_4_PBL_BA_PG_SZ_M,
V2_MPT_BYTE_4_PBL_BA_PG_SZ_S,
- mr->pbl_ba_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(mr->pbl_mtr.hem_cfg.ba_pg_shift));
roce_set_field(mpt_entry->byte_4_pd_hop_st, V2_MPT_BYTE_4_PD_M,
V2_MPT_BYTE_4_PD_S, mr->pd);
@@ -2599,11 +2674,19 @@ static int hns_roce_v2_rereg_write_mtpt(struct hns_roce_dev *hr_dev,
static int hns_roce_v2_frmr_write_mtpt(void *mb_buf, struct hns_roce_mr *mr)
{
+ struct hns_roce_dev *hr_dev = to_hr_dev(mr->ibmr.device);
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct hns_roce_v2_mpt_entry *mpt_entry;
+ dma_addr_t pbl_ba = 0;
mpt_entry = mb_buf;
memset(mpt_entry, 0, sizeof(*mpt_entry));
+ if (hns_roce_mtr_find(hr_dev, &mr->pbl_mtr, 0, NULL, 0, &pbl_ba) < 0) {
+ ibdev_err(ibdev, "failed to find frmr mtr.\n");
+ return -ENOBUFS;
+ }
+
roce_set_field(mpt_entry->byte_4_pd_hop_st, V2_MPT_BYTE_4_MPT_ST_M,
V2_MPT_BYTE_4_MPT_ST_S, V2_MPT_ST_FREE);
roce_set_field(mpt_entry->byte_4_pd_hop_st, V2_MPT_BYTE_4_PBL_HOP_NUM_M,
@@ -2611,7 +2694,7 @@ static int hns_roce_v2_frmr_write_mtpt(void *mb_buf, struct hns_roce_mr *mr)
roce_set_field(mpt_entry->byte_4_pd_hop_st,
V2_MPT_BYTE_4_PBL_BA_PG_SZ_M,
V2_MPT_BYTE_4_PBL_BA_PG_SZ_S,
- mr->pbl_ba_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(mr->pbl_mtr.hem_cfg.ba_pg_shift));
roce_set_field(mpt_entry->byte_4_pd_hop_st, V2_MPT_BYTE_4_PD_M,
V2_MPT_BYTE_4_PD_S, mr->pd);
@@ -2624,17 +2707,17 @@ static int hns_roce_v2_frmr_write_mtpt(void *mb_buf, struct hns_roce_mr *mr)
roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_MR_MW_S, 0);
roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_BPD_S, 1);
- mpt_entry->pbl_size = cpu_to_le32(mr->pbl_size);
+ mpt_entry->pbl_size = cpu_to_le32(mr->npages);
- mpt_entry->pbl_ba_l = cpu_to_le32(lower_32_bits(mr->pbl_ba >> 3));
+ mpt_entry->pbl_ba_l = cpu_to_le32(lower_32_bits(pbl_ba >> 3));
roce_set_field(mpt_entry->byte_48_mode_ba, V2_MPT_BYTE_48_PBL_BA_H_M,
V2_MPT_BYTE_48_PBL_BA_H_S,
- upper_32_bits(mr->pbl_ba >> 3));
+ upper_32_bits(pbl_ba >> 3));
roce_set_field(mpt_entry->byte_64_buf_pa1,
V2_MPT_BYTE_64_PBL_BUF_PG_SZ_M,
V2_MPT_BYTE_64_PBL_BUF_PG_SZ_S,
- mr->pbl_buf_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(mr->pbl_mtr.hem_cfg.buf_pg_shift));
return 0;
}
@@ -2680,7 +2763,8 @@ static int hns_roce_v2_mw_write_mtpt(void *mb_buf, struct hns_roce_mw *mw)
static void *get_cqe_v2(struct hns_roce_cq *hr_cq, int n)
{
- return hns_roce_buf_offset(&hr_cq->buf, n * HNS_ROCE_V2_CQE_ENTRY_SIZE);
+ return hns_roce_buf_offset(hr_cq->mtr.kmem,
+ n * HNS_ROCE_V2_CQE_ENTRY_SIZE);
}
static void *get_sw_cqe_v2(struct hns_roce_cq *hr_cq, int n)
@@ -2692,30 +2776,9 @@ static void *get_sw_cqe_v2(struct hns_roce_cq *hr_cq, int n)
!!(n & hr_cq->cq_depth)) ? cqe : NULL;
}
-static struct hns_roce_v2_cqe *next_cqe_sw_v2(struct hns_roce_cq *hr_cq)
+static inline void hns_roce_v2_cq_set_ci(struct hns_roce_cq *hr_cq, u32 ci)
{
- return get_sw_cqe_v2(hr_cq, hr_cq->cons_index);
-}
-
-static void *get_srq_wqe(struct hns_roce_srq *srq, int n)
-{
- return hns_roce_buf_offset(&srq->buf, n << srq->wqe_shift);
-}
-
-static void hns_roce_free_srq_wqe(struct hns_roce_srq *srq, int wqe_index)
-{
- /* always called with interrupts disabled. */
- spin_lock(&srq->lock);
-
- bitmap_clear(srq->idx_que.bitmap, wqe_index, 1);
- srq->tail++;
-
- spin_unlock(&srq->lock);
-}
-
-static void hns_roce_v2_cq_set_ci(struct hns_roce_cq *hr_cq, u32 cons_index)
-{
- *hr_cq->set_ci_db = cons_index & V2_CQ_DB_PARAMETER_CONS_IDX_M;
+ *hr_cq->set_ci_db = ci & V2_CQ_DB_PARAMETER_CONS_IDX_M;
}
static void __hns_roce_v2_cq_clean(struct hns_roce_cq *hr_cq, u32 qpn,
@@ -2801,39 +2864,39 @@ static void hns_roce_v2_write_cqc(struct hns_roce_dev *hr_dev,
roce_set_field(cq_context->byte_8_cqn, V2_CQC_BYTE_8_CQN_M,
V2_CQC_BYTE_8_CQN_S, hr_cq->cqn);
- cq_context->cqe_cur_blk_addr = cpu_to_le32(mtts[0] >> PAGE_ADDR_SHIFT);
+ cq_context->cqe_cur_blk_addr = cpu_to_le32(to_hr_hw_page_addr(mtts[0]));
roce_set_field(cq_context->byte_16_hop_addr,
V2_CQC_BYTE_16_CQE_CUR_BLK_ADDR_M,
V2_CQC_BYTE_16_CQE_CUR_BLK_ADDR_S,
- mtts[0] >> (32 + PAGE_ADDR_SHIFT));
+ upper_32_bits(to_hr_hw_page_addr(mtts[0])));
roce_set_field(cq_context->byte_16_hop_addr,
V2_CQC_BYTE_16_CQE_HOP_NUM_M,
V2_CQC_BYTE_16_CQE_HOP_NUM_S, hr_dev->caps.cqe_hop_num ==
HNS_ROCE_HOP_NUM_0 ? 0 : hr_dev->caps.cqe_hop_num);
- cq_context->cqe_nxt_blk_addr = cpu_to_le32(mtts[1] >> PAGE_ADDR_SHIFT);
+ cq_context->cqe_nxt_blk_addr = cpu_to_le32(to_hr_hw_page_addr(mtts[1]));
roce_set_field(cq_context->byte_24_pgsz_addr,
V2_CQC_BYTE_24_CQE_NXT_BLK_ADDR_M,
V2_CQC_BYTE_24_CQE_NXT_BLK_ADDR_S,
- mtts[1] >> (32 + PAGE_ADDR_SHIFT));
+ upper_32_bits(to_hr_hw_page_addr(mtts[1])));
roce_set_field(cq_context->byte_24_pgsz_addr,
V2_CQC_BYTE_24_CQE_BA_PG_SZ_M,
V2_CQC_BYTE_24_CQE_BA_PG_SZ_S,
- hr_dev->caps.cqe_ba_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(hr_cq->mtr.hem_cfg.ba_pg_shift));
roce_set_field(cq_context->byte_24_pgsz_addr,
V2_CQC_BYTE_24_CQE_BUF_PG_SZ_M,
V2_CQC_BYTE_24_CQE_BUF_PG_SZ_S,
- hr_dev->caps.cqe_buf_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(hr_cq->mtr.hem_cfg.buf_pg_shift));
cq_context->cqe_ba = cpu_to_le32(dma_handle >> 3);
roce_set_field(cq_context->byte_40_cqe_ba, V2_CQC_BYTE_40_CQE_BA_M,
V2_CQC_BYTE_40_CQE_BA_S, (dma_handle >> (32 + 3)));
- if (hr_cq->db_en)
- roce_set_bit(cq_context->byte_44_db_record,
- V2_CQC_BYTE_44_DB_RECORD_EN_S, 1);
+ roce_set_bit(cq_context->byte_44_db_record,
+ V2_CQC_BYTE_44_DB_RECORD_EN_S,
+ (hr_cq->flags & HNS_ROCE_CQ_FLAG_RECORD_DB) ? 1 : 0);
roce_set_field(cq_context->byte_44_db_record,
V2_CQC_BYTE_44_DB_RECORD_ADDR_M,
@@ -2873,8 +2936,7 @@ static int hns_roce_v2_req_notify_cq(struct ib_cq *ibcq,
roce_set_field(doorbell[0], V2_CQ_DB_BYTE_4_CMD_M, V2_DB_BYTE_4_CMD_S,
HNS_ROCE_V2_CQ_DB_NTR);
roce_set_field(doorbell[1], V2_CQ_DB_PARAMETER_CONS_IDX_M,
- V2_CQ_DB_PARAMETER_CONS_IDX_S,
- hr_cq->cons_index & ((hr_cq->cq_depth << 1) - 1));
+ V2_CQ_DB_PARAMETER_CONS_IDX_S, hr_cq->cons_index);
roce_set_field(doorbell[1], V2_CQ_DB_PARAMETER_CMD_SN_M,
V2_CQ_DB_PARAMETER_CMD_SN_S, hr_cq->arm_sn & 0x3);
roce_set_bit(doorbell[1], V2_CQ_DB_PARAMETER_NOTIFY_S,
@@ -2911,7 +2973,7 @@ static int hns_roce_handle_recv_inl_wqe(struct hns_roce_v2_cqe *cqe,
wqe_buf += size;
}
- if (data_len) {
+ if (unlikely(data_len)) {
wc->status = IB_WC_LOC_LEN_ERR;
return -EAGAIN;
}
@@ -2968,6 +3030,62 @@ out:
return npolled;
}
+static void get_cqe_status(struct hns_roce_dev *hr_dev, struct hns_roce_qp *qp,
+ struct hns_roce_v2_cqe *cqe, struct ib_wc *wc)
+{
+ static const struct {
+ u32 cqe_status;
+ enum ib_wc_status wc_status;
+ } map[] = {
+ { HNS_ROCE_CQE_V2_SUCCESS, IB_WC_SUCCESS },
+ { HNS_ROCE_CQE_V2_LOCAL_LENGTH_ERR, IB_WC_LOC_LEN_ERR },
+ { HNS_ROCE_CQE_V2_LOCAL_QP_OP_ERR, IB_WC_LOC_QP_OP_ERR },
+ { HNS_ROCE_CQE_V2_LOCAL_PROT_ERR, IB_WC_LOC_PROT_ERR },
+ { HNS_ROCE_CQE_V2_WR_FLUSH_ERR, IB_WC_WR_FLUSH_ERR },
+ { HNS_ROCE_CQE_V2_MW_BIND_ERR, IB_WC_MW_BIND_ERR },
+ { HNS_ROCE_CQE_V2_BAD_RESP_ERR, IB_WC_BAD_RESP_ERR },
+ { HNS_ROCE_CQE_V2_LOCAL_ACCESS_ERR, IB_WC_LOC_ACCESS_ERR },
+ { HNS_ROCE_CQE_V2_REMOTE_INVAL_REQ_ERR, IB_WC_REM_INV_REQ_ERR },
+ { HNS_ROCE_CQE_V2_REMOTE_ACCESS_ERR, IB_WC_REM_ACCESS_ERR },
+ { HNS_ROCE_CQE_V2_REMOTE_OP_ERR, IB_WC_REM_OP_ERR },
+ { HNS_ROCE_CQE_V2_TRANSPORT_RETRY_EXC_ERR,
+ IB_WC_RETRY_EXC_ERR },
+ { HNS_ROCE_CQE_V2_RNR_RETRY_EXC_ERR, IB_WC_RNR_RETRY_EXC_ERR },
+ { HNS_ROCE_CQE_V2_REMOTE_ABORT_ERR, IB_WC_REM_ABORT_ERR },
+ };
+
+ u32 cqe_status = roce_get_field(cqe->byte_4, V2_CQE_BYTE_4_STATUS_M,
+ V2_CQE_BYTE_4_STATUS_S);
+ int i;
+
+ wc->status = IB_WC_GENERAL_ERR;
+ for (i = 0; i < ARRAY_SIZE(map); i++)
+ if (cqe_status == map[i].cqe_status) {
+ wc->status = map[i].wc_status;
+ break;
+ }
+
+ if (likely(wc->status == IB_WC_SUCCESS ||
+ wc->status == IB_WC_WR_FLUSH_ERR))
+ return;
+
+ ibdev_err(&hr_dev->ib_dev, "error cqe status 0x%x:\n", cqe_status);
+ print_hex_dump(KERN_ERR, "", DUMP_PREFIX_NONE, 16, 4, cqe,
+ sizeof(*cqe), false);
+
+ /*
+ * Hip08 hardware cannot flush the WQEs in SQ/RQ if the QP state gets
+ * into errored mode. Hence, as a workaround to this hardware
+ * limitation, driver needs to assist in flushing. But the flushing
+ * operation uses mailbox to convey the QP state to the hardware and
+ * which can sleep due to the mutex protection around the mailbox calls.
+ * Hence, use the deferred flush for now. Once wc error detected, the
+ * flushing operation is needed.
+ */
+ if (!test_and_set_bit(HNS_ROCE_FLUSH_FLAG, &qp->flush_flag))
+ init_flush_work(hr_dev, qp);
+}
+
static int hns_roce_v2_poll_one(struct hns_roce_cq *hr_cq,
struct hns_roce_qp **cur_qp, struct ib_wc *wc)
{
@@ -2979,12 +3097,11 @@ static int hns_roce_v2_poll_one(struct hns_roce_cq *hr_cq,
int is_send;
u16 wqe_ctr;
u32 opcode;
- u32 status;
int qpn;
int ret;
/* Find cqe according to consumer index */
- cqe = next_cqe_sw_v2(hr_cq);
+ cqe = get_sw_cqe_v2(hr_cq, hr_cq->cons_index);
if (!cqe)
return -EAGAIN;
@@ -3009,7 +3126,6 @@ static int hns_roce_v2_poll_one(struct hns_roce_cq *hr_cq,
*cur_qp = hr_qp;
}
- hr_qp = *cur_qp;
wc->qp = &(*cur_qp)->ibqp;
wc->vendor_err = 0;
@@ -3044,77 +3160,8 @@ static int hns_roce_v2_poll_one(struct hns_roce_cq *hr_cq,
++wq->tail;
}
- status = roce_get_field(cqe->byte_4, V2_CQE_BYTE_4_STATUS_M,
- V2_CQE_BYTE_4_STATUS_S);
- switch (status & HNS_ROCE_V2_CQE_STATUS_MASK) {
- case HNS_ROCE_CQE_V2_SUCCESS:
- wc->status = IB_WC_SUCCESS;
- break;
- case HNS_ROCE_CQE_V2_LOCAL_LENGTH_ERR:
- wc->status = IB_WC_LOC_LEN_ERR;
- break;
- case HNS_ROCE_CQE_V2_LOCAL_QP_OP_ERR:
- wc->status = IB_WC_LOC_QP_OP_ERR;
- break;
- case HNS_ROCE_CQE_V2_LOCAL_PROT_ERR:
- wc->status = IB_WC_LOC_PROT_ERR;
- break;
- case HNS_ROCE_CQE_V2_WR_FLUSH_ERR:
- wc->status = IB_WC_WR_FLUSH_ERR;
- break;
- case HNS_ROCE_CQE_V2_MW_BIND_ERR:
- wc->status = IB_WC_MW_BIND_ERR;
- break;
- case HNS_ROCE_CQE_V2_BAD_RESP_ERR:
- wc->status = IB_WC_BAD_RESP_ERR;
- break;
- case HNS_ROCE_CQE_V2_LOCAL_ACCESS_ERR:
- wc->status = IB_WC_LOC_ACCESS_ERR;
- break;
- case HNS_ROCE_CQE_V2_REMOTE_INVAL_REQ_ERR:
- wc->status = IB_WC_REM_INV_REQ_ERR;
- break;
- case HNS_ROCE_CQE_V2_REMOTE_ACCESS_ERR:
- wc->status = IB_WC_REM_ACCESS_ERR;
- break;
- case HNS_ROCE_CQE_V2_REMOTE_OP_ERR:
- wc->status = IB_WC_REM_OP_ERR;
- break;
- case HNS_ROCE_CQE_V2_TRANSPORT_RETRY_EXC_ERR:
- wc->status = IB_WC_RETRY_EXC_ERR;
- break;
- case HNS_ROCE_CQE_V2_RNR_RETRY_EXC_ERR:
- wc->status = IB_WC_RNR_RETRY_EXC_ERR;
- break;
- case HNS_ROCE_CQE_V2_REMOTE_ABORT_ERR:
- wc->status = IB_WC_REM_ABORT_ERR;
- break;
- default:
- wc->status = IB_WC_GENERAL_ERR;
- break;
- }
-
- /*
- * Hip08 hardware cannot flush the WQEs in SQ/RQ if the QP state gets
- * into errored mode. Hence, as a workaround to this hardware
- * limitation, driver needs to assist in flushing. But the flushing
- * operation uses mailbox to convey the QP state to the hardware and
- * which can sleep due to the mutex protection around the mailbox calls.
- * Hence, use the deferred flush for now. Once wc error detected, the
- * flushing operation is needed.
- */
- if (wc->status != IB_WC_SUCCESS &&
- wc->status != IB_WC_WR_FLUSH_ERR) {
- ibdev_err(&hr_dev->ib_dev, "error cqe status is: 0x%x\n",
- status & HNS_ROCE_V2_CQE_STATUS_MASK);
-
- if (!test_and_set_bit(HNS_ROCE_FLUSH_FLAG, &hr_qp->flush_flag))
- init_flush_work(hr_dev, hr_qp);
-
- return 0;
- }
-
- if (wc->status == IB_WC_WR_FLUSH_ERR)
+ get_cqe_status(hr_dev, *cur_qp, cqe, wc);
+ if (unlikely(wc->status != IB_WC_SUCCESS))
return 0;
if (is_send) {
@@ -3213,7 +3260,7 @@ static int hns_roce_v2_poll_one(struct hns_roce_cq *hr_cq,
opcode == HNS_ROCE_V2_OPCODE_SEND_WITH_INV) &&
(roce_get_bit(cqe->byte_4, V2_CQE_BYTE_4_RQ_INLINE_S))) {
ret = hns_roce_handle_recv_inl_wqe(cqe, cur_qp, wc);
- if (ret)
+ if (unlikely(ret))
return -EAGAIN;
}
@@ -3514,29 +3561,18 @@ static void set_qpc_wqe_cnt(struct hns_roce_qp *hr_qp,
struct hns_roce_v2_qp_context *context,
struct hns_roce_v2_qp_context *qpc_mask)
{
- if (hr_qp->ibqp.qp_type == IB_QPT_GSI)
- roce_set_field(context->byte_4_sqpn_tst,
- V2_QPC_BYTE_4_SGE_SHIFT_M,
- V2_QPC_BYTE_4_SGE_SHIFT_S,
- ilog2((unsigned int)hr_qp->sge.sge_cnt));
- else
- roce_set_field(context->byte_4_sqpn_tst,
- V2_QPC_BYTE_4_SGE_SHIFT_M,
- V2_QPC_BYTE_4_SGE_SHIFT_S,
- hr_qp->sq.max_gs >
- HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE ?
- ilog2((unsigned int)hr_qp->sge.sge_cnt) : 0);
+ roce_set_field(context->byte_4_sqpn_tst,
+ V2_QPC_BYTE_4_SGE_SHIFT_M, V2_QPC_BYTE_4_SGE_SHIFT_S,
+ to_hr_hem_entries_shift(hr_qp->sge.sge_cnt,
+ hr_qp->sge.sge_shift));
roce_set_field(context->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_SQ_SHIFT_M, V2_QPC_BYTE_20_SQ_SHIFT_S,
- ilog2((unsigned int)hr_qp->sq.wqe_cnt));
+ ilog2(hr_qp->sq.wqe_cnt));
roce_set_field(context->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_RQ_SHIFT_M, V2_QPC_BYTE_20_RQ_SHIFT_S,
- (hr_qp->ibqp.qp_type == IB_QPT_XRC_INI ||
- hr_qp->ibqp.qp_type == IB_QPT_XRC_TGT ||
- hr_qp->ibqp.srq) ? 0 :
- ilog2((unsigned int)hr_qp->rq.wqe_cnt));
+ ilog2(hr_qp->rq.wqe_cnt));
}
static void modify_qp_reset_to_init(struct ib_qp *ibqp,
@@ -3572,7 +3608,7 @@ static void modify_qp_reset_to_init(struct ib_qp *ibqp,
roce_set_field(context->byte_24_mtu_tc, V2_QPC_BYTE_24_VLAN_ID_M,
V2_QPC_BYTE_24_VLAN_ID_S, 0xfff);
- if (hr_qp->rdb_en)
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
roce_set_bit(context->byte_68_rq_db,
V2_QPC_BYTE_68_RQ_RECORD_EN_S, 1);
@@ -3734,30 +3770,19 @@ static bool check_wqe_rq_mtt_count(struct hns_roce_dev *hr_dev,
return true;
}
-static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
- const struct ib_qp_attr *attr, int attr_mask,
- struct hns_roce_v2_qp_context *context,
- struct hns_roce_v2_qp_context *qpc_mask)
+static int config_qp_rq_buf(struct hns_roce_dev *hr_dev,
+ struct hns_roce_qp *hr_qp,
+ struct hns_roce_v2_qp_context *context,
+ struct hns_roce_v2_qp_context *qpc_mask)
{
- const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr);
- struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
- struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
- struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct ib_qp *ibqp = &hr_qp->ibqp;
u64 mtts[MTT_MIN_COUNT] = { 0 };
- dma_addr_t dma_handle_3;
- dma_addr_t dma_handle_2;
u64 wqe_sge_ba;
u32 page_size;
- u8 port_num;
- u64 *mtts_3;
- u64 *mtts_2;
int count;
- u8 *dmac;
- u8 *smac;
- int port;
/* Search qp buf's mtts */
- page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
+ page_size = 1 << hr_qp->mtr.hem_cfg.buf_pg_shift;
count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr,
hr_qp->rq.offset / page_size, mtts,
MTT_MIN_COUNT, &wqe_sge_ba);
@@ -3765,29 +3790,6 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
if (!check_wqe_rq_mtt_count(hr_dev, hr_qp, count, page_size))
return -EINVAL;
- /* Search IRRL's mtts */
- mtts_2 = hns_roce_table_find(hr_dev, &hr_dev->qp_table.irrl_table,
- hr_qp->qpn, &dma_handle_2);
- if (!mtts_2) {
- ibdev_err(ibdev, "failed to find QP irrl_table\n");
- return -EINVAL;
- }
-
- /* Search TRRL's mtts */
- mtts_3 = hns_roce_table_find(hr_dev, &hr_dev->qp_table.trrl_table,
- hr_qp->qpn, &dma_handle_3);
- if (!mtts_3) {
- ibdev_err(ibdev, "failed to find QP trrl_table\n");
- return -EINVAL;
- }
-
- if (attr_mask & IB_QP_ALT_PATH) {
- ibdev_err(ibdev, "INIT2RTR attr_mask (0x%x) error\n",
- attr_mask);
- return -EINVAL;
- }
-
- dmac = (u8 *)attr->ah_attr.roce.dmac;
context->wqe_sge_ba = cpu_to_le32(wqe_sge_ba >> 3);
qpc_mask->wqe_sge_ba = 0;
@@ -3804,17 +3806,16 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
roce_set_field(context->byte_12_sq_hop, V2_QPC_BYTE_12_SQ_HOP_NUM_M,
V2_QPC_BYTE_12_SQ_HOP_NUM_S,
- hr_dev->caps.wqe_sq_hop_num == HNS_ROCE_HOP_NUM_0 ?
- 0 : hr_dev->caps.wqe_sq_hop_num);
+ to_hr_hem_hopnum(hr_dev->caps.wqe_sq_hop_num,
+ hr_qp->sq.wqe_cnt));
roce_set_field(qpc_mask->byte_12_sq_hop, V2_QPC_BYTE_12_SQ_HOP_NUM_M,
V2_QPC_BYTE_12_SQ_HOP_NUM_S, 0);
roce_set_field(context->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_SGE_HOP_NUM_M,
V2_QPC_BYTE_20_SGE_HOP_NUM_S,
- ((ibqp->qp_type == IB_QPT_GSI) ||
- hr_qp->sq.max_gs > HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE) ?
- hr_dev->caps.wqe_sge_hop_num : 0);
+ to_hr_hem_hopnum(hr_dev->caps.wqe_sge_hop_num,
+ hr_qp->sge.sge_cnt));
roce_set_field(qpc_mask->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_SGE_HOP_NUM_M,
V2_QPC_BYTE_20_SGE_HOP_NUM_S, 0);
@@ -3822,8 +3823,9 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
roce_set_field(context->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_RQ_HOP_NUM_M,
V2_QPC_BYTE_20_RQ_HOP_NUM_S,
- hr_dev->caps.wqe_rq_hop_num == HNS_ROCE_HOP_NUM_0 ?
- 0 : hr_dev->caps.wqe_rq_hop_num);
+ to_hr_hem_hopnum(hr_dev->caps.wqe_rq_hop_num,
+ hr_qp->rq.wqe_cnt));
+
roce_set_field(qpc_mask->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_RQ_HOP_NUM_M,
V2_QPC_BYTE_20_RQ_HOP_NUM_S, 0);
@@ -3831,7 +3833,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
roce_set_field(context->byte_16_buf_ba_pg_sz,
V2_QPC_BYTE_16_WQE_SGE_BA_PG_SZ_M,
V2_QPC_BYTE_16_WQE_SGE_BA_PG_SZ_S,
- hr_qp->wqe_bt_pg_shift + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(hr_qp->mtr.hem_cfg.ba_pg_shift));
roce_set_field(qpc_mask->byte_16_buf_ba_pg_sz,
V2_QPC_BYTE_16_WQE_SGE_BA_PG_SZ_M,
V2_QPC_BYTE_16_WQE_SGE_BA_PG_SZ_S, 0);
@@ -3839,50 +3841,181 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
roce_set_field(context->byte_16_buf_ba_pg_sz,
V2_QPC_BYTE_16_WQE_SGE_BUF_PG_SZ_M,
V2_QPC_BYTE_16_WQE_SGE_BUF_PG_SZ_S,
- hr_dev->caps.mtt_buf_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(hr_qp->mtr.hem_cfg.buf_pg_shift));
roce_set_field(qpc_mask->byte_16_buf_ba_pg_sz,
V2_QPC_BYTE_16_WQE_SGE_BUF_PG_SZ_M,
V2_QPC_BYTE_16_WQE_SGE_BUF_PG_SZ_S, 0);
- context->rq_cur_blk_addr = cpu_to_le32(mtts[0] >> PAGE_ADDR_SHIFT);
+ context->rq_cur_blk_addr = cpu_to_le32(to_hr_hw_page_addr(mtts[0]));
qpc_mask->rq_cur_blk_addr = 0;
roce_set_field(context->byte_92_srq_info,
V2_QPC_BYTE_92_RQ_CUR_BLK_ADDR_M,
V2_QPC_BYTE_92_RQ_CUR_BLK_ADDR_S,
- mtts[0] >> (32 + PAGE_ADDR_SHIFT));
+ upper_32_bits(to_hr_hw_page_addr(mtts[0])));
roce_set_field(qpc_mask->byte_92_srq_info,
V2_QPC_BYTE_92_RQ_CUR_BLK_ADDR_M,
V2_QPC_BYTE_92_RQ_CUR_BLK_ADDR_S, 0);
- context->rq_nxt_blk_addr = cpu_to_le32(mtts[1] >> PAGE_ADDR_SHIFT);
+ context->rq_nxt_blk_addr = cpu_to_le32(to_hr_hw_page_addr(mtts[1]));
qpc_mask->rq_nxt_blk_addr = 0;
roce_set_field(context->byte_104_rq_sge,
V2_QPC_BYTE_104_RQ_NXT_BLK_ADDR_M,
V2_QPC_BYTE_104_RQ_NXT_BLK_ADDR_S,
- mtts[1] >> (32 + PAGE_ADDR_SHIFT));
+ upper_32_bits(to_hr_hw_page_addr(mtts[1])));
roce_set_field(qpc_mask->byte_104_rq_sge,
V2_QPC_BYTE_104_RQ_NXT_BLK_ADDR_M,
V2_QPC_BYTE_104_RQ_NXT_BLK_ADDR_S, 0);
+ roce_set_field(context->byte_84_rq_ci_pi,
+ V2_QPC_BYTE_84_RQ_PRODUCER_IDX_M,
+ V2_QPC_BYTE_84_RQ_PRODUCER_IDX_S, hr_qp->rq.head);
+ roce_set_field(qpc_mask->byte_84_rq_ci_pi,
+ V2_QPC_BYTE_84_RQ_PRODUCER_IDX_M,
+ V2_QPC_BYTE_84_RQ_PRODUCER_IDX_S, 0);
+
+ roce_set_field(qpc_mask->byte_84_rq_ci_pi,
+ V2_QPC_BYTE_84_RQ_CONSUMER_IDX_M,
+ V2_QPC_BYTE_84_RQ_CONSUMER_IDX_S, 0);
+
+ return 0;
+}
+
+static int config_qp_sq_buf(struct hns_roce_dev *hr_dev,
+ struct hns_roce_qp *hr_qp,
+ struct hns_roce_v2_qp_context *context,
+ struct hns_roce_v2_qp_context *qpc_mask)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ u64 sge_cur_blk = 0;
+ u64 sq_cur_blk = 0;
+ u32 page_size;
+ int count;
+
+ /* search qp buf's mtts */
+ count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr, 0, &sq_cur_blk, 1, NULL);
+ if (count < 1) {
+ ibdev_err(ibdev, "failed to find QP(0x%lx) SQ buf.\n",
+ hr_qp->qpn);
+ return -EINVAL;
+ }
+ if (hr_qp->sge.sge_cnt > 0) {
+ page_size = 1 << hr_qp->mtr.hem_cfg.buf_pg_shift;
+ count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr,
+ hr_qp->sge.offset / page_size,
+ &sge_cur_blk, 1, NULL);
+ if (count < 1) {
+ ibdev_err(ibdev, "failed to find QP(0x%lx) SGE buf.\n",
+ hr_qp->qpn);
+ return -EINVAL;
+ }
+ }
+
+ /*
+ * In v2 engine, software pass context and context mask to hardware
+ * when modifying qp. If software need modify some fields in context,
+ * we should set all bits of the relevant fields in context mask to
+ * 0 at the same time, else set them to 0x1.
+ */
+ context->sq_cur_blk_addr = cpu_to_le32(to_hr_hw_page_addr(sq_cur_blk));
+ roce_set_field(context->byte_168_irrl_idx,
+ V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_M,
+ V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_S,
+ upper_32_bits(to_hr_hw_page_addr(sq_cur_blk)));
+ qpc_mask->sq_cur_blk_addr = 0;
+ roce_set_field(qpc_mask->byte_168_irrl_idx,
+ V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_M,
+ V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_S, 0);
+
+ context->sq_cur_sge_blk_addr =
+ cpu_to_le32(to_hr_hw_page_addr(sge_cur_blk));
+ roce_set_field(context->byte_184_irrl_idx,
+ V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_M,
+ V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_S,
+ upper_32_bits(to_hr_hw_page_addr(sge_cur_blk)));
+ qpc_mask->sq_cur_sge_blk_addr = 0;
+ roce_set_field(qpc_mask->byte_184_irrl_idx,
+ V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_M,
+ V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_S, 0);
+
+ context->rx_sq_cur_blk_addr =
+ cpu_to_le32(to_hr_hw_page_addr(sq_cur_blk));
+ roce_set_field(context->byte_232_irrl_sge,
+ V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_M,
+ V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_S,
+ upper_32_bits(to_hr_hw_page_addr(sq_cur_blk)));
+ qpc_mask->rx_sq_cur_blk_addr = 0;
+ roce_set_field(qpc_mask->byte_232_irrl_sge,
+ V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_M,
+ V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_S, 0);
+
+ return 0;
+}
+
+static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
+ const struct ib_qp_attr *attr, int attr_mask,
+ struct hns_roce_v2_qp_context *context,
+ struct hns_roce_v2_qp_context *qpc_mask)
+{
+ const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr);
+ struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
+ struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ dma_addr_t trrl_ba;
+ dma_addr_t irrl_ba;
+ u8 port_num;
+ u64 *mtts;
+ u8 *dmac;
+ u8 *smac;
+ int port;
+ int ret;
+
+ ret = config_qp_rq_buf(hr_dev, hr_qp, context, qpc_mask);
+ if (ret) {
+ ibdev_err(ibdev, "failed to config rq buf, ret = %d.\n", ret);
+ return ret;
+ }
+
+ /* Search IRRL's mtts */
+ mtts = hns_roce_table_find(hr_dev, &hr_dev->qp_table.irrl_table,
+ hr_qp->qpn, &irrl_ba);
+ if (!mtts) {
+ ibdev_err(ibdev, "failed to find qp irrl_table.\n");
+ return -EINVAL;
+ }
+
+ /* Search TRRL's mtts */
+ mtts = hns_roce_table_find(hr_dev, &hr_dev->qp_table.trrl_table,
+ hr_qp->qpn, &trrl_ba);
+ if (!mtts) {
+ ibdev_err(ibdev, "failed to find qp trrl_table.\n");
+ return -EINVAL;
+ }
+
+ if (attr_mask & IB_QP_ALT_PATH) {
+ ibdev_err(ibdev, "INIT2RTR attr_mask (0x%x) error.\n",
+ attr_mask);
+ return -EINVAL;
+ }
+
roce_set_field(context->byte_132_trrl, V2_QPC_BYTE_132_TRRL_BA_M,
- V2_QPC_BYTE_132_TRRL_BA_S, dma_handle_3 >> 4);
+ V2_QPC_BYTE_132_TRRL_BA_S, trrl_ba >> 4);
roce_set_field(qpc_mask->byte_132_trrl, V2_QPC_BYTE_132_TRRL_BA_M,
V2_QPC_BYTE_132_TRRL_BA_S, 0);
- context->trrl_ba = cpu_to_le32(dma_handle_3 >> (16 + 4));
+ context->trrl_ba = cpu_to_le32(trrl_ba >> (16 + 4));
qpc_mask->trrl_ba = 0;
roce_set_field(context->byte_140_raq, V2_QPC_BYTE_140_TRRL_BA_M,
V2_QPC_BYTE_140_TRRL_BA_S,
- (u32)(dma_handle_3 >> (32 + 16 + 4)));
+ (u32)(trrl_ba >> (32 + 16 + 4)));
roce_set_field(qpc_mask->byte_140_raq, V2_QPC_BYTE_140_TRRL_BA_M,
V2_QPC_BYTE_140_TRRL_BA_S, 0);
- context->irrl_ba = cpu_to_le32(dma_handle_2 >> 6);
+ context->irrl_ba = cpu_to_le32(irrl_ba >> 6);
qpc_mask->irrl_ba = 0;
roce_set_field(context->byte_208_irrl, V2_QPC_BYTE_208_IRRL_BA_M,
V2_QPC_BYTE_208_IRRL_BA_S,
- dma_handle_2 >> (32 + 6));
+ irrl_ba >> (32 + 6));
roce_set_field(qpc_mask->byte_208_irrl, V2_QPC_BYTE_208_IRRL_BA_M,
V2_QPC_BYTE_208_IRRL_BA_S, 0);
@@ -3897,6 +4030,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
port = (attr_mask & IB_QP_PORT) ? (attr->port_num - 1) : hr_qp->port;
smac = (u8 *)hr_dev->dev_addr[port];
+ dmac = (u8 *)attr->ah_attr.roce.dmac;
/* when dmac equals smac or loop_idc is 1, it should loopback */
if (ether_addr_equal_unaligned(dmac, smac) ||
hr_dev->loop_idc == 0x1) {
@@ -3919,6 +4053,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
grh->sgid_index));
roce_set_field(qpc_mask->byte_20_smac_sgid_idx,
V2_QPC_BYTE_20_SGID_IDX_M, V2_QPC_BYTE_20_SGID_IDX_S, 0);
+
memcpy(&(context->dmac), dmac, sizeof(u32));
roce_set_field(context->byte_52_udpspn_dmac, V2_QPC_BYTE_52_DMAC_M,
V2_QPC_BYTE_52_DMAC_S, *((u16 *)(&dmac[4])));
@@ -3928,7 +4063,8 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
/* mtu*(2^LP_PKTN_INI) should not bigger than 1 message length 64kb */
roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_LP_PKTN_INI_M,
- V2_QPC_BYTE_56_LP_PKTN_INI_S, 0);
+ V2_QPC_BYTE_56_LP_PKTN_INI_S,
+ ilog2(hr_dev->caps.max_sq_inline / IB_MTU_4096));
roce_set_field(qpc_mask->byte_56_dqpn_err, V2_QPC_BYTE_56_LP_PKTN_INI_M,
V2_QPC_BYTE_56_LP_PKTN_INI_S, 0);
@@ -3942,16 +4078,6 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp,
roce_set_field(qpc_mask->byte_24_mtu_tc, V2_QPC_BYTE_24_MTU_M,
V2_QPC_BYTE_24_MTU_S, 0);
- roce_set_field(context->byte_84_rq_ci_pi,
- V2_QPC_BYTE_84_RQ_PRODUCER_IDX_M,
- V2_QPC_BYTE_84_RQ_PRODUCER_IDX_S, hr_qp->rq.head);
- roce_set_field(qpc_mask->byte_84_rq_ci_pi,
- V2_QPC_BYTE_84_RQ_PRODUCER_IDX_M,
- V2_QPC_BYTE_84_RQ_PRODUCER_IDX_S, 0);
-
- roce_set_field(qpc_mask->byte_84_rq_ci_pi,
- V2_QPC_BYTE_84_RQ_CONSUMER_IDX_M,
- V2_QPC_BYTE_84_RQ_CONSUMER_IDX_S, 0);
roce_set_bit(qpc_mask->byte_108_rx_reqepsn,
V2_QPC_BYTE_108_RX_REQ_PSN_ERR_S, 0);
roce_set_field(qpc_mask->byte_96_rx_reqmsn, V2_QPC_BYTE_96_RX_REQ_MSN_M,
@@ -3987,30 +4113,7 @@ static int modify_qp_rtr_to_rts(struct ib_qp *ibqp,
struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
struct ib_device *ibdev = &hr_dev->ib_dev;
- u64 sge_cur_blk = 0;
- u64 sq_cur_blk = 0;
- u32 page_size;
- int count;
-
- /* Search qp buf's mtts */
- count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr, 0, &sq_cur_blk, 1, NULL);
- if (count < 1) {
- ibdev_err(ibdev, "failed to find buf pa of QP(0x%lx)\n",
- hr_qp->qpn);
- return -EINVAL;
- }
-
- if (hr_qp->sge.offset) {
- page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
- count = hns_roce_mtr_find(hr_dev, &hr_qp->mtr,
- hr_qp->sge.offset / page_size,
- &sge_cur_blk, 1, NULL);
- if (count < 1) {
- ibdev_err(ibdev, "failed to find sge pa of QP(0x%lx)\n",
- hr_qp->qpn);
- return -EINVAL;
- }
- }
+ int ret;
/* Not support alternate path and path migration */
if (attr_mask & (IB_QP_ALT_PATH | IB_QP_PATH_MIG_STATE)) {
@@ -4018,48 +4121,11 @@ static int modify_qp_rtr_to_rts(struct ib_qp *ibqp,
return -EINVAL;
}
- /*
- * In v2 engine, software pass context and context mask to hardware
- * when modifying qp. If software need modify some fields in context,
- * we should set all bits of the relevant fields in context mask to
- * 0 at the same time, else set them to 0x1.
- */
- context->sq_cur_blk_addr = cpu_to_le32(sq_cur_blk >> PAGE_ADDR_SHIFT);
- roce_set_field(context->byte_168_irrl_idx,
- V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_M,
- V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_S,
- sq_cur_blk >> (32 + PAGE_ADDR_SHIFT));
- qpc_mask->sq_cur_blk_addr = 0;
- roce_set_field(qpc_mask->byte_168_irrl_idx,
- V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_M,
- V2_QPC_BYTE_168_SQ_CUR_BLK_ADDR_S, 0);
-
- context->sq_cur_sge_blk_addr = ((ibqp->qp_type == IB_QPT_GSI) ||
- hr_qp->sq.max_gs > HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE) ?
- cpu_to_le32(sge_cur_blk >>
- PAGE_ADDR_SHIFT) : 0;
- roce_set_field(context->byte_184_irrl_idx,
- V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_M,
- V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_S,
- ((ibqp->qp_type == IB_QPT_GSI) || hr_qp->sq.max_gs >
- HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE) ?
- (sge_cur_blk >>
- (32 + PAGE_ADDR_SHIFT)) : 0);
- qpc_mask->sq_cur_sge_blk_addr = 0;
- roce_set_field(qpc_mask->byte_184_irrl_idx,
- V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_M,
- V2_QPC_BYTE_184_SQ_CUR_SGE_BLK_ADDR_S, 0);
-
- context->rx_sq_cur_blk_addr =
- cpu_to_le32(sq_cur_blk >> PAGE_ADDR_SHIFT);
- roce_set_field(context->byte_232_irrl_sge,
- V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_M,
- V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_S,
- sq_cur_blk >> (32 + PAGE_ADDR_SHIFT));
- qpc_mask->rx_sq_cur_blk_addr = 0;
- roce_set_field(qpc_mask->byte_232_irrl_sge,
- V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_M,
- V2_QPC_BYTE_232_RX_SQ_CUR_BLK_ADDR_S, 0);
+ ret = config_qp_sq_buf(hr_dev, hr_qp, context, qpc_mask);
+ if (ret) {
+ ibdev_err(ibdev, "failed to config sq buf, ret %d\n", ret);
+ return ret;
+ }
/*
* Set some fields in context to zero, Because the default values
@@ -4108,21 +4174,6 @@ static int modify_qp_rtr_to_rts(struct ib_qp *ibqp,
return 0;
}
-static inline bool hns_roce_v2_check_qp_stat(enum ib_qp_state cur_state,
- enum ib_qp_state new_state)
-{
-
- if ((cur_state != IB_QPS_RESET &&
- (new_state == IB_QPS_ERR || new_state == IB_QPS_RESET)) ||
- ((cur_state == IB_QPS_RTS || cur_state == IB_QPS_SQD) &&
- (new_state == IB_QPS_RTS || new_state == IB_QPS_SQD)) ||
- (cur_state == IB_QPS_SQE && new_state == IB_QPS_RTS))
- return true;
-
- return false;
-
-}
-
static int hns_roce_v2_set_path(struct ib_qp *ibqp,
const struct ib_qp_attr *attr,
int attr_mask,
@@ -4226,6 +4277,28 @@ static int hns_roce_v2_set_path(struct ib_qp *ibqp,
return 0;
}
+static bool check_qp_state(enum ib_qp_state cur_state,
+ enum ib_qp_state new_state)
+{
+ static const bool sm[][IB_QPS_ERR + 1] = {
+ [IB_QPS_RESET] = { [IB_QPS_RESET] = true,
+ [IB_QPS_INIT] = true },
+ [IB_QPS_INIT] = { [IB_QPS_RESET] = true,
+ [IB_QPS_INIT] = true,
+ [IB_QPS_RTR] = true,
+ [IB_QPS_ERR] = true },
+ [IB_QPS_RTR] = { [IB_QPS_RESET] = true,
+ [IB_QPS_RTS] = true,
+ [IB_QPS_ERR] = true },
+ [IB_QPS_RTS] = { [IB_QPS_RESET] = true, [IB_QPS_ERR] = true },
+ [IB_QPS_SQD] = {},
+ [IB_QPS_SQE] = {},
+ [IB_QPS_ERR] = { [IB_QPS_RESET] = true, [IB_QPS_ERR] = true }
+ };
+
+ return sm[cur_state][new_state];
+}
+
static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp,
const struct ib_qp_attr *attr,
int attr_mask,
@@ -4237,6 +4310,11 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp,
struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
int ret = 0;
+ if (!check_qp_state(cur_state, new_state)) {
+ ibdev_err(&hr_dev->ib_dev, "Illegal state for QP!\n");
+ return -EINVAL;
+ }
+
if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {
memset(qpc_mask, 0, sizeof(*qpc_mask));
modify_qp_reset_to_init(ibqp, attr, attr_mask, context,
@@ -4247,23 +4325,11 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp,
} else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_RTR) {
ret = modify_qp_init_to_rtr(ibqp, attr, attr_mask, context,
qpc_mask);
- if (ret)
- goto out;
} else if (cur_state == IB_QPS_RTR && new_state == IB_QPS_RTS) {
ret = modify_qp_rtr_to_rts(ibqp, attr, attr_mask, context,
qpc_mask);
- if (ret)
- goto out;
- } else if (hns_roce_v2_check_qp_stat(cur_state, new_state)) {
- /* Nothing */
- ;
- } else {
- ibdev_err(&hr_dev->ib_dev, "Illegal state for QP!\n");
- ret = -EINVAL;
- goto out;
}
-out:
return ret;
}
@@ -4554,19 +4620,20 @@ out:
return ret;
}
-static inline enum ib_qp_state to_ib_qp_st(enum hns_roce_v2_qp_state state)
+static int to_ib_qp_st(enum hns_roce_v2_qp_state state)
{
- switch (state) {
- case HNS_ROCE_QP_ST_RST: return IB_QPS_RESET;
- case HNS_ROCE_QP_ST_INIT: return IB_QPS_INIT;
- case HNS_ROCE_QP_ST_RTR: return IB_QPS_RTR;
- case HNS_ROCE_QP_ST_RTS: return IB_QPS_RTS;
- case HNS_ROCE_QP_ST_SQ_DRAINING:
- case HNS_ROCE_QP_ST_SQD: return IB_QPS_SQD;
- case HNS_ROCE_QP_ST_SQER: return IB_QPS_SQE;
- case HNS_ROCE_QP_ST_ERR: return IB_QPS_ERR;
- default: return -1;
- }
+ static const enum ib_qp_state map[] = {
+ [HNS_ROCE_QP_ST_RST] = IB_QPS_RESET,
+ [HNS_ROCE_QP_ST_INIT] = IB_QPS_INIT,
+ [HNS_ROCE_QP_ST_RTR] = IB_QPS_RTR,
+ [HNS_ROCE_QP_ST_RTS] = IB_QPS_RTS,
+ [HNS_ROCE_QP_ST_SQD] = IB_QPS_SQD,
+ [HNS_ROCE_QP_ST_SQER] = IB_QPS_SQE,
+ [HNS_ROCE_QP_ST_ERR] = IB_QPS_ERR,
+ [HNS_ROCE_QP_ST_SQ_DRAINING] = IB_QPS_SQD
+ };
+
+ return (state < ARRAY_SIZE(map)) ? map[state] : -1;
}
static int hns_roce_v2_query_qpc(struct hns_roce_dev *hr_dev,
@@ -4639,7 +4706,7 @@ static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
qp_attr->path_mig_state = IB_MIG_ARMED;
qp_attr->ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE;
if (hr_qp->ibqp.qp_type == IB_QPT_UD)
- qp_attr->qkey = V2_QKEY_VAL;
+ qp_attr->qkey = le32_to_cpu(context.qkey_xrcd);
qp_attr->rq_psn = roce_get_field(context.byte_108_rx_reqepsn,
V2_QPC_BYTE_108_RX_REQ_EPSN_M,
@@ -4838,6 +4905,184 @@ out:
return ret;
}
+static void hns_roce_v2_write_srqc(struct hns_roce_dev *hr_dev,
+ struct hns_roce_srq *srq, u32 pdn, u16 xrcd,
+ u32 cqn, void *mb_buf, u64 *mtts_wqe,
+ u64 *mtts_idx, dma_addr_t dma_handle_wqe,
+ dma_addr_t dma_handle_idx)
+{
+ struct hns_roce_srq_context *srq_context;
+
+ srq_context = mb_buf;
+ memset(srq_context, 0, sizeof(*srq_context));
+
+ roce_set_field(srq_context->byte_4_srqn_srqst, SRQC_BYTE_4_SRQ_ST_M,
+ SRQC_BYTE_4_SRQ_ST_S, 1);
+
+ roce_set_field(srq_context->byte_4_srqn_srqst,
+ SRQC_BYTE_4_SRQ_WQE_HOP_NUM_M,
+ SRQC_BYTE_4_SRQ_WQE_HOP_NUM_S,
+ to_hr_hem_hopnum(hr_dev->caps.srqwqe_hop_num,
+ srq->wqe_cnt));
+ roce_set_field(srq_context->byte_4_srqn_srqst,
+ SRQC_BYTE_4_SRQ_SHIFT_M, SRQC_BYTE_4_SRQ_SHIFT_S,
+ ilog2(srq->wqe_cnt));
+
+ roce_set_field(srq_context->byte_4_srqn_srqst, SRQC_BYTE_4_SRQN_M,
+ SRQC_BYTE_4_SRQN_S, srq->srqn);
+
+ roce_set_field(srq_context->byte_8_limit_wl, SRQC_BYTE_8_SRQ_LIMIT_WL_M,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_S, 0);
+
+ roce_set_field(srq_context->byte_12_xrcd, SRQC_BYTE_12_SRQ_XRCD_M,
+ SRQC_BYTE_12_SRQ_XRCD_S, xrcd);
+
+ srq_context->wqe_bt_ba = cpu_to_le32((u32)(dma_handle_wqe >> 3));
+
+ roce_set_field(srq_context->byte_24_wqe_bt_ba,
+ SRQC_BYTE_24_SRQ_WQE_BT_BA_M,
+ SRQC_BYTE_24_SRQ_WQE_BT_BA_S,
+ dma_handle_wqe >> 35);
+
+ roce_set_field(srq_context->byte_28_rqws_pd, SRQC_BYTE_28_PD_M,
+ SRQC_BYTE_28_PD_S, pdn);
+ roce_set_field(srq_context->byte_28_rqws_pd, SRQC_BYTE_28_RQWS_M,
+ SRQC_BYTE_28_RQWS_S, srq->max_gs <= 0 ? 0 :
+ fls(srq->max_gs - 1));
+
+ srq_context->idx_bt_ba = cpu_to_le32(dma_handle_idx >> 3);
+ roce_set_field(srq_context->rsv_idx_bt_ba,
+ SRQC_BYTE_36_SRQ_IDX_BT_BA_M,
+ SRQC_BYTE_36_SRQ_IDX_BT_BA_S,
+ dma_handle_idx >> 35);
+
+ srq_context->idx_cur_blk_addr =
+ cpu_to_le32(to_hr_hw_page_addr(mtts_idx[0]));
+ roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
+ SRQC_BYTE_44_SRQ_IDX_CUR_BLK_ADDR_M,
+ SRQC_BYTE_44_SRQ_IDX_CUR_BLK_ADDR_S,
+ upper_32_bits(to_hr_hw_page_addr(mtts_idx[0])));
+ roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
+ SRQC_BYTE_44_SRQ_IDX_HOP_NUM_M,
+ SRQC_BYTE_44_SRQ_IDX_HOP_NUM_S,
+ to_hr_hem_hopnum(hr_dev->caps.idx_hop_num,
+ srq->wqe_cnt));
+
+ roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
+ SRQC_BYTE_44_SRQ_IDX_BA_PG_SZ_M,
+ SRQC_BYTE_44_SRQ_IDX_BA_PG_SZ_S,
+ to_hr_hw_page_shift(srq->idx_que.mtr.hem_cfg.ba_pg_shift));
+ roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
+ SRQC_BYTE_44_SRQ_IDX_BUF_PG_SZ_M,
+ SRQC_BYTE_44_SRQ_IDX_BUF_PG_SZ_S,
+ to_hr_hw_page_shift(srq->idx_que.mtr.hem_cfg.buf_pg_shift));
+
+ srq_context->idx_nxt_blk_addr =
+ cpu_to_le32(to_hr_hw_page_addr(mtts_idx[1]));
+ roce_set_field(srq_context->rsv_idxnxtblkaddr,
+ SRQC_BYTE_52_SRQ_IDX_NXT_BLK_ADDR_M,
+ SRQC_BYTE_52_SRQ_IDX_NXT_BLK_ADDR_S,
+ upper_32_bits(to_hr_hw_page_addr(mtts_idx[1])));
+ roce_set_field(srq_context->byte_56_xrc_cqn,
+ SRQC_BYTE_56_SRQ_XRC_CQN_M, SRQC_BYTE_56_SRQ_XRC_CQN_S,
+ cqn);
+ roce_set_field(srq_context->byte_56_xrc_cqn,
+ SRQC_BYTE_56_SRQ_WQE_BA_PG_SZ_M,
+ SRQC_BYTE_56_SRQ_WQE_BA_PG_SZ_S,
+ to_hr_hw_page_shift(srq->buf_mtr.hem_cfg.ba_pg_shift));
+ roce_set_field(srq_context->byte_56_xrc_cqn,
+ SRQC_BYTE_56_SRQ_WQE_BUF_PG_SZ_M,
+ SRQC_BYTE_56_SRQ_WQE_BUF_PG_SZ_S,
+ to_hr_hw_page_shift(srq->buf_mtr.hem_cfg.buf_pg_shift));
+
+ roce_set_bit(srq_context->db_record_addr_record_en,
+ SRQC_BYTE_60_SRQ_RECORD_EN_S, 0);
+}
+
+static int hns_roce_v2_modify_srq(struct ib_srq *ibsrq,
+ struct ib_srq_attr *srq_attr,
+ enum ib_srq_attr_mask srq_attr_mask,
+ struct ib_udata *udata)
+{
+ struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
+ struct hns_roce_srq *srq = to_hr_srq(ibsrq);
+ struct hns_roce_srq_context *srq_context;
+ struct hns_roce_srq_context *srqc_mask;
+ struct hns_roce_cmd_mailbox *mailbox;
+ int ret;
+
+ if (srq_attr_mask & IB_SRQ_LIMIT) {
+ if (srq_attr->srq_limit >= srq->wqe_cnt)
+ return -EINVAL;
+
+ mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+
+ srq_context = mailbox->buf;
+ srqc_mask = (struct hns_roce_srq_context *)mailbox->buf + 1;
+
+ memset(srqc_mask, 0xff, sizeof(*srqc_mask));
+
+ roce_set_field(srq_context->byte_8_limit_wl,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_M,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_S, srq_attr->srq_limit);
+ roce_set_field(srqc_mask->byte_8_limit_wl,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_M,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_S, 0);
+
+ ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, srq->srqn, 0,
+ HNS_ROCE_CMD_MODIFY_SRQC,
+ HNS_ROCE_CMD_TIMEOUT_MSECS);
+ hns_roce_free_cmd_mailbox(hr_dev, mailbox);
+ if (ret) {
+ ibdev_err(&hr_dev->ib_dev,
+ "failed to handle cmd of modifying SRQ, ret = %d.\n",
+ ret);
+ return ret;
+ }
+ }
+
+ return 0;
+}
+
+static int hns_roce_v2_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr)
+{
+ struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
+ struct hns_roce_srq *srq = to_hr_srq(ibsrq);
+ struct hns_roce_srq_context *srq_context;
+ struct hns_roce_cmd_mailbox *mailbox;
+ int limit_wl;
+ int ret;
+
+ mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
+ if (IS_ERR(mailbox))
+ return PTR_ERR(mailbox);
+
+ srq_context = mailbox->buf;
+ ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, srq->srqn, 0,
+ HNS_ROCE_CMD_QUERY_SRQC,
+ HNS_ROCE_CMD_TIMEOUT_MSECS);
+ if (ret) {
+ ibdev_err(&hr_dev->ib_dev,
+ "failed to process cmd of querying SRQ, ret = %d.\n",
+ ret);
+ goto out;
+ }
+
+ limit_wl = roce_get_field(srq_context->byte_8_limit_wl,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_M,
+ SRQC_BYTE_8_SRQ_LIMIT_WL_S);
+
+ attr->srq_limit = limit_wl;
+ attr->max_wr = srq->wqe_cnt - 1;
+ attr->max_sge = srq->max_gs - HNS_ROCE_RESERVED_SGE;
+
+out:
+ hns_roce_free_cmd_mailbox(hr_dev, mailbox);
+ return ret;
+}
+
static int hns_roce_v2_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
{
struct hns_roce_dev *hr_dev = to_hr_dev(cq->device);
@@ -4989,24 +5234,14 @@ static void set_eq_cons_index_v2(struct hns_roce_eq *eq)
hns_roce_write64(hr_dev, doorbell, eq->doorbell);
}
-static inline void *get_eqe_buf(struct hns_roce_eq *eq, unsigned long offset)
-{
- u32 buf_chk_sz;
-
- buf_chk_sz = 1 << (eq->eqe_buf_pg_sz + PAGE_SHIFT);
- if (eq->buf.nbufs == 1)
- return eq->buf.direct.buf + offset % buf_chk_sz;
- else
- return eq->buf.page_list[offset / buf_chk_sz].buf +
- offset % buf_chk_sz;
-}
-
static struct hns_roce_aeqe *next_aeqe_sw_v2(struct hns_roce_eq *eq)
{
struct hns_roce_aeqe *aeqe;
- aeqe = get_eqe_buf(eq, (eq->cons_index & (eq->entries - 1)) *
- HNS_ROCE_AEQ_ENTRY_SIZE);
+ aeqe = hns_roce_buf_offset(eq->mtr.kmem,
+ (eq->cons_index & (eq->entries - 1)) *
+ HNS_ROCE_AEQ_ENTRY_SIZE);
+
return (roce_get_bit(aeqe->asyn, HNS_ROCE_V2_AEQ_AEQE_OWNER_S) ^
!!(eq->cons_index & eq->entries)) ? aeqe : NULL;
}
@@ -5103,8 +5338,9 @@ static struct hns_roce_ceqe *next_ceqe_sw_v2(struct hns_roce_eq *eq)
{
struct hns_roce_ceqe *ceqe;
- ceqe = get_eqe_buf(eq, (eq->cons_index & (eq->entries - 1)) *
- HNS_ROCE_CEQ_ENTRY_SIZE);
+ ceqe = hns_roce_buf_offset(eq->mtr.kmem,
+ (eq->cons_index & (eq->entries - 1)) *
+ HNS_ROCE_CEQ_ENTRY_SIZE);
return (!!(roce_get_bit(ceqe->comp, HNS_ROCE_V2_CEQ_CEQE_OWNER_S))) ^
(!!(eq->cons_index & eq->entries)) ? ceqe : NULL;
}
@@ -5263,17 +5499,15 @@ static void hns_roce_v2_destroy_eqc(struct hns_roce_dev *hr_dev, int eqn)
static void free_eq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_eq *eq)
{
- if (!eq->hop_num || eq->hop_num == HNS_ROCE_HOP_NUM_0)
- hns_roce_mtr_cleanup(hr_dev, &eq->mtr);
- hns_roce_buf_free(hr_dev, eq->buf.size, &eq->buf);
+ hns_roce_mtr_destroy(hr_dev, &eq->mtr);
}
-static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
- struct hns_roce_eq *eq,
- void *mb_buf)
+static int config_eqc(struct hns_roce_dev *hr_dev, struct hns_roce_eq *eq,
+ void *mb_buf)
{
+ u64 eqe_ba[MTT_MIN_COUNT] = { 0 };
struct hns_roce_eq_context *eqc;
- u64 ba[MTT_MIN_COUNT] = { 0 };
+ u64 bt_ba = 0;
int count;
eqc = mb_buf;
@@ -5281,31 +5515,18 @@ static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
/* init eqc */
eq->doorbell = hr_dev->reg_base + ROCEE_VF_EQ_DB_CFG0_REG;
- eq->hop_num = hr_dev->caps.eqe_hop_num;
eq->cons_index = 0;
eq->over_ignore = HNS_ROCE_V2_EQ_OVER_IGNORE_0;
eq->coalesce = HNS_ROCE_V2_EQ_COALESCE_0;
eq->arm_st = HNS_ROCE_V2_EQ_ALWAYS_ARMED;
- eq->eqe_ba_pg_sz = hr_dev->caps.eqe_ba_pg_sz;
- eq->eqe_buf_pg_sz = hr_dev->caps.eqe_buf_pg_sz;
eq->shift = ilog2((unsigned int)eq->entries);
- /* if not muti-hop, eqe buffer only use one trunk */
- if (!eq->hop_num || eq->hop_num == HNS_ROCE_HOP_NUM_0) {
- eq->eqe_ba = eq->buf.direct.map;
- eq->cur_eqe_ba = eq->eqe_ba;
- if (eq->buf.npages > 1)
- eq->nxt_eqe_ba = eq->eqe_ba + (1 << eq->eqe_buf_pg_sz);
- else
- eq->nxt_eqe_ba = eq->eqe_ba;
- } else {
- count = hns_roce_mtr_find(hr_dev, &eq->mtr, 0, ba,
- MTT_MIN_COUNT, &eq->eqe_ba);
- eq->cur_eqe_ba = ba[0];
- if (count > 1)
- eq->nxt_eqe_ba = ba[1];
- else
- eq->nxt_eqe_ba = ba[0];
+ /* if not multi-hop, eqe buffer only use one trunk */
+ count = hns_roce_mtr_find(hr_dev, &eq->mtr, 0, eqe_ba, MTT_MIN_COUNT,
+ &bt_ba);
+ if (count < 1) {
+ dev_err(hr_dev->dev, "failed to find EQE mtr\n");
+ return -ENOBUFS;
}
/* set eqc state */
@@ -5339,12 +5560,12 @@ static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
/* set eqe_ba_pg_sz */
roce_set_field(eqc->byte_8, HNS_ROCE_EQC_BA_PG_SZ_M,
HNS_ROCE_EQC_BA_PG_SZ_S,
- eq->eqe_ba_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(eq->mtr.hem_cfg.ba_pg_shift));
/* set eqe_buf_pg_sz */
roce_set_field(eqc->byte_8, HNS_ROCE_EQC_BUF_PG_SZ_M,
HNS_ROCE_EQC_BUF_PG_SZ_S,
- eq->eqe_buf_pg_sz + PG_SHIFT_OFFSET);
+ to_hr_hw_page_shift(eq->mtr.hem_cfg.buf_pg_shift));
/* set eq_producer_idx */
roce_set_field(eqc->byte_8, HNS_ROCE_EQC_PROD_INDX_M,
@@ -5363,13 +5584,13 @@ static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
HNS_ROCE_EQC_REPORT_TIMER_S,
HNS_ROCE_EQ_INIT_REPORT_TIMER);
- /* set eqe_ba [34:3] */
+ /* set bt_ba [34:3] */
roce_set_field(eqc->eqe_ba0, HNS_ROCE_EQC_EQE_BA_L_M,
- HNS_ROCE_EQC_EQE_BA_L_S, eq->eqe_ba >> 3);
+ HNS_ROCE_EQC_EQE_BA_L_S, bt_ba >> 3);
- /* set eqe_ba [64:35] */
+ /* set bt_ba [64:35] */
roce_set_field(eqc->eqe_ba1, HNS_ROCE_EQC_EQE_BA_H_M,
- HNS_ROCE_EQC_EQE_BA_H_S, eq->eqe_ba >> 35);
+ HNS_ROCE_EQC_EQE_BA_H_S, bt_ba >> 35);
/* set eq shift */
roce_set_field(eqc->byte_28, HNS_ROCE_EQC_SHIFT_M, HNS_ROCE_EQC_SHIFT_S,
@@ -5381,15 +5602,15 @@ static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
/* set cur_eqe_ba [27:12] */
roce_set_field(eqc->byte_28, HNS_ROCE_EQC_CUR_EQE_BA_L_M,
- HNS_ROCE_EQC_CUR_EQE_BA_L_S, eq->cur_eqe_ba >> 12);
+ HNS_ROCE_EQC_CUR_EQE_BA_L_S, eqe_ba[0] >> 12);
/* set cur_eqe_ba [59:28] */
roce_set_field(eqc->byte_32, HNS_ROCE_EQC_CUR_EQE_BA_M_M,
- HNS_ROCE_EQC_CUR_EQE_BA_M_S, eq->cur_eqe_ba >> 28);
+ HNS_ROCE_EQC_CUR_EQE_BA_M_S, eqe_ba[0] >> 28);
/* set cur_eqe_ba [63:60] */
roce_set_field(eqc->byte_36, HNS_ROCE_EQC_CUR_EQE_BA_H_M,
- HNS_ROCE_EQC_CUR_EQE_BA_H_S, eq->cur_eqe_ba >> 60);
+ HNS_ROCE_EQC_CUR_EQE_BA_H_S, eqe_ba[0] >> 60);
/* set eq consumer idx */
roce_set_field(eqc->byte_36, HNS_ROCE_EQC_CONS_INDX_M,
@@ -5397,97 +5618,38 @@ static void hns_roce_config_eqc(struct hns_roce_dev *hr_dev,
/* set nex_eqe_ba[43:12] */
roce_set_field(eqc->nxt_eqe_ba0, HNS_ROCE_EQC_NXT_EQE_BA_L_M,
- HNS_ROCE_EQC_NXT_EQE_BA_L_S, eq->nxt_eqe_ba >> 12);
+ HNS_ROCE_EQC_NXT_EQE_BA_L_S, eqe_ba[1] >> 12);
/* set nex_eqe_ba[63:44] */
roce_set_field(eqc->nxt_eqe_ba1, HNS_ROCE_EQC_NXT_EQE_BA_H_M,
- HNS_ROCE_EQC_NXT_EQE_BA_H_S, eq->nxt_eqe_ba >> 44);
-}
+ HNS_ROCE_EQC_NXT_EQE_BA_H_S, eqe_ba[1] >> 44);
-static int map_eq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_eq *eq,
- u32 page_shift)
-{
- struct hns_roce_buf_region region = {};
- dma_addr_t *buf_list = NULL;
- int ba_num;
- int ret;
-
- ba_num = DIV_ROUND_UP(PAGE_ALIGN(eq->entries * eq->eqe_size),
- 1 << page_shift);
- hns_roce_init_buf_region(&region, hr_dev->caps.eqe_hop_num, 0, ba_num);
-
- /* alloc a tmp list for storing eq buf address */
- ret = hns_roce_alloc_buf_list(&region, &buf_list, 1);
- if (ret) {
- dev_err(hr_dev->dev, "alloc eq buf_list error\n");
- return ret;
- }
-
- ba_num = hns_roce_get_kmem_bufs(hr_dev, buf_list, region.count,
- region.offset, &eq->buf);
- if (ba_num != region.count) {
- dev_err(hr_dev->dev, "get eqe buf err,expect %d,ret %d.\n",
- region.count, ba_num);
- ret = -ENOBUFS;
- goto done;
- }
-
- hns_roce_mtr_init(&eq->mtr, PAGE_SHIFT + hr_dev->caps.eqe_ba_pg_sz,
- page_shift);
- ret = hns_roce_mtr_attach(hr_dev, &eq->mtr, &buf_list, &region, 1);
- if (ret)
- dev_err(hr_dev->dev, "mtr attach error for eqe\n");
-
- goto done;
-
- hns_roce_mtr_cleanup(hr_dev, &eq->mtr);
-done:
- hns_roce_free_buf_list(&buf_list, 1);
-
- return ret;
+ return 0;
}
static int alloc_eq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_eq *eq)
{
- struct hns_roce_buf *buf = &eq->buf;
- bool is_mhop = false;
- u32 page_shift;
- u32 mhop_num;
- u32 max_size;
- int ret;
+ struct hns_roce_buf_attr buf_attr = {};
+ int err;
- page_shift = PAGE_SHIFT + hr_dev->caps.eqe_buf_pg_sz;
- mhop_num = hr_dev->caps.eqe_hop_num;
- if (!mhop_num) {
- max_size = 1 << page_shift;
- buf->size = max_size;
- } else if (mhop_num == HNS_ROCE_HOP_NUM_0) {
- max_size = eq->entries * eq->eqe_size;
- buf->size = max_size;
- } else {
- max_size = 1 << page_shift;
- buf->size = PAGE_ALIGN(eq->entries * eq->eqe_size);
- is_mhop = true;
- }
+ if (hr_dev->caps.eqe_hop_num == HNS_ROCE_HOP_NUM_0)
+ eq->hop_num = 0;
+ else
+ eq->hop_num = hr_dev->caps.eqe_hop_num;
- ret = hns_roce_buf_alloc(hr_dev, buf->size, max_size, buf, page_shift);
- if (ret) {
- dev_err(hr_dev->dev, "alloc eq buf error\n");
- return ret;
- }
+ buf_attr.page_shift = hr_dev->caps.eqe_buf_pg_sz + HNS_HW_PAGE_SHIFT;
+ buf_attr.region[0].size = eq->entries * eq->eqe_size;
+ buf_attr.region[0].hopnum = eq->hop_num;
+ buf_attr.region_count = 1;
+ buf_attr.fixed_page = true;
- if (is_mhop) {
- ret = map_eq_buf(hr_dev, eq, page_shift);
- if (ret) {
- dev_err(hr_dev->dev, "map roce buf error\n");
- goto err_alloc;
- }
- }
+ err = hns_roce_mtr_create(hr_dev, &eq->mtr, &buf_attr,
+ hr_dev->caps.eqe_ba_pg_sz +
+ HNS_HW_PAGE_SHIFT, NULL, 0);
+ if (err)
+ dev_err(hr_dev->dev, "Failed to alloc EQE mtr, err %d\n", err);
- return 0;
-err_alloc:
- hns_roce_buf_free(hr_dev, buf->size, buf);
- return ret;
+ return err;
}
static int hns_roce_v2_create_eq(struct hns_roce_dev *hr_dev,
@@ -5499,15 +5661,16 @@ static int hns_roce_v2_create_eq(struct hns_roce_dev *hr_dev,
/* Allocate mailbox memory */
mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
- if (IS_ERR(mailbox))
- return PTR_ERR(mailbox);
+ if (IS_ERR_OR_NULL(mailbox))
+ return -ENOMEM;
ret = alloc_eq_buf(hr_dev, eq);
- if (ret) {
- ret = -ENOMEM;
+ if (ret)
goto free_cmd_mbox;
- }
- hns_roce_config_eqc(hr_dev, eq, mailbox->buf);
+
+ ret = config_eqc(hr_dev, eq, mailbox->buf);
+ if (ret)
+ goto err_cmd_mbox;
ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, eq->eqn, 0,
eq_cmd, HNS_ROCE_CMD_TIMEOUT_MSECS);
@@ -5731,294 +5894,6 @@ static void hns_roce_v2_cleanup_eq_table(struct hns_roce_dev *hr_dev)
destroy_workqueue(hr_dev->irq_workq);
}
-static void hns_roce_v2_write_srqc(struct hns_roce_dev *hr_dev,
- struct hns_roce_srq *srq, u32 pdn, u16 xrcd,
- u32 cqn, void *mb_buf, u64 *mtts_wqe,
- u64 *mtts_idx, dma_addr_t dma_handle_wqe,
- dma_addr_t dma_handle_idx)
-{
- struct hns_roce_srq_context *srq_context;
-
- srq_context = mb_buf;
- memset(srq_context, 0, sizeof(*srq_context));
-
- roce_set_field(srq_context->byte_4_srqn_srqst, SRQC_BYTE_4_SRQ_ST_M,
- SRQC_BYTE_4_SRQ_ST_S, 1);
-
- roce_set_field(srq_context->byte_4_srqn_srqst,
- SRQC_BYTE_4_SRQ_WQE_HOP_NUM_M,
- SRQC_BYTE_4_SRQ_WQE_HOP_NUM_S,
- (hr_dev->caps.srqwqe_hop_num == HNS_ROCE_HOP_NUM_0 ? 0 :
- hr_dev->caps.srqwqe_hop_num));
- roce_set_field(srq_context->byte_4_srqn_srqst,
- SRQC_BYTE_4_SRQ_SHIFT_M, SRQC_BYTE_4_SRQ_SHIFT_S,
- ilog2(srq->wqe_cnt));
-
- roce_set_field(srq_context->byte_4_srqn_srqst, SRQC_BYTE_4_SRQN_M,
- SRQC_BYTE_4_SRQN_S, srq->srqn);
-
- roce_set_field(srq_context->byte_8_limit_wl, SRQC_BYTE_8_SRQ_LIMIT_WL_M,
- SRQC_BYTE_8_SRQ_LIMIT_WL_S, 0);
-
- roce_set_field(srq_context->byte_12_xrcd, SRQC_BYTE_12_SRQ_XRCD_M,
- SRQC_BYTE_12_SRQ_XRCD_S, xrcd);
-
- srq_context->wqe_bt_ba = cpu_to_le32((u32)(dma_handle_wqe >> 3));
-
- roce_set_field(srq_context->byte_24_wqe_bt_ba,
- SRQC_BYTE_24_SRQ_WQE_BT_BA_M,
- SRQC_BYTE_24_SRQ_WQE_BT_BA_S,
- dma_handle_wqe >> 35);
-
- roce_set_field(srq_context->byte_28_rqws_pd, SRQC_BYTE_28_PD_M,
- SRQC_BYTE_28_PD_S, pdn);
- roce_set_field(srq_context->byte_28_rqws_pd, SRQC_BYTE_28_RQWS_M,
- SRQC_BYTE_28_RQWS_S, srq->max_gs <= 0 ? 0 :
- fls(srq->max_gs - 1));
-
- srq_context->idx_bt_ba = cpu_to_le32(dma_handle_idx >> 3);
- roce_set_field(srq_context->rsv_idx_bt_ba,
- SRQC_BYTE_36_SRQ_IDX_BT_BA_M,
- SRQC_BYTE_36_SRQ_IDX_BT_BA_S,
- dma_handle_idx >> 35);
-
- srq_context->idx_cur_blk_addr =
- cpu_to_le32(mtts_idx[0] >> PAGE_ADDR_SHIFT);
- roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
- SRQC_BYTE_44_SRQ_IDX_CUR_BLK_ADDR_M,
- SRQC_BYTE_44_SRQ_IDX_CUR_BLK_ADDR_S,
- mtts_idx[0] >> (32 + PAGE_ADDR_SHIFT));
- roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
- SRQC_BYTE_44_SRQ_IDX_HOP_NUM_M,
- SRQC_BYTE_44_SRQ_IDX_HOP_NUM_S,
- hr_dev->caps.idx_hop_num == HNS_ROCE_HOP_NUM_0 ? 0 :
- hr_dev->caps.idx_hop_num);
-
- roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
- SRQC_BYTE_44_SRQ_IDX_BA_PG_SZ_M,
- SRQC_BYTE_44_SRQ_IDX_BA_PG_SZ_S,
- hr_dev->caps.idx_ba_pg_sz + PG_SHIFT_OFFSET);
- roce_set_field(srq_context->byte_44_idxbufpgsz_addr,
- SRQC_BYTE_44_SRQ_IDX_BUF_PG_SZ_M,
- SRQC_BYTE_44_SRQ_IDX_BUF_PG_SZ_S,
- hr_dev->caps.idx_buf_pg_sz + PG_SHIFT_OFFSET);
-
- srq_context->idx_nxt_blk_addr =
- cpu_to_le32(mtts_idx[1] >> PAGE_ADDR_SHIFT);
- roce_set_field(srq_context->rsv_idxnxtblkaddr,
- SRQC_BYTE_52_SRQ_IDX_NXT_BLK_ADDR_M,
- SRQC_BYTE_52_SRQ_IDX_NXT_BLK_ADDR_S,
- mtts_idx[1] >> (32 + PAGE_ADDR_SHIFT));
- roce_set_field(srq_context->byte_56_xrc_cqn,
- SRQC_BYTE_56_SRQ_XRC_CQN_M, SRQC_BYTE_56_SRQ_XRC_CQN_S,
- cqn);
- roce_set_field(srq_context->byte_56_xrc_cqn,
- SRQC_BYTE_56_SRQ_WQE_BA_PG_SZ_M,
- SRQC_BYTE_56_SRQ_WQE_BA_PG_SZ_S,
- hr_dev->caps.srqwqe_ba_pg_sz + PG_SHIFT_OFFSET);
- roce_set_field(srq_context->byte_56_xrc_cqn,
- SRQC_BYTE_56_SRQ_WQE_BUF_PG_SZ_M,
- SRQC_BYTE_56_SRQ_WQE_BUF_PG_SZ_S,
- hr_dev->caps.srqwqe_buf_pg_sz + PG_SHIFT_OFFSET);
-
- roce_set_bit(srq_context->db_record_addr_record_en,
- SRQC_BYTE_60_SRQ_RECORD_EN_S, 0);
-}
-
-static int hns_roce_v2_modify_srq(struct ib_srq *ibsrq,
- struct ib_srq_attr *srq_attr,
- enum ib_srq_attr_mask srq_attr_mask,
- struct ib_udata *udata)
-{
- struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
- struct hns_roce_srq *srq = to_hr_srq(ibsrq);
- struct hns_roce_srq_context *srq_context;
- struct hns_roce_srq_context *srqc_mask;
- struct hns_roce_cmd_mailbox *mailbox;
- int ret;
-
- if (srq_attr_mask & IB_SRQ_LIMIT) {
- if (srq_attr->srq_limit >= srq->wqe_cnt)
- return -EINVAL;
-
- mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
- if (IS_ERR(mailbox))
- return PTR_ERR(mailbox);
-
- srq_context = mailbox->buf;
- srqc_mask = (struct hns_roce_srq_context *)mailbox->buf + 1;
-
- memset(srqc_mask, 0xff, sizeof(*srqc_mask));
-
- roce_set_field(srq_context->byte_8_limit_wl,
- SRQC_BYTE_8_SRQ_LIMIT_WL_M,
- SRQC_BYTE_8_SRQ_LIMIT_WL_S, srq_attr->srq_limit);
- roce_set_field(srqc_mask->byte_8_limit_wl,
- SRQC_BYTE_8_SRQ_LIMIT_WL_M,
- SRQC_BYTE_8_SRQ_LIMIT_WL_S, 0);
-
- ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, srq->srqn, 0,
- HNS_ROCE_CMD_MODIFY_SRQC,
- HNS_ROCE_CMD_TIMEOUT_MSECS);
- hns_roce_free_cmd_mailbox(hr_dev, mailbox);
- if (ret) {
- ibdev_err(&hr_dev->ib_dev,
- "failed to process cmd when modifying SRQ, ret = %d\n",
- ret);
- return ret;
- }
- }
-
- return 0;
-}
-
-static int hns_roce_v2_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr)
-{
- struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
- struct hns_roce_srq *srq = to_hr_srq(ibsrq);
- struct hns_roce_srq_context *srq_context;
- struct hns_roce_cmd_mailbox *mailbox;
- int limit_wl;
- int ret;
-
- mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
- if (IS_ERR(mailbox))
- return PTR_ERR(mailbox);
-
- srq_context = mailbox->buf;
- ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, srq->srqn, 0,
- HNS_ROCE_CMD_QUERY_SRQC,
- HNS_ROCE_CMD_TIMEOUT_MSECS);
- if (ret) {
- ibdev_err(&hr_dev->ib_dev,
- "failed to process cmd when querying SRQ, ret = %d\n",
- ret);
- goto out;
- }
-
- limit_wl = roce_get_field(srq_context->byte_8_limit_wl,
- SRQC_BYTE_8_SRQ_LIMIT_WL_M,
- SRQC_BYTE_8_SRQ_LIMIT_WL_S);
-
- attr->srq_limit = limit_wl;
- attr->max_wr = srq->wqe_cnt - 1;
- attr->max_sge = srq->max_gs;
-
- memcpy(srq_context, mailbox->buf, sizeof(*srq_context));
-
-out:
- hns_roce_free_cmd_mailbox(hr_dev, mailbox);
- return ret;
-}
-
-static int find_empty_entry(struct hns_roce_idx_que *idx_que,
- unsigned long size)
-{
- int wqe_idx;
-
- if (unlikely(bitmap_full(idx_que->bitmap, size)))
- return -ENOSPC;
-
- wqe_idx = find_first_zero_bit(idx_que->bitmap, size);
-
- bitmap_set(idx_que->bitmap, wqe_idx, 1);
-
- return wqe_idx;
-}
-
-static void fill_idx_queue(struct hns_roce_idx_que *idx_que,
- int cur_idx, int wqe_idx)
-{
- unsigned int *addr;
-
- addr = (unsigned int *)hns_roce_buf_offset(&idx_que->idx_buf,
- cur_idx * idx_que->entry_sz);
- *addr = wqe_idx;
-}
-
-static int hns_roce_v2_post_srq_recv(struct ib_srq *ibsrq,
- const struct ib_recv_wr *wr,
- const struct ib_recv_wr **bad_wr)
-{
- struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
- struct hns_roce_srq *srq = to_hr_srq(ibsrq);
- struct hns_roce_v2_wqe_data_seg *dseg;
- struct hns_roce_v2_db srq_db;
- unsigned long flags;
- int ret = 0;
- int wqe_idx;
- void *wqe;
- int nreq;
- int ind;
- int i;
-
- spin_lock_irqsave(&srq->lock, flags);
-
- ind = srq->head & (srq->wqe_cnt - 1);
-
- for (nreq = 0; wr; ++nreq, wr = wr->next) {
- if (unlikely(wr->num_sge > srq->max_gs)) {
- ret = -EINVAL;
- *bad_wr = wr;
- break;
- }
-
- if (unlikely(srq->head == srq->tail)) {
- ret = -ENOMEM;
- *bad_wr = wr;
- break;
- }
-
- wqe_idx = find_empty_entry(&srq->idx_que, srq->wqe_cnt);
- if (wqe_idx < 0) {
- ret = -ENOMEM;
- *bad_wr = wr;
- break;
- }
-
- fill_idx_queue(&srq->idx_que, ind, wqe_idx);
- wqe = get_srq_wqe(srq, wqe_idx);
- dseg = (struct hns_roce_v2_wqe_data_seg *)wqe;
-
- for (i = 0; i < wr->num_sge; ++i) {
- dseg[i].len = cpu_to_le32(wr->sg_list[i].length);
- dseg[i].lkey = cpu_to_le32(wr->sg_list[i].lkey);
- dseg[i].addr = cpu_to_le64(wr->sg_list[i].addr);
- }
-
- if (i < srq->max_gs) {
- dseg[i].len = 0;
- dseg[i].lkey = cpu_to_le32(0x100);
- dseg[i].addr = 0;
- }
-
- srq->wrid[wqe_idx] = wr->wr_id;
- ind = (ind + 1) & (srq->wqe_cnt - 1);
- }
-
- if (likely(nreq)) {
- srq->head += nreq;
-
- /*
- * Make sure that descriptors are written before
- * doorbell record.
- */
- wmb();
-
- srq_db.byte_4 =
- cpu_to_le32(HNS_ROCE_V2_SRQ_DB << V2_DB_BYTE_4_CMD_S |
- (srq->srqn & V2_DB_BYTE_4_TAG_M));
- srq_db.parameter = cpu_to_le32(srq->head);
-
- hns_roce_write64(hr_dev, (__le32 *)&srq_db, srq->db_reg_l);
-
- }
-
- spin_unlock_irqrestore(&srq->lock, flags);
-
- return ret;
-}
-
static const struct hns_roce_dfx_hw hns_roce_dfx_hw_v2 = {
.query_cqc_info = hns_roce_v2_query_cqc_info,
};
@@ -6161,7 +6036,7 @@ error_failed_kzalloc:
static void __hns_roce_hw_v2_uninit_instance(struct hnae3_handle *handle,
bool reset)
{
- struct hns_roce_dev *hr_dev = (struct hns_roce_dev *)handle->priv;
+ struct hns_roce_dev *hr_dev = handle->priv;
if (!hr_dev)
return;
@@ -6241,7 +6116,7 @@ static int hns_roce_hw_v2_reset_notify_down(struct hnae3_handle *handle)
handle->rinfo.reset_state = HNS_ROCE_STATE_RST_DOWN;
clear_bit(HNS_ROCE_RST_DIRECT_RETURN, &handle->rinfo.state);
- hr_dev = (struct hns_roce_dev *)handle->priv;
+ hr_dev = handle->priv;
if (!hr_dev)
return 0;
diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h
index 82dd9f6f4845..e176b0aaa4ac 100644
--- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h
+++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h
@@ -92,7 +92,9 @@
#define HNS_ROCE_V2_CQC_TIMER_ENTRY_SZ PAGE_SIZE
#define HNS_ROCE_V2_PAGE_SIZE_SUPPORTED 0xFFFFF000
#define HNS_ROCE_V2_MAX_INNER_MTPT_NUM 2
-#define HNS_ROCE_INVALID_LKEY 0x100
+#define HNS_ROCE_INVALID_LKEY 0x0
+#define HNS_ROCE_INVALID_SGE_LENGTH 0x80000000
+
#define HNS_ROCE_CMQ_TX_TIMEOUT 30000
#define HNS_ROCE_V2_UC_RC_SGE_NUM_IN_WQE 2
#define HNS_ROCE_V2_RSV_QPS 8
@@ -1241,10 +1243,9 @@ struct hns_roce_func_clear {
};
#define FUNC_CLEAR_RST_FUN_DONE_S 0
-/* Each physical function manages up to 248 virtual functionsï¼›
- * it takes up to 100ms for each function to execute clearï¼›
- * if an abnormal reset occurs, it is executed twice at most;
- * so it takes up to 249 * 2 * 100ms.
+/* Each physical function manages up to 248 virtual functions, it takes up to
+ * 100ms for each function to execute clear. If an abnormal reset occurs, it is
+ * executed twice at most, so it takes up to 249 * 2 * 100ms.
*/
#define HNS_ROCE_V2_FUNC_CLEAR_TIMEOUT_MSECS (249 * 2 * 100)
#define HNS_ROCE_V2_READ_FUNC_CLEAR_FLAG_INTERVAL 40
@@ -1648,7 +1649,7 @@ struct hns_roce_query_pf_caps_c {
struct hns_roce_query_pf_caps_d {
__le32 wq_hop_num_max_srqs;
__le16 srq_depth;
- __le16 rsv;
+ __le16 cap_flags_ex;
__le32 num_ceqs_ceq_depth;
__le32 arm_st_aeq_depth;
__le32 num_uars_rsv_pds;
@@ -1978,7 +1979,7 @@ int hns_roce_v2_query_cqc_info(struct hns_roce_dev *hr_dev, u32 cqn,
static inline void hns_roce_write64(struct hns_roce_dev *hr_dev, __le32 val[2],
void __iomem *dest)
{
- struct hns_roce_v2_priv *priv = (struct hns_roce_v2_priv *)hr_dev->priv;
+ struct hns_roce_v2_priv *priv = hr_dev->priv;
struct hnae3_handle *handle = priv->handle;
const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c
index d0031d559213..50763cf4fa3d 100644
--- a/drivers/infiniband/hw/hns/hns_roce_main.c
+++ b/drivers/infiniband/hw/hns/hns_roce_main.c
@@ -233,7 +233,6 @@ static int hns_roce_query_port(struct ib_device *ib_dev, u8 port_num,
enum ib_mtu mtu;
u8 port;
- assert(port_num > 0);
port = port_num - 1;
/* props being zeroed by the caller, avoid zeroing it here */
@@ -579,33 +578,12 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
int ret;
struct device *dev = hr_dev->dev;
- ret = hns_roce_init_hem_table(hr_dev, &hr_dev->mr_table.mtt_table,
- HEM_TYPE_MTT, hr_dev->caps.mtt_entry_sz,
- hr_dev->caps.num_mtt_segs, 1);
- if (ret) {
- dev_err(dev, "Failed to init MTT context memory, aborting.\n");
- return ret;
- }
-
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE)) {
- ret = hns_roce_init_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_cqe_table,
- HEM_TYPE_CQE,
- hr_dev->caps.mtt_entry_sz,
- hr_dev->caps.num_cqe_segs, 1);
- if (ret) {
- dev_err(dev,
- "Failed to init CQE context memory, aborting.\n");
- goto err_unmap_cqe;
- }
- }
-
ret = hns_roce_init_hem_table(hr_dev, &hr_dev->mr_table.mtpt_table,
HEM_TYPE_MTPT, hr_dev->caps.mtpt_entry_sz,
hr_dev->caps.num_mtpts, 1);
if (ret) {
dev_err(dev, "Failed to init MTPT context memory, aborting.\n");
- goto err_unmap_mtt;
+ return ret;
}
ret = hns_roce_init_hem_table(hr_dev, &hr_dev->qp_table.qp_table,
@@ -660,32 +638,6 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
}
}
- if (hr_dev->caps.num_srqwqe_segs) {
- ret = hns_roce_init_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_srqwqe_table,
- HEM_TYPE_SRQWQE,
- hr_dev->caps.mtt_entry_sz,
- hr_dev->caps.num_srqwqe_segs, 1);
- if (ret) {
- dev_err(dev,
- "Failed to init MTT srqwqe memory, aborting.\n");
- goto err_unmap_srq;
- }
- }
-
- if (hr_dev->caps.num_idx_segs) {
- ret = hns_roce_init_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_idx_table,
- HEM_TYPE_IDX,
- hr_dev->caps.idx_entry_sz,
- hr_dev->caps.num_idx_segs, 1);
- if (ret) {
- dev_err(dev,
- "Failed to init MTT idx memory, aborting.\n");
- goto err_unmap_srqwqe;
- }
- }
-
if (hr_dev->caps.sccc_entry_sz) {
ret = hns_roce_init_hem_table(hr_dev,
&hr_dev->qp_table.sccc_table,
@@ -695,7 +647,7 @@ static int hns_roce_init_hem(struct hns_roce_dev *hr_dev)
if (ret) {
dev_err(dev,
"Failed to init SCC context memory, aborting.\n");
- goto err_unmap_idx;
+ goto err_unmap_srq;
}
}
@@ -733,17 +685,6 @@ err_unmap_ctx:
if (hr_dev->caps.sccc_entry_sz)
hns_roce_cleanup_hem_table(hr_dev,
&hr_dev->qp_table.sccc_table);
-
-err_unmap_idx:
- if (hr_dev->caps.num_idx_segs)
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_idx_table);
-
-err_unmap_srqwqe:
- if (hr_dev->caps.num_srqwqe_segs)
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_srqwqe_table);
-
err_unmap_srq:
if (hr_dev->caps.srqc_entry_sz)
hns_roce_cleanup_hem_table(hr_dev, &hr_dev->srq_table.table);
@@ -765,14 +706,6 @@ err_unmap_qp:
err_unmap_dmpt:
hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtpt_table);
-err_unmap_mtt:
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- hns_roce_cleanup_hem_table(hr_dev,
- &hr_dev->mr_table.mtt_cqe_table);
-
-err_unmap_cqe:
- hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table);
-
return ret;
}
diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c
index 176f34692f88..4c0bbb12770d 100644
--- a/drivers/infiniband/hw/hns/hns_roce_mr.c
+++ b/drivers/infiniband/hw/hns/hns_roce_mr.c
@@ -66,645 +66,89 @@ int hns_roce_hw_destroy_mpt(struct hns_roce_dev *hr_dev,
HNS_ROCE_CMD_TIMEOUT_MSECS);
}
-static int hns_roce_buddy_alloc(struct hns_roce_buddy *buddy, int order,
- unsigned long *seg)
+static int alloc_mr_key(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr,
+ u32 pd, u64 iova, u64 size, u32 access)
{
- int o;
- u32 m;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ unsigned long obj = 0;
+ int err;
- spin_lock(&buddy->lock);
-
- for (o = order; o <= buddy->max_order; ++o) {
- if (buddy->num_free[o]) {
- m = 1 << (buddy->max_order - o);
- *seg = find_first_bit(buddy->bits[o], m);
- if (*seg < m)
- goto found;
- }
- }
- spin_unlock(&buddy->lock);
- return -EINVAL;
-
- found:
- clear_bit(*seg, buddy->bits[o]);
- --buddy->num_free[o];
-
- while (o > order) {
- --o;
- *seg <<= 1;
- set_bit(*seg ^ 1, buddy->bits[o]);
- ++buddy->num_free[o];
- }
-
- spin_unlock(&buddy->lock);
-
- *seg <<= order;
- return 0;
-}
-
-static void hns_roce_buddy_free(struct hns_roce_buddy *buddy, unsigned long seg,
- int order)
-{
- seg >>= order;
-
- spin_lock(&buddy->lock);
-
- while (test_bit(seg ^ 1, buddy->bits[order])) {
- clear_bit(seg ^ 1, buddy->bits[order]);
- --buddy->num_free[order];
- seg >>= 1;
- ++order;
- }
-
- set_bit(seg, buddy->bits[order]);
- ++buddy->num_free[order];
-
- spin_unlock(&buddy->lock);
-}
-
-static int hns_roce_buddy_init(struct hns_roce_buddy *buddy, int max_order)
-{
- int i, s;
-
- buddy->max_order = max_order;
- spin_lock_init(&buddy->lock);
- buddy->bits = kcalloc(buddy->max_order + 1,
- sizeof(*buddy->bits),
- GFP_KERNEL);
- buddy->num_free = kcalloc(buddy->max_order + 1,
- sizeof(*buddy->num_free),
- GFP_KERNEL);
- if (!buddy->bits || !buddy->num_free)
- goto err_out;
-
- for (i = 0; i <= buddy->max_order; ++i) {
- s = BITS_TO_LONGS(1 << (buddy->max_order - i));
- buddy->bits[i] = kcalloc(s, sizeof(long), GFP_KERNEL |
- __GFP_NOWARN);
- if (!buddy->bits[i]) {
- buddy->bits[i] = vzalloc(array_size(s, sizeof(long)));
- if (!buddy->bits[i])
- goto err_out_free;
- }
- }
-
- set_bit(0, buddy->bits[buddy->max_order]);
- buddy->num_free[buddy->max_order] = 1;
-
- return 0;
-
-err_out_free:
- for (i = 0; i <= buddy->max_order; ++i)
- kvfree(buddy->bits[i]);
-
-err_out:
- kfree(buddy->bits);
- kfree(buddy->num_free);
- return -ENOMEM;
-}
-
-static void hns_roce_buddy_cleanup(struct hns_roce_buddy *buddy)
-{
- int i;
-
- for (i = 0; i <= buddy->max_order; ++i)
- kvfree(buddy->bits[i]);
-
- kfree(buddy->bits);
- kfree(buddy->num_free);
-}
-
-static int hns_roce_alloc_mtt_range(struct hns_roce_dev *hr_dev, int order,
- unsigned long *seg, u32 mtt_type)
-{
- struct hns_roce_mr_table *mr_table = &hr_dev->mr_table;
- struct hns_roce_hem_table *table;
- struct hns_roce_buddy *buddy;
- int ret;
-
- switch (mtt_type) {
- case MTT_TYPE_WQE:
- buddy = &mr_table->mtt_buddy;
- table = &mr_table->mtt_table;
- break;
- case MTT_TYPE_CQE:
- buddy = &mr_table->mtt_cqe_buddy;
- table = &mr_table->mtt_cqe_table;
- break;
- case MTT_TYPE_SRQWQE:
- buddy = &mr_table->mtt_srqwqe_buddy;
- table = &mr_table->mtt_srqwqe_table;
- break;
- case MTT_TYPE_IDX:
- buddy = &mr_table->mtt_idx_buddy;
- table = &mr_table->mtt_idx_table;
- break;
- default:
- dev_err(hr_dev->dev, "Unsupport MTT table type: %d\n",
- mtt_type);
- return -EINVAL;
- }
-
- ret = hns_roce_buddy_alloc(buddy, order, seg);
- if (ret)
- return ret;
-
- ret = hns_roce_table_get_range(hr_dev, table, *seg,
- *seg + (1 << order) - 1);
- if (ret) {
- hns_roce_buddy_free(buddy, *seg, order);
- return ret;
- }
-
- return 0;
-}
-
-int hns_roce_mtt_init(struct hns_roce_dev *hr_dev, int npages, int page_shift,
- struct hns_roce_mtt *mtt)
-{
- int ret;
- int i;
-
- /* Page num is zero, correspond to DMA memory register */
- if (!npages) {
- mtt->order = -1;
- mtt->page_shift = HNS_ROCE_HEM_PAGE_SHIFT;
- return 0;
- }
-
- /* Note: if page_shift is zero, FAST memory register */
- mtt->page_shift = page_shift;
-
- /* Compute MTT entry necessary */
- for (mtt->order = 0, i = HNS_ROCE_MTT_ENTRY_PER_SEG; i < npages;
- i <<= 1)
- ++mtt->order;
-
- /* Allocate MTT entry */
- ret = hns_roce_alloc_mtt_range(hr_dev, mtt->order, &mtt->first_seg,
- mtt->mtt_type);
- if (ret)
- return -ENOMEM;
-
- return 0;
-}
-
-void hns_roce_mtt_cleanup(struct hns_roce_dev *hr_dev, struct hns_roce_mtt *mtt)
-{
- struct hns_roce_mr_table *mr_table = &hr_dev->mr_table;
-
- if (mtt->order < 0)
- return;
-
- switch (mtt->mtt_type) {
- case MTT_TYPE_WQE:
- hns_roce_buddy_free(&mr_table->mtt_buddy, mtt->first_seg,
- mtt->order);
- hns_roce_table_put_range(hr_dev, &mr_table->mtt_table,
- mtt->first_seg,
- mtt->first_seg + (1 << mtt->order) - 1);
- break;
- case MTT_TYPE_CQE:
- hns_roce_buddy_free(&mr_table->mtt_cqe_buddy, mtt->first_seg,
- mtt->order);
- hns_roce_table_put_range(hr_dev, &mr_table->mtt_cqe_table,
- mtt->first_seg,
- mtt->first_seg + (1 << mtt->order) - 1);
- break;
- case MTT_TYPE_SRQWQE:
- hns_roce_buddy_free(&mr_table->mtt_srqwqe_buddy, mtt->first_seg,
- mtt->order);
- hns_roce_table_put_range(hr_dev, &mr_table->mtt_srqwqe_table,
- mtt->first_seg,
- mtt->first_seg + (1 << mtt->order) - 1);
- break;
- case MTT_TYPE_IDX:
- hns_roce_buddy_free(&mr_table->mtt_idx_buddy, mtt->first_seg,
- mtt->order);
- hns_roce_table_put_range(hr_dev, &mr_table->mtt_idx_table,
- mtt->first_seg,
- mtt->first_seg + (1 << mtt->order) - 1);
- break;
- default:
- dev_err(hr_dev->dev,
- "Unsupport mtt type %d, clean mtt failed\n",
- mtt->mtt_type);
- break;
- }
-}
-
-static void hns_roce_loop_free(struct hns_roce_dev *hr_dev,
- struct hns_roce_mr *mr, int err_loop_index,
- int loop_i, int loop_j)
-{
- struct device *dev = hr_dev->dev;
- u32 mhop_num;
- u32 pbl_bt_sz;
- u64 bt_idx;
- int i, j;
-
- pbl_bt_sz = 1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT);
- mhop_num = hr_dev->caps.pbl_hop_num;
-
- i = loop_i;
- if (mhop_num == 3 && err_loop_index == 2) {
- for (; i >= 0; i--) {
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
-
- for (j = 0; j < pbl_bt_sz / BA_BYTE_LEN; j++) {
- if (i == loop_i && j >= loop_j)
- break;
-
- bt_idx = i * pbl_bt_sz / BA_BYTE_LEN + j;
- dma_free_coherent(dev, pbl_bt_sz,
- mr->pbl_bt_l2[bt_idx],
- mr->pbl_l2_dma_addr[bt_idx]);
- }
- }
- } else if (mhop_num == 3 && err_loop_index == 1) {
- for (i -= 1; i >= 0; i--) {
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
-
- for (j = 0; j < pbl_bt_sz / BA_BYTE_LEN; j++) {
- bt_idx = i * pbl_bt_sz / BA_BYTE_LEN + j;
- dma_free_coherent(dev, pbl_bt_sz,
- mr->pbl_bt_l2[bt_idx],
- mr->pbl_l2_dma_addr[bt_idx]);
- }
- }
- } else if (mhop_num == 2 && err_loop_index == 1) {
- for (i -= 1; i >= 0; i--)
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
- } else {
- dev_warn(dev, "not support: mhop_num=%d, err_loop_index=%d.",
- mhop_num, err_loop_index);
- return;
- }
-
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l0, mr->pbl_l0_dma_addr);
- mr->pbl_bt_l0 = NULL;
- mr->pbl_l0_dma_addr = 0;
-}
-static int pbl_1hop_alloc(struct hns_roce_dev *hr_dev, int npages,
- struct hns_roce_mr *mr, u32 pbl_bt_sz)
-{
- struct device *dev = hr_dev->dev;
-
- if (npages > pbl_bt_sz / 8) {
- dev_err(dev, "npages %d is larger than buf_pg_sz!",
- npages);
- return -EINVAL;
- }
- mr->pbl_buf = dma_alloc_coherent(dev, npages * 8,
- &(mr->pbl_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_buf)
+ /* Allocate a key for mr from mr_table */
+ err = hns_roce_bitmap_alloc(&hr_dev->mr_table.mtpt_bitmap, &obj);
+ if (err) {
+ ibdev_err(ibdev,
+ "failed to alloc bitmap for MR key, ret = %d.\n",
+ err);
return -ENOMEM;
-
- mr->pbl_size = npages;
- mr->pbl_ba = mr->pbl_dma_addr;
- mr->pbl_hop_num = 1;
- mr->pbl_ba_pg_sz = hr_dev->caps.pbl_ba_pg_sz;
- mr->pbl_buf_pg_sz = hr_dev->caps.pbl_buf_pg_sz;
- return 0;
-
-}
-
-
-static int pbl_2hop_alloc(struct hns_roce_dev *hr_dev, int npages,
- struct hns_roce_mr *mr, u32 pbl_bt_sz)
-{
- struct device *dev = hr_dev->dev;
- int npages_allocated;
- u64 pbl_last_bt_num;
- u64 pbl_bt_cnt = 0;
- u64 size;
- int i;
-
- pbl_last_bt_num = (npages + pbl_bt_sz / 8 - 1) / (pbl_bt_sz / 8);
-
- /* alloc L1 BT */
- for (i = 0; i < pbl_bt_sz / 8; i++) {
- if (pbl_bt_cnt + 1 < pbl_last_bt_num) {
- size = pbl_bt_sz;
- } else {
- npages_allocated = i * (pbl_bt_sz / 8);
- size = (npages - npages_allocated) * 8;
- }
- mr->pbl_bt_l1[i] = dma_alloc_coherent(dev, size,
- &(mr->pbl_l1_dma_addr[i]),
- GFP_KERNEL);
- if (!mr->pbl_bt_l1[i]) {
- hns_roce_loop_free(hr_dev, mr, 1, i, 0);
- return -ENOMEM;
- }
-
- *(mr->pbl_bt_l0 + i) = mr->pbl_l1_dma_addr[i];
-
- pbl_bt_cnt++;
- if (pbl_bt_cnt >= pbl_last_bt_num)
- break;
}
- mr->l0_chunk_last_num = i + 1;
-
- return 0;
-}
-
-static int pbl_3hop_alloc(struct hns_roce_dev *hr_dev, int npages,
- struct hns_roce_mr *mr, u32 pbl_bt_sz)
-{
- struct device *dev = hr_dev->dev;
- int mr_alloc_done = 0;
- int npages_allocated;
- u64 pbl_last_bt_num;
- u64 pbl_bt_cnt = 0;
- u64 bt_idx;
- u64 size;
- int i;
- int j = 0;
-
- pbl_last_bt_num = (npages + pbl_bt_sz / 8 - 1) / (pbl_bt_sz / 8);
-
- mr->pbl_l2_dma_addr = kcalloc(pbl_last_bt_num,
- sizeof(*mr->pbl_l2_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_l2_dma_addr)
- return -ENOMEM;
-
- mr->pbl_bt_l2 = kcalloc(pbl_last_bt_num,
- sizeof(*mr->pbl_bt_l2),
- GFP_KERNEL);
- if (!mr->pbl_bt_l2)
- goto err_kcalloc_bt_l2;
-
- /* alloc L1, L2 BT */
- for (i = 0; i < pbl_bt_sz / 8; i++) {
- mr->pbl_bt_l1[i] = dma_alloc_coherent(dev, pbl_bt_sz,
- &(mr->pbl_l1_dma_addr[i]),
- GFP_KERNEL);
- if (!mr->pbl_bt_l1[i]) {
- hns_roce_loop_free(hr_dev, mr, 1, i, 0);
- goto err_dma_alloc_l0;
- }
-
- *(mr->pbl_bt_l0 + i) = mr->pbl_l1_dma_addr[i];
-
- for (j = 0; j < pbl_bt_sz / 8; j++) {
- bt_idx = i * pbl_bt_sz / 8 + j;
-
- if (pbl_bt_cnt + 1 < pbl_last_bt_num) {
- size = pbl_bt_sz;
- } else {
- npages_allocated = bt_idx *
- (pbl_bt_sz / 8);
- size = (npages - npages_allocated) * 8;
- }
- mr->pbl_bt_l2[bt_idx] = dma_alloc_coherent(
- dev, size,
- &(mr->pbl_l2_dma_addr[bt_idx]),
- GFP_KERNEL);
- if (!mr->pbl_bt_l2[bt_idx]) {
- hns_roce_loop_free(hr_dev, mr, 2, i, j);
- goto err_dma_alloc_l0;
- }
-
- *(mr->pbl_bt_l1[i] + j) =
- mr->pbl_l2_dma_addr[bt_idx];
-
- pbl_bt_cnt++;
- if (pbl_bt_cnt >= pbl_last_bt_num) {
- mr_alloc_done = 1;
- break;
- }
- }
+ mr->iova = iova; /* MR va starting addr */
+ mr->size = size; /* MR addr range */
+ mr->pd = pd; /* MR num */
+ mr->access = access; /* MR access permit */
+ mr->enabled = 0; /* MR active status */
+ mr->key = hw_index_to_key(obj); /* MR key */
- if (mr_alloc_done)
- break;
+ err = hns_roce_table_get(hr_dev, &hr_dev->mr_table.mtpt_table, obj);
+ if (err) {
+ ibdev_err(ibdev, "failed to alloc mtpt, ret = %d.\n", err);
+ goto err_free_bitmap;
}
- mr->l0_chunk_last_num = i + 1;
- mr->l1_chunk_last_num = j + 1;
-
-
return 0;
-
-err_dma_alloc_l0:
- kfree(mr->pbl_bt_l2);
- mr->pbl_bt_l2 = NULL;
-
-err_kcalloc_bt_l2:
- kfree(mr->pbl_l2_dma_addr);
- mr->pbl_l2_dma_addr = NULL;
-
- return -ENOMEM;
+err_free_bitmap:
+ hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap, obj, BITMAP_NO_RR);
+ return err;
}
-
-/* PBL multi hop addressing */
-static int hns_roce_mhop_alloc(struct hns_roce_dev *hr_dev, int npages,
- struct hns_roce_mr *mr)
+static void free_mr_key(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr)
{
- struct device *dev = hr_dev->dev;
- u32 pbl_bt_sz;
- u32 mhop_num;
-
- mhop_num = (mr->type == MR_TYPE_FRMR ? 1 : hr_dev->caps.pbl_hop_num);
- pbl_bt_sz = 1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT);
-
- if (mhop_num == HNS_ROCE_HOP_NUM_0)
- return 0;
-
- if (mhop_num == 1)
- return pbl_1hop_alloc(hr_dev, npages, mr, pbl_bt_sz);
-
- mr->pbl_l1_dma_addr = kcalloc(pbl_bt_sz / 8,
- sizeof(*mr->pbl_l1_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_l1_dma_addr)
- return -ENOMEM;
-
- mr->pbl_bt_l1 = kcalloc(pbl_bt_sz / 8, sizeof(*mr->pbl_bt_l1),
- GFP_KERNEL);
- if (!mr->pbl_bt_l1)
- goto err_kcalloc_bt_l1;
-
- /* alloc L0 BT */
- mr->pbl_bt_l0 = dma_alloc_coherent(dev, pbl_bt_sz,
- &(mr->pbl_l0_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_bt_l0)
- goto err_kcalloc_l2_dma;
-
- if (mhop_num == 2) {
- if (pbl_2hop_alloc(hr_dev, npages, mr, pbl_bt_sz))
- goto err_kcalloc_l2_dma;
- }
-
- if (mhop_num == 3) {
- if (pbl_3hop_alloc(hr_dev, npages, mr, pbl_bt_sz))
- goto err_kcalloc_l2_dma;
- }
-
+ unsigned long obj = key_to_hw_index(mr->key);
- mr->pbl_size = npages;
- mr->pbl_ba = mr->pbl_l0_dma_addr;
- mr->pbl_hop_num = hr_dev->caps.pbl_hop_num;
- mr->pbl_ba_pg_sz = hr_dev->caps.pbl_ba_pg_sz;
- mr->pbl_buf_pg_sz = hr_dev->caps.pbl_buf_pg_sz;
-
- return 0;
-
-err_kcalloc_l2_dma:
- kfree(mr->pbl_bt_l1);
- mr->pbl_bt_l1 = NULL;
-
-err_kcalloc_bt_l1:
- kfree(mr->pbl_l1_dma_addr);
- mr->pbl_l1_dma_addr = NULL;
-
- return -ENOMEM;
+ hns_roce_table_put(hr_dev, &hr_dev->mr_table.mtpt_table, obj);
+ hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap, obj, BITMAP_NO_RR);
}
-static int hns_roce_mr_alloc(struct hns_roce_dev *hr_dev, u32 pd, u64 iova,
- u64 size, u32 access, int npages,
- struct hns_roce_mr *mr)
+static int alloc_mr_pbl(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr,
+ size_t length, struct ib_udata *udata, u64 start,
+ int access)
{
- struct device *dev = hr_dev->dev;
- unsigned long index = 0;
- int ret;
-
- /* Allocate a key for mr from mr_table */
- ret = hns_roce_bitmap_alloc(&hr_dev->mr_table.mtpt_bitmap, &index);
- if (ret)
- return -ENOMEM;
-
- mr->iova = iova; /* MR va starting addr */
- mr->size = size; /* MR addr range */
- mr->pd = pd; /* MR num */
- mr->access = access; /* MR access permit */
- mr->enabled = 0; /* MR active status */
- mr->key = hw_index_to_key(index); /* MR key */
-
- if (size == ~0ull) {
- mr->pbl_buf = NULL;
- mr->pbl_dma_addr = 0;
- /* PBL multi-hop addressing parameters */
- mr->pbl_bt_l2 = NULL;
- mr->pbl_bt_l1 = NULL;
- mr->pbl_bt_l0 = NULL;
- mr->pbl_l2_dma_addr = NULL;
- mr->pbl_l1_dma_addr = NULL;
- mr->pbl_l0_dma_addr = 0;
- } else {
- if (!hr_dev->caps.pbl_hop_num) {
- mr->pbl_buf = dma_alloc_coherent(dev,
- npages * BA_BYTE_LEN,
- &(mr->pbl_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_buf)
- return -ENOMEM;
- } else {
- ret = hns_roce_mhop_alloc(hr_dev, npages, mr);
- }
- }
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ bool is_fast = mr->type == MR_TYPE_FRMR;
+ struct hns_roce_buf_attr buf_attr = {};
+ int err;
+
+ mr->pbl_hop_num = is_fast ? 1 : hr_dev->caps.pbl_hop_num;
+ buf_attr.page_shift = is_fast ? PAGE_SHIFT :
+ hr_dev->caps.pbl_buf_pg_sz + HNS_HW_PAGE_SHIFT;
+ buf_attr.region[0].size = length;
+ buf_attr.region[0].hopnum = mr->pbl_hop_num;
+ buf_attr.region_count = 1;
+ buf_attr.fixed_page = true;
+ buf_attr.user_access = access;
+ /* fast MR's buffer is alloced before mapping, not at creation */
+ buf_attr.mtt_only = is_fast;
+
+ err = hns_roce_mtr_create(hr_dev, &mr->pbl_mtr, &buf_attr,
+ hr_dev->caps.pbl_ba_pg_sz + HNS_HW_PAGE_SHIFT,
+ udata, start);
+ if (err)
+ ibdev_err(ibdev, "failed to alloc pbl mtr, ret = %d.\n", err);
+ else
+ mr->npages = mr->pbl_mtr.hem_cfg.buf_pg_count;
- return ret;
+ return err;
}
-static void hns_roce_mhop_free(struct hns_roce_dev *hr_dev,
- struct hns_roce_mr *mr)
+static void free_mr_pbl(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr)
{
- struct device *dev = hr_dev->dev;
- int npages_allocated;
- int npages;
- int i, j;
- u32 pbl_bt_sz;
- u32 mhop_num;
- u64 bt_idx;
-
- npages = mr->pbl_size;
- pbl_bt_sz = 1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT);
- mhop_num = (mr->type == MR_TYPE_FRMR) ? 1 : hr_dev->caps.pbl_hop_num;
-
- if (mhop_num == HNS_ROCE_HOP_NUM_0)
- return;
-
- if (mhop_num == 1) {
- dma_free_coherent(dev, (unsigned int)(npages * BA_BYTE_LEN),
- mr->pbl_buf, mr->pbl_dma_addr);
- return;
- }
-
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l0,
- mr->pbl_l0_dma_addr);
-
- if (mhop_num == 2) {
- for (i = 0; i < mr->l0_chunk_last_num; i++) {
- if (i == mr->l0_chunk_last_num - 1) {
- npages_allocated =
- i * (pbl_bt_sz / BA_BYTE_LEN);
-
- dma_free_coherent(dev,
- (npages - npages_allocated) * BA_BYTE_LEN,
- mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
-
- break;
- }
-
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
- }
- } else if (mhop_num == 3) {
- for (i = 0; i < mr->l0_chunk_last_num; i++) {
- dma_free_coherent(dev, pbl_bt_sz, mr->pbl_bt_l1[i],
- mr->pbl_l1_dma_addr[i]);
-
- for (j = 0; j < pbl_bt_sz / BA_BYTE_LEN; j++) {
- bt_idx = i * (pbl_bt_sz / BA_BYTE_LEN) + j;
-
- if ((i == mr->l0_chunk_last_num - 1)
- && j == mr->l1_chunk_last_num - 1) {
- npages_allocated = bt_idx *
- (pbl_bt_sz / BA_BYTE_LEN);
-
- dma_free_coherent(dev,
- (npages - npages_allocated) *
- BA_BYTE_LEN,
- mr->pbl_bt_l2[bt_idx],
- mr->pbl_l2_dma_addr[bt_idx]);
-
- break;
- }
-
- dma_free_coherent(dev, pbl_bt_sz,
- mr->pbl_bt_l2[bt_idx],
- mr->pbl_l2_dma_addr[bt_idx]);
- }
- }
- }
-
- kfree(mr->pbl_bt_l1);
- kfree(mr->pbl_l1_dma_addr);
- mr->pbl_bt_l1 = NULL;
- mr->pbl_l1_dma_addr = NULL;
- if (mhop_num == 3) {
- kfree(mr->pbl_bt_l2);
- kfree(mr->pbl_l2_dma_addr);
- mr->pbl_bt_l2 = NULL;
- mr->pbl_l2_dma_addr = NULL;
- }
+ hns_roce_mtr_destroy(hr_dev, &mr->pbl_mtr);
}
static void hns_roce_mr_free(struct hns_roce_dev *hr_dev,
struct hns_roce_mr *mr)
{
- struct device *dev = hr_dev->dev;
- int npages = 0;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
int ret;
if (mr->enabled) {
@@ -712,27 +156,12 @@ static void hns_roce_mr_free(struct hns_roce_dev *hr_dev,
key_to_hw_index(mr->key) &
(hr_dev->caps.num_mtpts - 1));
if (ret)
- dev_warn(dev, "DESTROY_MPT failed (%d)\n", ret);
+ ibdev_warn(ibdev, "failed to destroy mpt, ret = %d.\n",
+ ret);
}
- if (mr->size != ~0ULL) {
- if (mr->type == MR_TYPE_MR)
- npages = ib_umem_page_count(mr->umem);
-
- if (!hr_dev->caps.pbl_hop_num)
- dma_free_coherent(dev,
- (unsigned int)(npages * BA_BYTE_LEN),
- mr->pbl_buf, mr->pbl_dma_addr);
- else
- hns_roce_mhop_free(hr_dev, mr);
- }
-
- if (mr->enabled)
- hns_roce_table_put(hr_dev, &hr_dev->mr_table.mtpt_table,
- key_to_hw_index(mr->key));
-
- hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap,
- key_to_hw_index(mr->key), BITMAP_NO_RR);
+ free_mr_pbl(hr_dev, mr);
+ free_mr_key(hr_dev, mr);
}
static int hns_roce_mr_enable(struct hns_roce_dev *hr_dev,
@@ -742,18 +171,12 @@ static int hns_roce_mr_enable(struct hns_roce_dev *hr_dev,
unsigned long mtpt_idx = key_to_hw_index(mr->key);
struct device *dev = hr_dev->dev;
struct hns_roce_cmd_mailbox *mailbox;
- struct hns_roce_mr_table *mr_table = &hr_dev->mr_table;
-
- /* Prepare HEM entry memory */
- ret = hns_roce_table_get(hr_dev, &mr_table->mtpt_table, mtpt_idx);
- if (ret)
- return ret;
/* Allocate mailbox memory */
mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
if (IS_ERR(mailbox)) {
ret = PTR_ERR(mailbox);
- goto err_table;
+ return ret;
}
if (mr->type != MR_TYPE_FRMR)
@@ -780,137 +203,6 @@ static int hns_roce_mr_enable(struct hns_roce_dev *hr_dev,
err_page:
hns_roce_free_cmd_mailbox(hr_dev, mailbox);
-err_table:
- hns_roce_table_put(hr_dev, &mr_table->mtpt_table, mtpt_idx);
- return ret;
-}
-
-static int hns_roce_write_mtt_chunk(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, u32 start_index,
- u32 npages, u64 *page_list)
-{
- struct hns_roce_hem_table *table;
- dma_addr_t dma_handle;
- __le64 *mtts;
- u32 bt_page_size;
- u32 i;
-
- switch (mtt->mtt_type) {
- case MTT_TYPE_WQE:
- table = &hr_dev->mr_table.mtt_table;
- bt_page_size = 1 << (hr_dev->caps.mtt_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_CQE:
- table = &hr_dev->mr_table.mtt_cqe_table;
- bt_page_size = 1 << (hr_dev->caps.cqe_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_SRQWQE:
- table = &hr_dev->mr_table.mtt_srqwqe_table;
- bt_page_size = 1 << (hr_dev->caps.srqwqe_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_IDX:
- table = &hr_dev->mr_table.mtt_idx_table;
- bt_page_size = 1 << (hr_dev->caps.idx_ba_pg_sz + PAGE_SHIFT);
- break;
- default:
- return -EINVAL;
- }
-
- /* All MTTs must fit in the same page */
- if (start_index / (bt_page_size / sizeof(u64)) !=
- (start_index + npages - 1) / (bt_page_size / sizeof(u64)))
- return -EINVAL;
-
- if (start_index & (HNS_ROCE_MTT_ENTRY_PER_SEG - 1))
- return -EINVAL;
-
- mtts = hns_roce_table_find(hr_dev, table,
- mtt->first_seg +
- start_index / HNS_ROCE_MTT_ENTRY_PER_SEG,
- &dma_handle);
- if (!mtts)
- return -ENOMEM;
-
- /* Save page addr, low 12 bits : 0 */
- for (i = 0; i < npages; ++i) {
- if (!hr_dev->caps.mtt_hop_num)
- mtts[i] = cpu_to_le64(page_list[i] >> PAGE_ADDR_SHIFT);
- else
- mtts[i] = cpu_to_le64(page_list[i]);
- }
-
- return 0;
-}
-
-static int hns_roce_write_mtt(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, u32 start_index,
- u32 npages, u64 *page_list)
-{
- int chunk;
- int ret;
- u32 bt_page_size;
-
- if (mtt->order < 0)
- return -EINVAL;
-
- switch (mtt->mtt_type) {
- case MTT_TYPE_WQE:
- bt_page_size = 1 << (hr_dev->caps.mtt_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_CQE:
- bt_page_size = 1 << (hr_dev->caps.cqe_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_SRQWQE:
- bt_page_size = 1 << (hr_dev->caps.srqwqe_ba_pg_sz + PAGE_SHIFT);
- break;
- case MTT_TYPE_IDX:
- bt_page_size = 1 << (hr_dev->caps.idx_ba_pg_sz + PAGE_SHIFT);
- break;
- default:
- dev_err(hr_dev->dev,
- "Unsupport mtt type %d, write mtt failed\n",
- mtt->mtt_type);
- return -EINVAL;
- }
-
- while (npages > 0) {
- chunk = min_t(int, bt_page_size / sizeof(u64), npages);
-
- ret = hns_roce_write_mtt_chunk(hr_dev, mtt, start_index, chunk,
- page_list);
- if (ret)
- return ret;
-
- npages -= chunk;
- start_index += chunk;
- page_list += chunk;
- }
-
- return 0;
-}
-
-int hns_roce_buf_write_mtt(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, struct hns_roce_buf *buf)
-{
- u64 *page_list;
- int ret;
- u32 i;
-
- page_list = kmalloc_array(buf->npages, sizeof(*page_list), GFP_KERNEL);
- if (!page_list)
- return -ENOMEM;
-
- for (i = 0; i < buf->npages; ++i) {
- if (buf->nbufs == 1)
- page_list[i] = buf->direct.map + (i << buf->page_shift);
- else
- page_list[i] = buf->page_list[i].map;
-
- }
- ret = hns_roce_write_mtt(hr_dev, mtt, 0, buf->npages, page_list);
-
- kfree(page_list);
-
return ret;
}
@@ -923,50 +215,6 @@ int hns_roce_init_mr_table(struct hns_roce_dev *hr_dev)
hr_dev->caps.num_mtpts,
hr_dev->caps.num_mtpts - 1,
hr_dev->caps.reserved_mrws, 0);
- if (ret)
- return ret;
-
- ret = hns_roce_buddy_init(&mr_table->mtt_buddy,
- ilog2(hr_dev->caps.num_mtt_segs));
- if (ret)
- goto err_buddy;
-
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE)) {
- ret = hns_roce_buddy_init(&mr_table->mtt_cqe_buddy,
- ilog2(hr_dev->caps.num_cqe_segs));
- if (ret)
- goto err_buddy_cqe;
- }
-
- if (hr_dev->caps.num_srqwqe_segs) {
- ret = hns_roce_buddy_init(&mr_table->mtt_srqwqe_buddy,
- ilog2(hr_dev->caps.num_srqwqe_segs));
- if (ret)
- goto err_buddy_srqwqe;
- }
-
- if (hr_dev->caps.num_idx_segs) {
- ret = hns_roce_buddy_init(&mr_table->mtt_idx_buddy,
- ilog2(hr_dev->caps.num_idx_segs));
- if (ret)
- goto err_buddy_idx;
- }
-
- return 0;
-
-err_buddy_idx:
- if (hr_dev->caps.num_srqwqe_segs)
- hns_roce_buddy_cleanup(&mr_table->mtt_srqwqe_buddy);
-
-err_buddy_srqwqe:
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- hns_roce_buddy_cleanup(&mr_table->mtt_cqe_buddy);
-
-err_buddy_cqe:
- hns_roce_buddy_cleanup(&mr_table->mtt_buddy);
-
-err_buddy:
- hns_roce_bitmap_cleanup(&mr_table->mtpt_bitmap);
return ret;
}
@@ -974,30 +222,24 @@ void hns_roce_cleanup_mr_table(struct hns_roce_dev *hr_dev)
{
struct hns_roce_mr_table *mr_table = &hr_dev->mr_table;
- if (hr_dev->caps.num_idx_segs)
- hns_roce_buddy_cleanup(&mr_table->mtt_idx_buddy);
- if (hr_dev->caps.num_srqwqe_segs)
- hns_roce_buddy_cleanup(&mr_table->mtt_srqwqe_buddy);
- hns_roce_buddy_cleanup(&mr_table->mtt_buddy);
- if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE))
- hns_roce_buddy_cleanup(&mr_table->mtt_cqe_buddy);
hns_roce_bitmap_cleanup(&mr_table->mtpt_bitmap);
}
struct ib_mr *hns_roce_get_dma_mr(struct ib_pd *pd, int acc)
{
+ struct hns_roce_dev *hr_dev = to_hr_dev(pd->device);
struct hns_roce_mr *mr;
int ret;
- mr = kmalloc(sizeof(*mr), GFP_KERNEL);
+ mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (mr == NULL)
return ERR_PTR(-ENOMEM);
mr->type = MR_TYPE_DMA;
/* Allocate memory region key */
- ret = hns_roce_mr_alloc(to_hr_dev(pd->device), to_hr_pd(pd)->pdn, 0,
- ~0ULL, acc, 0, mr);
+ hns_roce_hem_list_init(&mr->pbl_mtr.hem_list);
+ ret = alloc_mr_key(hr_dev, mr, to_hr_pd(pd)->pdn, 0, 0, acc);
if (ret)
goto err_free;
@@ -1006,203 +248,52 @@ struct ib_mr *hns_roce_get_dma_mr(struct ib_pd *pd, int acc)
goto err_mr;
mr->ibmr.rkey = mr->ibmr.lkey = mr->key;
- mr->umem = NULL;
return &mr->ibmr;
-
err_mr:
- hns_roce_mr_free(to_hr_dev(pd->device), mr);
+ free_mr_key(hr_dev, mr);
err_free:
kfree(mr);
return ERR_PTR(ret);
}
-int hns_roce_ib_umem_write_mtt(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtt *mtt, struct ib_umem *umem)
-{
- struct device *dev = hr_dev->dev;
- struct sg_dma_page_iter sg_iter;
- unsigned int order;
- int npage = 0;
- int ret = 0;
- int i;
- u64 page_addr;
- u64 *pages;
- u32 bt_page_size;
- u32 n;
-
- switch (mtt->mtt_type) {
- case MTT_TYPE_WQE:
- order = hr_dev->caps.mtt_ba_pg_sz;
- break;
- case MTT_TYPE_CQE:
- order = hr_dev->caps.cqe_ba_pg_sz;
- break;
- case MTT_TYPE_SRQWQE:
- order = hr_dev->caps.srqwqe_ba_pg_sz;
- break;
- case MTT_TYPE_IDX:
- order = hr_dev->caps.idx_ba_pg_sz;
- break;
- default:
- dev_err(dev, "Unsupport mtt type %d, write mtt failed\n",
- mtt->mtt_type);
- return -EINVAL;
- }
-
- bt_page_size = 1 << (order + PAGE_SHIFT);
-
- pages = (u64 *) __get_free_pages(GFP_KERNEL, order);
- if (!pages)
- return -ENOMEM;
-
- i = n = 0;
-
- for_each_sg_dma_page(umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
- page_addr = sg_page_iter_dma_address(&sg_iter);
- if (!(npage % (1 << (mtt->page_shift - PAGE_SHIFT)))) {
- if (page_addr & ((1 << mtt->page_shift) - 1)) {
- dev_err(dev,
- "page_addr is not page_shift %d alignment!\n",
- mtt->page_shift);
- ret = -EINVAL;
- goto out;
- }
- pages[i++] = page_addr;
- }
- npage++;
- if (i == bt_page_size / sizeof(u64)) {
- ret = hns_roce_write_mtt(hr_dev, mtt, n, i, pages);
- if (ret)
- goto out;
- n += i;
- i = 0;
- }
- }
-
- if (i)
- ret = hns_roce_write_mtt(hr_dev, mtt, n, i, pages);
-
-out:
- free_pages((unsigned long) pages, order);
- return ret;
-}
-
-static int hns_roce_ib_umem_write_mr(struct hns_roce_dev *hr_dev,
- struct hns_roce_mr *mr,
- struct ib_umem *umem)
-{
- struct sg_dma_page_iter sg_iter;
- int i = 0, j = 0;
- u64 page_addr;
- u32 pbl_bt_sz;
-
- if (hr_dev->caps.pbl_hop_num == HNS_ROCE_HOP_NUM_0)
- return 0;
-
- pbl_bt_sz = 1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT);
- for_each_sg_dma_page(umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
- page_addr = sg_page_iter_dma_address(&sg_iter);
- if (!hr_dev->caps.pbl_hop_num) {
- /* for hip06, page addr is aligned to 4K */
- mr->pbl_buf[i++] = page_addr >> 12;
- } else if (hr_dev->caps.pbl_hop_num == 1) {
- mr->pbl_buf[i++] = page_addr;
- } else {
- if (hr_dev->caps.pbl_hop_num == 2)
- mr->pbl_bt_l1[i][j] = page_addr;
- else if (hr_dev->caps.pbl_hop_num == 3)
- mr->pbl_bt_l2[i][j] = page_addr;
-
- j++;
- if (j >= (pbl_bt_sz / BA_BYTE_LEN)) {
- i++;
- j = 0;
- }
- }
- }
-
- /* Memory barrier */
- mb();
-
- return 0;
-}
-
struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
u64 virt_addr, int access_flags,
struct ib_udata *udata)
{
struct hns_roce_dev *hr_dev = to_hr_dev(pd->device);
- struct device *dev = hr_dev->dev;
struct hns_roce_mr *mr;
- int bt_size;
int ret;
- int n;
- int i;
- mr = kmalloc(sizeof(*mr), GFP_KERNEL);
+ mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (!mr)
return ERR_PTR(-ENOMEM);
- mr->umem = ib_umem_get(pd->device, start, length, access_flags);
- if (IS_ERR(mr->umem)) {
- ret = PTR_ERR(mr->umem);
- goto err_free;
- }
-
- n = ib_umem_page_count(mr->umem);
-
- if (!hr_dev->caps.pbl_hop_num) {
- if (n > HNS_ROCE_MAX_MTPT_PBL_NUM) {
- dev_err(dev,
- " MR len %lld err. MR is limited to 4G at most!\n",
- length);
- ret = -EINVAL;
- goto err_umem;
- }
- } else {
- u64 pbl_size = 1;
-
- bt_size = (1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT)) /
- BA_BYTE_LEN;
- for (i = 0; i < hr_dev->caps.pbl_hop_num; i++)
- pbl_size *= bt_size;
- if (n > pbl_size) {
- dev_err(dev,
- " MR len %lld err. MR page num is limited to %lld!\n",
- length, pbl_size);
- ret = -EINVAL;
- goto err_umem;
- }
- }
-
mr->type = MR_TYPE_MR;
-
- ret = hns_roce_mr_alloc(hr_dev, to_hr_pd(pd)->pdn, virt_addr, length,
- access_flags, n, mr);
+ ret = alloc_mr_key(hr_dev, mr, to_hr_pd(pd)->pdn, virt_addr, length,
+ access_flags);
if (ret)
- goto err_umem;
+ goto err_alloc_mr;
- ret = hns_roce_ib_umem_write_mr(hr_dev, mr, mr->umem);
+ ret = alloc_mr_pbl(hr_dev, mr, length, udata, start, access_flags);
if (ret)
- goto err_mr;
+ goto err_alloc_key;
ret = hns_roce_mr_enable(hr_dev, mr);
if (ret)
- goto err_mr;
+ goto err_alloc_pbl;
mr->ibmr.rkey = mr->ibmr.lkey = mr->key;
+ mr->ibmr.length = length;
return &mr->ibmr;
-err_mr:
- hns_roce_mr_free(hr_dev, mr);
-
-err_umem:
- ib_umem_release(mr->umem);
-
-err_free:
+err_alloc_pbl:
+ free_mr_pbl(hr_dev, mr);
+err_alloc_key:
+ free_mr_key(hr_dev, mr);
+err_alloc_mr:
kfree(mr);
return ERR_PTR(ret);
}
@@ -1214,84 +305,36 @@ static int rereg_mr_trans(struct ib_mr *ibmr, int flags,
u32 pdn, struct ib_udata *udata)
{
struct hns_roce_dev *hr_dev = to_hr_dev(ibmr->device);
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct hns_roce_mr *mr = to_hr_mr(ibmr);
- struct device *dev = hr_dev->dev;
- int npages;
int ret;
- if (mr->size != ~0ULL) {
- npages = ib_umem_page_count(mr->umem);
-
- if (hr_dev->caps.pbl_hop_num)
- hns_roce_mhop_free(hr_dev, mr);
- else
- dma_free_coherent(dev, npages * 8,
- mr->pbl_buf, mr->pbl_dma_addr);
- }
- ib_umem_release(mr->umem);
-
- mr->umem = ib_umem_get(ibmr->device, start, length, mr_access_flags);
- if (IS_ERR(mr->umem)) {
- ret = PTR_ERR(mr->umem);
- mr->umem = NULL;
- return -ENOMEM;
- }
- npages = ib_umem_page_count(mr->umem);
-
- if (hr_dev->caps.pbl_hop_num) {
- ret = hns_roce_mhop_alloc(hr_dev, npages, mr);
- if (ret)
- goto release_umem;
- } else {
- mr->pbl_buf = dma_alloc_coherent(dev, npages * 8,
- &(mr->pbl_dma_addr),
- GFP_KERNEL);
- if (!mr->pbl_buf) {
- ret = -ENOMEM;
- goto release_umem;
- }
+ free_mr_pbl(hr_dev, mr);
+ ret = alloc_mr_pbl(hr_dev, mr, length, udata, start, mr_access_flags);
+ if (ret) {
+ ibdev_err(ibdev, "failed to create mr PBL, ret = %d.\n", ret);
+ return ret;
}
ret = hr_dev->hw->rereg_write_mtpt(hr_dev, mr, flags, pdn,
mr_access_flags, virt_addr,
length, mailbox->buf);
- if (ret)
- goto release_umem;
-
-
- ret = hns_roce_ib_umem_write_mr(hr_dev, mr, mr->umem);
if (ret) {
- if (mr->size != ~0ULL) {
- npages = ib_umem_page_count(mr->umem);
-
- if (hr_dev->caps.pbl_hop_num)
- hns_roce_mhop_free(hr_dev, mr);
- else
- dma_free_coherent(dev, npages * 8,
- mr->pbl_buf,
- mr->pbl_dma_addr);
- }
-
- goto release_umem;
+ ibdev_err(ibdev, "failed to write mtpt, ret = %d.\n", ret);
+ free_mr_pbl(hr_dev, mr);
}
- return 0;
-
-release_umem:
- ib_umem_release(mr->umem);
return ret;
-
}
-
int hns_roce_rereg_user_mr(struct ib_mr *ibmr, int flags, u64 start, u64 length,
u64 virt_addr, int mr_access_flags, struct ib_pd *pd,
struct ib_udata *udata)
{
struct hns_roce_dev *hr_dev = to_hr_dev(ibmr->device);
+ struct ib_device *ib_dev = &hr_dev->ib_dev;
struct hns_roce_mr *mr = to_hr_mr(ibmr);
struct hns_roce_cmd_mailbox *mailbox;
- struct device *dev = hr_dev->dev;
unsigned long mtpt_idx;
u32 pdn = 0;
int ret;
@@ -1312,7 +355,7 @@ int hns_roce_rereg_user_mr(struct ib_mr *ibmr, int flags, u64 start, u64 length,
ret = hns_roce_hw_destroy_mpt(hr_dev, NULL, mtpt_idx);
if (ret)
- dev_warn(dev, "DESTROY_MPT failed (%d)\n", ret);
+ ibdev_warn(ib_dev, "failed to destroy MPT, ret = %d.\n", ret);
mr->enabled = 0;
@@ -1336,8 +379,7 @@ int hns_roce_rereg_user_mr(struct ib_mr *ibmr, int flags, u64 start, u64 length,
ret = hns_roce_hw_create_mpt(hr_dev, mailbox, mtpt_idx);
if (ret) {
- dev_err(dev, "CREATE_MPT failed (%d)\n", ret);
- ib_umem_release(mr->umem);
+ ibdev_err(ib_dev, "failed to create MPT, ret = %d.\n", ret);
goto free_cmd_mbox;
}
@@ -1365,8 +407,6 @@ int hns_roce_dereg_mr(struct ib_mr *ibmr, struct ib_udata *udata)
ret = hr_dev->hw->dereg_mr(hr_dev, mr, udata);
} else {
hns_roce_mr_free(hr_dev, mr);
-
- ib_umem_release(mr->umem);
kfree(mr);
}
@@ -1380,12 +420,8 @@ struct ib_mr *hns_roce_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type,
struct device *dev = hr_dev->dev;
struct hns_roce_mr *mr;
u64 length;
- u32 page_size;
int ret;
- page_size = 1 << (hr_dev->caps.pbl_buf_pg_sz + PAGE_SHIFT);
- length = max_num_sg * page_size;
-
if (mr_type != IB_MR_TYPE_MEM_REG)
return ERR_PTR(-EINVAL);
@@ -1402,23 +438,28 @@ struct ib_mr *hns_roce_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type,
mr->type = MR_TYPE_FRMR;
/* Allocate memory region key */
- ret = hns_roce_mr_alloc(hr_dev, to_hr_pd(pd)->pdn, 0, length,
- 0, max_num_sg, mr);
+ length = max_num_sg * (1 << PAGE_SHIFT);
+ ret = alloc_mr_key(hr_dev, mr, to_hr_pd(pd)->pdn, 0, length, 0);
if (ret)
goto err_free;
+ ret = alloc_mr_pbl(hr_dev, mr, length, NULL, 0, 0);
+ if (ret)
+ goto err_key;
+
ret = hns_roce_mr_enable(hr_dev, mr);
if (ret)
- goto err_mr;
+ goto err_pbl;
mr->ibmr.rkey = mr->ibmr.lkey = mr->key;
- mr->umem = NULL;
+ mr->ibmr.length = length;
return &mr->ibmr;
-err_mr:
- hns_roce_mr_free(to_hr_dev(pd->device), mr);
-
+err_key:
+ free_mr_key(hr_dev, mr);
+err_pbl:
+ free_mr_pbl(hr_dev, mr);
err_free:
kfree(mr);
return ERR_PTR(ret);
@@ -1428,19 +469,54 @@ static int hns_roce_set_page(struct ib_mr *ibmr, u64 addr)
{
struct hns_roce_mr *mr = to_hr_mr(ibmr);
- mr->pbl_buf[mr->npages++] = addr;
+ if (likely(mr->npages < mr->pbl_mtr.hem_cfg.buf_pg_count)) {
+ mr->page_list[mr->npages++] = addr;
+ return 0;
+ }
- return 0;
+ return -ENOBUFS;
}
int hns_roce_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents,
unsigned int *sg_offset)
{
+ struct hns_roce_dev *hr_dev = to_hr_dev(ibmr->device);
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct hns_roce_mr *mr = to_hr_mr(ibmr);
+ struct hns_roce_mtr *mtr = &mr->pbl_mtr;
+ int ret = 0;
mr->npages = 0;
+ mr->page_list = kvcalloc(mr->pbl_mtr.hem_cfg.buf_pg_count,
+ sizeof(dma_addr_t), GFP_KERNEL);
+ if (!mr->page_list)
+ return ret;
+
+ ret = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, hns_roce_set_page);
+ if (ret < 1) {
+ ibdev_err(ibdev, "failed to store sg pages %d %d, cnt = %d.\n",
+ mr->npages, mr->pbl_mtr.hem_cfg.buf_pg_count, ret);
+ goto err_page_list;
+ }
+
+ mtr->hem_cfg.region[0].offset = 0;
+ mtr->hem_cfg.region[0].count = mr->npages;
+ mtr->hem_cfg.region[0].hopnum = mr->pbl_hop_num;
+ mtr->hem_cfg.region_count = 1;
+ ret = hns_roce_mtr_map(hr_dev, mtr, mr->page_list, mr->npages);
+ if (ret) {
+ ibdev_err(ibdev, "failed to map sg mtr, ret = %d.\n", ret);
+ ret = 0;
+ } else {
+ mr->pbl_mtr.hem_cfg.buf_pg_shift = ilog2(ibmr->page_size);
+ ret = mr->npages;
+ }
+
+err_page_list:
+ kvfree(mr->page_list);
+ mr->page_list = NULL;
- return ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, hns_roce_set_page);
+ return ret;
}
static void hns_roce_mw_free(struct hns_roce_dev *hr_dev,
@@ -1564,32 +640,23 @@ int hns_roce_dealloc_mw(struct ib_mw *ibmw)
return 0;
}
-void hns_roce_mtr_init(struct hns_roce_mtr *mtr, int bt_pg_shift,
- int buf_pg_shift)
-{
- hns_roce_hem_list_init(&mtr->hem_list, bt_pg_shift);
- mtr->buf_pg_shift = buf_pg_shift;
-}
-
-void hns_roce_mtr_cleanup(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtr *mtr)
-{
- hns_roce_hem_list_release(hr_dev, &mtr->hem_list);
-}
-
-static int hns_roce_write_mtr(struct hns_roce_dev *hr_dev,
- struct hns_roce_mtr *mtr, dma_addr_t *bufs,
- struct hns_roce_buf_region *r)
+static int mtr_map_region(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ dma_addr_t *pages, struct hns_roce_buf_region *region)
{
+ __le64 *mtts;
int offset;
int count;
int npage;
- u64 *mtts;
+ u64 addr;
int end;
int i;
- offset = r->offset;
- end = offset + r->count;
+ /* if hopnum is 0, buffer cannot store BAs, so skip write mtt */
+ if (!region->hopnum)
+ return 0;
+
+ offset = region->offset;
+ end = offset + region->count;
npage = 0;
while (offset < end) {
mtts = hns_roce_hem_list_find_mtt(hr_dev, &mtr->hem_list,
@@ -1597,13 +664,13 @@ static int hns_roce_write_mtr(struct hns_roce_dev *hr_dev,
if (!mtts)
return -ENOBUFS;
- /* Save page addr, low 12 bits : 0 */
for (i = 0; i < count; i++) {
if (hr_dev->hw_rev == HNS_ROCE_HW_VER1)
- mtts[i] = bufs[npage] >> PAGE_ADDR_SHIFT;
+ addr = to_hr_hw_page_addr(pages[npage]);
else
- mtts[i] = bufs[npage];
+ addr = pages[npage];
+ mtts[i] = cpu_to_le64(addr);
npage++;
}
offset += count;
@@ -1612,69 +679,416 @@ static int hns_roce_write_mtr(struct hns_roce_dev *hr_dev,
return 0;
}
-int hns_roce_mtr_attach(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
- dma_addr_t **bufs, struct hns_roce_buf_region *regions,
- int region_cnt)
+static inline bool mtr_has_mtt(struct hns_roce_buf_attr *attr)
{
- struct hns_roce_buf_region *r;
- int ret;
int i;
- ret = hns_roce_hem_list_request(hr_dev, &mtr->hem_list, regions,
- region_cnt);
- if (ret)
- return ret;
+ for (i = 0; i < attr->region_count; i++)
+ if (attr->region[i].hopnum != HNS_ROCE_HOP_NUM_0 &&
+ attr->region[i].hopnum > 0)
+ return true;
- for (i = 0; i < region_cnt; i++) {
- r = &regions[i];
- ret = hns_roce_write_mtr(hr_dev, mtr, bufs[i], r);
+ /* because the mtr only one root base address, when hopnum is 0 means
+ * root base address equals the first buffer address, thus all alloced
+ * memory must in a continuous space accessed by direct mode.
+ */
+ return false;
+}
+
+static inline size_t mtr_bufs_size(struct hns_roce_buf_attr *attr)
+{
+ size_t size = 0;
+ int i;
+
+ for (i = 0; i < attr->region_count; i++)
+ size += attr->region[i].size;
+
+ return size;
+}
+
+static inline int mtr_umem_page_count(struct ib_umem *umem,
+ unsigned int page_shift)
+{
+ int count = ib_umem_page_count(umem);
+
+ if (page_shift >= PAGE_SHIFT)
+ count >>= page_shift - PAGE_SHIFT;
+ else
+ count <<= PAGE_SHIFT - page_shift;
+
+ return count;
+}
+
+static inline size_t mtr_kmem_direct_size(bool is_direct, size_t alloc_size,
+ unsigned int page_shift)
+{
+ if (is_direct)
+ return ALIGN(alloc_size, 1 << page_shift);
+ else
+ return HNS_HW_DIRECT_PAGE_COUNT << page_shift;
+}
+
+/*
+ * check the given pages in continuous address space
+ * Returns 0 on success, or the error page num.
+ */
+static inline int mtr_check_direct_pages(dma_addr_t *pages, int page_count,
+ unsigned int page_shift)
+{
+ size_t page_size = 1 << page_shift;
+ int i;
+
+ for (i = 1; i < page_count; i++)
+ if (pages[i] - pages[i - 1] != page_size)
+ return i;
+
+ return 0;
+}
+
+static void mtr_free_bufs(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr)
+{
+ /* release user buffers */
+ if (mtr->umem) {
+ ib_umem_release(mtr->umem);
+ mtr->umem = NULL;
+ }
+
+ /* release kernel buffers */
+ if (mtr->kmem) {
+ hns_roce_buf_free(hr_dev, mtr->kmem);
+ kfree(mtr->kmem);
+ mtr->kmem = NULL;
+ }
+}
+
+static int mtr_alloc_bufs(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ struct hns_roce_buf_attr *buf_attr, bool is_direct,
+ struct ib_udata *udata, unsigned long user_addr)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ unsigned int max_pg_shift = buf_attr->page_shift;
+ unsigned int best_pg_shift = 0;
+ int all_pg_count = 0;
+ size_t direct_size;
+ size_t total_size;
+ unsigned long tmp;
+ int ret = 0;
+
+ total_size = mtr_bufs_size(buf_attr);
+ if (total_size < 1) {
+ ibdev_err(ibdev, "Failed to check mtr size\n");
+ return -EINVAL;
+ }
+
+ if (udata) {
+ mtr->kmem = NULL;
+ mtr->umem = ib_umem_get(ibdev, user_addr, total_size,
+ buf_attr->user_access);
+ if (IS_ERR_OR_NULL(mtr->umem)) {
+ ibdev_err(ibdev, "Failed to get umem, ret %ld\n",
+ PTR_ERR(mtr->umem));
+ return -ENOMEM;
+ }
+ if (buf_attr->fixed_page) {
+ best_pg_shift = max_pg_shift;
+ } else {
+ tmp = GENMASK(max_pg_shift, 0);
+ ret = ib_umem_find_best_pgsz(mtr->umem, tmp, user_addr);
+ best_pg_shift = (ret <= PAGE_SIZE) ?
+ PAGE_SHIFT : ilog2(ret);
+ }
+ all_pg_count = mtr_umem_page_count(mtr->umem, best_pg_shift);
+ ret = 0;
+ } else {
+ mtr->umem = NULL;
+ mtr->kmem = kzalloc(sizeof(*mtr->kmem), GFP_KERNEL);
+ if (!mtr->kmem) {
+ ibdev_err(ibdev, "Failed to alloc kmem\n");
+ return -ENOMEM;
+ }
+ direct_size = mtr_kmem_direct_size(is_direct, total_size,
+ max_pg_shift);
+ ret = hns_roce_buf_alloc(hr_dev, total_size, direct_size,
+ mtr->kmem, max_pg_shift);
if (ret) {
- dev_err(hr_dev->dev,
- "write mtr[%d/%d] err %d,offset=%d.\n",
- i, region_cnt, ret, r->offset);
- goto err_write;
+ ibdev_err(ibdev, "Failed to alloc kmem, ret %d\n", ret);
+ goto err_alloc_mem;
+ } else {
+ best_pg_shift = max_pg_shift;
+ all_pg_count = mtr->kmem->npages;
}
}
- return 0;
+ /* must bigger than minimum hardware page shift */
+ if (best_pg_shift < HNS_HW_PAGE_SHIFT || all_pg_count < 1) {
+ ret = -EINVAL;
+ ibdev_err(ibdev, "Failed to check mtr page shift %d count %d\n",
+ best_pg_shift, all_pg_count);
+ goto err_alloc_mem;
+ }
-err_write:
- hns_roce_hem_list_release(hr_dev, &mtr->hem_list);
+ mtr->hem_cfg.buf_pg_shift = best_pg_shift;
+ mtr->hem_cfg.buf_pg_count = all_pg_count;
+ return 0;
+err_alloc_mem:
+ mtr_free_bufs(hr_dev, mtr);
return ret;
}
+static int mtr_get_pages(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ dma_addr_t *pages, int count, unsigned int page_shift)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ int npage;
+ int err;
+
+ if (mtr->umem)
+ npage = hns_roce_get_umem_bufs(hr_dev, pages, count, 0,
+ mtr->umem, page_shift);
+ else
+ npage = hns_roce_get_kmem_bufs(hr_dev, pages, count, 0,
+ mtr->kmem);
+
+ if (mtr->hem_cfg.is_direct && npage > 1) {
+ err = mtr_check_direct_pages(pages, npage, page_shift);
+ if (err) {
+ ibdev_err(ibdev, "Failed to check %s direct page-%d\n",
+ mtr->umem ? "user" : "kernel", err);
+ npage = err;
+ }
+ }
+
+ return npage;
+}
+
+int hns_roce_mtr_map(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ dma_addr_t *pages, int page_cnt)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_buf_region *r;
+ int err;
+ int i;
+
+ for (i = 0; i < mtr->hem_cfg.region_count; i++) {
+ r = &mtr->hem_cfg.region[i];
+ if (r->offset + r->count > page_cnt) {
+ err = -EINVAL;
+ ibdev_err(ibdev,
+ "Failed to check mtr%d end %d + %d, max %d\n",
+ i, r->offset, r->count, page_cnt);
+ return err;
+ }
+
+ err = mtr_map_region(hr_dev, mtr, &pages[r->offset], r);
+ if (err) {
+ ibdev_err(ibdev,
+ "Failed to map mtr%d offset %d, err %d\n",
+ i, r->offset, err);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
int hns_roce_mtr_find(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
int offset, u64 *mtt_buf, int mtt_max, u64 *base_addr)
{
- u64 *mtts = mtt_buf;
int mtt_count;
int total = 0;
- u64 *addr;
+ __le64 *mtts;
int npage;
+ u64 addr;
int left;
- if (mtts == NULL || mtt_max < 1)
+ if (!mtt_buf || mtt_max < 1)
goto done;
+ /* no mtt memory in direct mode, so just return the buffer address */
+ if (mtr->hem_cfg.is_direct) {
+ npage = offset;
+ for (total = 0; total < mtt_max; total++, npage++) {
+ addr = mtr->hem_cfg.root_ba +
+ (npage << mtr->hem_cfg.buf_pg_shift);
+
+ if (hr_dev->hw_rev == HNS_ROCE_HW_VER1)
+ mtt_buf[total] = to_hr_hw_page_addr(addr);
+ else
+ mtt_buf[total] = addr;
+ }
+
+ goto done;
+ }
+
left = mtt_max;
while (left > 0) {
mtt_count = 0;
- addr = hns_roce_hem_list_find_mtt(hr_dev, &mtr->hem_list,
+ mtts = hns_roce_hem_list_find_mtt(hr_dev, &mtr->hem_list,
offset + total,
&mtt_count, NULL);
- if (!addr || !mtt_count)
+ if (!mtts || !mtt_count)
goto done;
npage = min(mtt_count, left);
- memcpy(&mtts[total], addr, BA_BYTE_LEN * npage);
left -= npage;
- total += npage;
+ for (mtt_count = 0; mtt_count < npage; mtt_count++)
+ mtt_buf[total++] = le64_to_cpu(mtts[mtt_count]);
}
done:
if (base_addr)
- *base_addr = mtr->hem_list.root_ba;
+ *base_addr = mtr->hem_cfg.root_ba;
return total;
}
+
+/* convert buffer size to page index and page count */
+static unsigned int mtr_init_region(struct hns_roce_buf_attr *attr,
+ int page_cnt,
+ struct hns_roce_buf_region *regions,
+ int region_cnt, unsigned int page_shift)
+{
+ unsigned int page_size = 1 << page_shift;
+ int max_region = attr->region_count;
+ struct hns_roce_buf_region *r;
+ unsigned int i = 0;
+ int page_idx = 0;
+
+ for (; i < region_cnt && i < max_region && page_idx < page_cnt; i++) {
+ r = &regions[i];
+ r->hopnum = attr->region[i].hopnum == HNS_ROCE_HOP_NUM_0 ?
+ 0 : attr->region[i].hopnum;
+ r->offset = page_idx;
+ r->count = DIV_ROUND_UP(attr->region[i].size, page_size);
+ page_idx += r->count;
+ }
+
+ return i;
+}
+
+/**
+ * hns_roce_mtr_create - Create hns memory translate region.
+ *
+ * @mtr: memory translate region
+ * @init_attr: init attribute for creating mtr
+ * @page_shift: page shift for multi-hop base address table
+ * @udata: user space context, if it's NULL, means kernel space
+ * @user_addr: userspace virtual address to start at
+ * @buf_alloced: mtr has private buffer, true means need to alloc
+ */
+int hns_roce_mtr_create(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr,
+ struct hns_roce_buf_attr *buf_attr,
+ unsigned int page_shift, struct ib_udata *udata,
+ unsigned long user_addr)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ dma_addr_t *pages = NULL;
+ int region_cnt = 0;
+ int all_pg_cnt;
+ int get_pg_cnt;
+ bool has_mtt;
+ int err = 0;
+
+ has_mtt = mtr_has_mtt(buf_attr);
+ /* if buffer only need mtt, just init the hem cfg */
+ if (buf_attr->mtt_only) {
+ mtr->hem_cfg.buf_pg_shift = buf_attr->page_shift;
+ mtr->hem_cfg.buf_pg_count = mtr_bufs_size(buf_attr) >>
+ buf_attr->page_shift;
+ mtr->umem = NULL;
+ mtr->kmem = NULL;
+ } else {
+ err = mtr_alloc_bufs(hr_dev, mtr, buf_attr, !has_mtt, udata,
+ user_addr);
+ if (err) {
+ ibdev_err(ibdev, "Failed to alloc mtr bufs, err %d\n",
+ err);
+ return err;
+ }
+ }
+
+ /* alloc mtt memory */
+ all_pg_cnt = mtr->hem_cfg.buf_pg_count;
+ hns_roce_hem_list_init(&mtr->hem_list);
+ mtr->hem_cfg.is_direct = !has_mtt;
+ mtr->hem_cfg.ba_pg_shift = page_shift;
+ mtr->hem_cfg.region_count = 0;
+ region_cnt = mtr_init_region(buf_attr, all_pg_cnt,
+ mtr->hem_cfg.region,
+ ARRAY_SIZE(mtr->hem_cfg.region),
+ mtr->hem_cfg.buf_pg_shift);
+ if (region_cnt < 1) {
+ err = -ENOBUFS;
+ ibdev_err(ibdev, "failed to init mtr region %d\n", region_cnt);
+ goto err_alloc_bufs;
+ }
+
+ mtr->hem_cfg.region_count = region_cnt;
+
+ if (has_mtt) {
+ err = hns_roce_hem_list_request(hr_dev, &mtr->hem_list,
+ mtr->hem_cfg.region, region_cnt,
+ page_shift);
+ if (err) {
+ ibdev_err(ibdev, "Failed to request mtr hem, err %d\n",
+ err);
+ goto err_alloc_bufs;
+ }
+ mtr->hem_cfg.root_ba = mtr->hem_list.root_ba;
+ }
+
+ /* no buffer to map */
+ if (buf_attr->mtt_only)
+ return 0;
+
+ /* alloc a tmp array to store buffer's dma address */
+ pages = kvcalloc(all_pg_cnt, sizeof(dma_addr_t), GFP_KERNEL);
+ if (!pages) {
+ err = -ENOMEM;
+ ibdev_err(ibdev, "Failed to alloc mtr page list %d\n",
+ all_pg_cnt);
+ goto err_alloc_hem_list;
+ }
+
+ get_pg_cnt = mtr_get_pages(hr_dev, mtr, pages, all_pg_cnt,
+ mtr->hem_cfg.buf_pg_shift);
+ if (get_pg_cnt != all_pg_cnt) {
+ ibdev_err(ibdev, "Failed to get mtr page %d != %d\n",
+ get_pg_cnt, all_pg_cnt);
+ err = -ENOBUFS;
+ goto err_alloc_page_list;
+ }
+
+ if (!has_mtt) {
+ mtr->hem_cfg.root_ba = pages[0];
+ } else {
+ /* write buffer's dma address to BA table */
+ err = hns_roce_mtr_map(hr_dev, mtr, pages, all_pg_cnt);
+ if (err) {
+ ibdev_err(ibdev, "Failed to map mtr pages, err %d\n",
+ err);
+ goto err_alloc_page_list;
+ }
+ }
+
+ /* drop tmp array */
+ kvfree(pages);
+ return 0;
+err_alloc_page_list:
+ kvfree(pages);
+err_alloc_hem_list:
+ hns_roce_hem_list_release(hr_dev, &mtr->hem_list);
+err_alloc_bufs:
+ mtr_free_bufs(hr_dev, mtr);
+ return err;
+}
+
+void hns_roce_mtr_destroy(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr)
+{
+ /* release multi-hop addressing resource */
+ hns_roce_hem_list_release(hr_dev, &mtr->hem_list);
+
+ /* free buffers */
+ mtr_free_bufs(hr_dev, mtr);
+}
diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c
index 6317901c4b4f..a0a47bd66975 100644
--- a/drivers/infiniband/hw/hns/hns_roce_qp.c
+++ b/drivers/infiniband/hw/hns/hns_roce_qp.c
@@ -355,16 +355,16 @@ static void free_qpn(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
hns_roce_bitmap_free_range(&qp_table->bitmap, hr_qp->qpn, 1, BITMAP_RR);
}
-static int set_rq_size(struct hns_roce_dev *hr_dev,
- struct ib_qp_cap *cap, bool is_user, int has_rq,
- struct hns_roce_qp *hr_qp)
+static int set_rq_size(struct hns_roce_dev *hr_dev, struct ib_qp_cap *cap,
+ struct hns_roce_qp *hr_qp, int has_rq)
{
- u32 max_cnt;
+ u32 cnt;
/* If srq exist, set zero for relative number of rq */
if (!has_rq) {
hr_qp->rq.wqe_cnt = 0;
hr_qp->rq.max_gs = 0;
+ hr_qp->rq_inl_buf.wqe_cnt = 0;
cap->max_recv_wr = 0;
cap->max_recv_sge = 0;
@@ -379,17 +379,15 @@ static int set_rq_size(struct hns_roce_dev *hr_dev,
return -EINVAL;
}
- max_cnt = max(cap->max_recv_wr, hr_dev->caps.min_wqes);
-
- hr_qp->rq.wqe_cnt = roundup_pow_of_two(max_cnt);
- if ((u32)hr_qp->rq.wqe_cnt > hr_dev->caps.max_wqes) {
+ cnt = roundup_pow_of_two(max(cap->max_recv_wr, hr_dev->caps.min_wqes));
+ if (cnt > hr_dev->caps.max_wqes) {
ibdev_err(&hr_dev->ib_dev, "rq depth %u too large\n",
cap->max_recv_wr);
return -EINVAL;
}
- max_cnt = max(1U, cap->max_recv_sge);
- hr_qp->rq.max_gs = roundup_pow_of_two(max_cnt);
+ hr_qp->rq.max_gs = roundup_pow_of_two(max(1U, cap->max_recv_sge) +
+ HNS_ROCE_RESERVED_SGE);
if (hr_dev->caps.max_rq_sg <= HNS_ROCE_SGE_IN_WQE)
hr_qp->rq.wqe_shift = ilog2(hr_dev->caps.max_rq_desc_sz);
@@ -397,8 +395,57 @@ static int set_rq_size(struct hns_roce_dev *hr_dev,
hr_qp->rq.wqe_shift = ilog2(hr_dev->caps.max_rq_desc_sz *
hr_qp->rq.max_gs);
- cap->max_recv_wr = hr_qp->rq.wqe_cnt;
- cap->max_recv_sge = hr_qp->rq.max_gs;
+ hr_qp->rq.wqe_cnt = cnt;
+ if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE)
+ hr_qp->rq_inl_buf.wqe_cnt = cnt;
+ else
+ hr_qp->rq_inl_buf.wqe_cnt = 0;
+
+ cap->max_recv_wr = cnt;
+ cap->max_recv_sge = hr_qp->rq.max_gs - HNS_ROCE_RESERVED_SGE;
+
+ return 0;
+}
+
+static int set_extend_sge_param(struct hns_roce_dev *hr_dev, u32 sq_wqe_cnt,
+ struct hns_roce_qp *hr_qp,
+ struct ib_qp_cap *cap)
+{
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ u32 cnt;
+
+ cnt = max(1U, cap->max_send_sge);
+ if (hr_dev->hw_rev == HNS_ROCE_HW_VER1) {
+ hr_qp->sq.max_gs = roundup_pow_of_two(cnt);
+ hr_qp->sge.sge_cnt = 0;
+
+ return 0;
+ }
+
+ hr_qp->sq.max_gs = cnt;
+
+ /* UD sqwqe's sge use extend sge */
+ if (hr_qp->ibqp.qp_type == IB_QPT_GSI ||
+ hr_qp->ibqp.qp_type == IB_QPT_UD) {
+ cnt = roundup_pow_of_two(sq_wqe_cnt * hr_qp->sq.max_gs);
+ } else if (hr_qp->sq.max_gs > HNS_ROCE_SGE_IN_WQE) {
+ cnt = roundup_pow_of_two(sq_wqe_cnt *
+ (hr_qp->sq.max_gs - HNS_ROCE_SGE_IN_WQE));
+
+ if (hr_dev->pci_dev->revision == PCI_REVISION_ID_HIP08_A) {
+ if (cnt > hr_dev->caps.max_extend_sg) {
+ ibdev_err(ibdev,
+ "failed to check exSGE num, exSGE num = %d.\n",
+ cnt);
+ return -EINVAL;
+ }
+ }
+ } else {
+ cnt = 0;
+ }
+
+ hr_qp->sge.sge_shift = HNS_ROCE_SGE_SHIFT;
+ hr_qp->sge.sge_cnt = cnt;
return 0;
}
@@ -430,174 +477,79 @@ static int set_user_sq_size(struct hns_roce_dev *hr_dev,
struct ib_qp_cap *cap, struct hns_roce_qp *hr_qp,
struct hns_roce_ib_create_qp *ucmd)
{
- u32 ex_sge_num;
- u32 page_size;
- u32 max_cnt;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ u32 cnt = 0;
int ret;
- if (check_shl_overflow(1, ucmd->log_sq_bb_count, &hr_qp->sq.wqe_cnt) ||
- hr_qp->sq.wqe_cnt > hr_dev->caps.max_wqes)
+ if (check_shl_overflow(1, ucmd->log_sq_bb_count, &cnt) ||
+ cnt > hr_dev->caps.max_wqes)
return -EINVAL;
ret = check_sq_size_with_integrity(hr_dev, cap, ucmd);
if (ret) {
- ibdev_err(&hr_dev->ib_dev, "Failed to check user SQ size limit\n");
+ ibdev_err(ibdev, "failed to check user SQ size, ret = %d.\n",
+ ret);
return ret;
}
- hr_qp->sq.wqe_shift = ucmd->log_sq_stride;
-
- max_cnt = max(1U, cap->max_send_sge);
- if (hr_dev->hw_rev == HNS_ROCE_HW_VER1)
- hr_qp->sq.max_gs = roundup_pow_of_two(max_cnt);
- else
- hr_qp->sq.max_gs = max_cnt;
-
- if (hr_qp->sq.max_gs > HNS_ROCE_SGE_IN_WQE)
- hr_qp->sge.sge_cnt = roundup_pow_of_two(hr_qp->sq.wqe_cnt *
- (hr_qp->sq.max_gs - 2));
-
- if (hr_qp->sq.max_gs > HNS_ROCE_SGE_IN_WQE &&
- hr_dev->pci_dev->revision == PCI_REVISION_ID_HIP08_A) {
- if (hr_qp->sge.sge_cnt > hr_dev->caps.max_extend_sg) {
- ibdev_err(&hr_dev->ib_dev,
- "Failed to check extended SGE size limit %d\n",
- hr_qp->sge.sge_cnt);
- return -EINVAL;
- }
- }
-
- hr_qp->sge.sge_shift = 4;
- ex_sge_num = hr_qp->sge.sge_cnt;
+ ret = set_extend_sge_param(hr_dev, cnt, hr_qp, cap);
+ if (ret)
+ return ret;
- /* Get buf size, SQ and RQ are aligned to page_szie */
- if (hr_dev->hw_rev == HNS_ROCE_HW_VER1) {
- hr_qp->buff_size = round_up((hr_qp->rq.wqe_cnt <<
- hr_qp->rq.wqe_shift), PAGE_SIZE) +
- round_up((hr_qp->sq.wqe_cnt <<
- hr_qp->sq.wqe_shift), PAGE_SIZE);
-
- hr_qp->sq.offset = 0;
- hr_qp->rq.offset = round_up((hr_qp->sq.wqe_cnt <<
- hr_qp->sq.wqe_shift), PAGE_SIZE);
- } else {
- page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
- hr_qp->sge.sge_cnt = ex_sge_num ?
- max(page_size / (1 << hr_qp->sge.sge_shift), ex_sge_num) : 0;
- hr_qp->buff_size = round_up((hr_qp->rq.wqe_cnt <<
- hr_qp->rq.wqe_shift), page_size) +
- round_up((hr_qp->sge.sge_cnt <<
- hr_qp->sge.sge_shift), page_size) +
- round_up((hr_qp->sq.wqe_cnt <<
- hr_qp->sq.wqe_shift), page_size);
-
- hr_qp->sq.offset = 0;
- if (ex_sge_num) {
- hr_qp->sge.offset = round_up((hr_qp->sq.wqe_cnt <<
- hr_qp->sq.wqe_shift),
- page_size);
- hr_qp->rq.offset = hr_qp->sge.offset +
- round_up((hr_qp->sge.sge_cnt <<
- hr_qp->sge.sge_shift),
- page_size);
- } else {
- hr_qp->rq.offset = round_up((hr_qp->sq.wqe_cnt <<
- hr_qp->sq.wqe_shift),
- page_size);
- }
- }
+ hr_qp->sq.wqe_shift = ucmd->log_sq_stride;
+ hr_qp->sq.wqe_cnt = cnt;
return 0;
}
-static int split_wqe_buf_region(struct hns_roce_dev *hr_dev,
- struct hns_roce_qp *hr_qp,
- struct hns_roce_buf_region *regions,
- int region_max, int page_shift)
+static int set_wqe_buf_attr(struct hns_roce_dev *hr_dev,
+ struct hns_roce_qp *hr_qp,
+ struct hns_roce_buf_attr *buf_attr)
{
- int page_size = 1 << page_shift;
- bool is_extend_sge;
- int region_cnt = 0;
int buf_size;
- int buf_cnt;
+ int idx = 0;
- if (hr_qp->buff_size < 1 || region_max < 1)
- return region_cnt;
+ hr_qp->buff_size = 0;
- if (hr_qp->sge.sge_cnt > 0)
- is_extend_sge = true;
- else
- is_extend_sge = false;
-
- /* sq region */
- if (is_extend_sge)
- buf_size = hr_qp->sge.offset - hr_qp->sq.offset;
- else
- buf_size = hr_qp->rq.offset - hr_qp->sq.offset;
-
- if (buf_size > 0 && region_cnt < region_max) {
- buf_cnt = DIV_ROUND_UP(buf_size, page_size);
- hns_roce_init_buf_region(&regions[region_cnt],
- hr_dev->caps.wqe_sq_hop_num,
- hr_qp->sq.offset / page_size,
- buf_cnt);
- region_cnt++;
- }
-
- /* sge region */
- if (is_extend_sge) {
- buf_size = hr_qp->rq.offset - hr_qp->sge.offset;
- if (buf_size > 0 && region_cnt < region_max) {
- buf_cnt = DIV_ROUND_UP(buf_size, page_size);
- hns_roce_init_buf_region(&regions[region_cnt],
- hr_dev->caps.wqe_sge_hop_num,
- hr_qp->sge.offset / page_size,
- buf_cnt);
- region_cnt++;
- }
- }
-
- /* rq region */
- buf_size = hr_qp->buff_size - hr_qp->rq.offset;
- if (buf_size > 0) {
- buf_cnt = DIV_ROUND_UP(buf_size, page_size);
- hns_roce_init_buf_region(&regions[region_cnt],
- hr_dev->caps.wqe_rq_hop_num,
- hr_qp->rq.offset / page_size,
- buf_cnt);
- region_cnt++;
- }
-
- return region_cnt;
-}
-
-static int set_extend_sge_param(struct hns_roce_dev *hr_dev,
- struct hns_roce_qp *hr_qp)
-{
- struct device *dev = hr_dev->dev;
-
- if (hr_qp->sq.max_gs > 2) {
- hr_qp->sge.sge_cnt = roundup_pow_of_two(hr_qp->sq.wqe_cnt *
- (hr_qp->sq.max_gs - 2));
- hr_qp->sge.sge_shift = 4;
- }
-
- /* ud sqwqe's sge use extend sge */
- if (hr_dev->hw_rev != HNS_ROCE_HW_VER1 &&
- hr_qp->ibqp.qp_type == IB_QPT_GSI) {
- hr_qp->sge.sge_cnt = roundup_pow_of_two(hr_qp->sq.wqe_cnt *
- hr_qp->sq.max_gs);
- hr_qp->sge.sge_shift = 4;
- }
+ /* SQ WQE */
+ hr_qp->sq.offset = 0;
+ buf_size = to_hr_hem_entries_size(hr_qp->sq.wqe_cnt,
+ hr_qp->sq.wqe_shift);
+ if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
+ buf_attr->region[idx].size = buf_size;
+ buf_attr->region[idx].hopnum = hr_dev->caps.wqe_sq_hop_num;
+ idx++;
+ hr_qp->buff_size += buf_size;
+ }
+
+ /* extend SGE WQE in SQ */
+ hr_qp->sge.offset = hr_qp->buff_size;
+ buf_size = to_hr_hem_entries_size(hr_qp->sge.sge_cnt,
+ hr_qp->sge.sge_shift);
+ if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
+ buf_attr->region[idx].size = buf_size;
+ buf_attr->region[idx].hopnum = hr_dev->caps.wqe_sge_hop_num;
+ idx++;
+ hr_qp->buff_size += buf_size;
+ }
+
+ /* RQ WQE */
+ hr_qp->rq.offset = hr_qp->buff_size;
+ buf_size = to_hr_hem_entries_size(hr_qp->rq.wqe_cnt,
+ hr_qp->rq.wqe_shift);
+ if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
+ buf_attr->region[idx].size = buf_size;
+ buf_attr->region[idx].hopnum = hr_dev->caps.wqe_rq_hop_num;
+ idx++;
+ hr_qp->buff_size += buf_size;
+ }
+
+ if (hr_qp->buff_size < 1)
+ return -EINVAL;
- if (hr_qp->sq.max_gs > 2 &&
- hr_dev->pci_dev->revision == PCI_REVISION_ID_HIP08_A) {
- if (hr_qp->sge.sge_cnt > hr_dev->caps.max_extend_sg) {
- dev_err(dev, "The extended sge cnt error! sge_cnt=%d\n",
- hr_qp->sge.sge_cnt);
- return -EINVAL;
- }
- }
+ buf_attr->page_shift = HNS_HW_PAGE_SHIFT + hr_dev->caps.mtt_buf_pg_sz;
+ buf_attr->fixed_page = true;
+ buf_attr->region_count = idx;
return 0;
}
@@ -605,62 +557,35 @@ static int set_extend_sge_param(struct hns_roce_dev *hr_dev,
static int set_kernel_sq_size(struct hns_roce_dev *hr_dev,
struct ib_qp_cap *cap, struct hns_roce_qp *hr_qp)
{
- u32 page_size;
- u32 max_cnt;
- int size;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ u32 cnt;
int ret;
if (!cap->max_send_wr || cap->max_send_wr > hr_dev->caps.max_wqes ||
cap->max_send_sge > hr_dev->caps.max_sq_sg ||
cap->max_inline_data > hr_dev->caps.max_sq_inline) {
- ibdev_err(&hr_dev->ib_dev,
- "SQ WR or sge or inline data error!\n");
+ ibdev_err(ibdev,
+ "failed to check SQ WR, SGE or inline num, ret = %d.\n",
+ -EINVAL);
return -EINVAL;
}
- hr_qp->sq.wqe_shift = ilog2(hr_dev->caps.max_sq_desc_sz);
-
- max_cnt = max(cap->max_send_wr, hr_dev->caps.min_wqes);
-
- hr_qp->sq.wqe_cnt = roundup_pow_of_two(max_cnt);
- if ((u32)hr_qp->sq.wqe_cnt > hr_dev->caps.max_wqes) {
- ibdev_err(&hr_dev->ib_dev,
- "while setting kernel sq size, sq.wqe_cnt too large\n");
+ cnt = roundup_pow_of_two(max(cap->max_send_wr, hr_dev->caps.min_wqes));
+ if (cnt > hr_dev->caps.max_wqes) {
+ ibdev_err(ibdev, "failed to check WQE num, WQE num = %d.\n",
+ cnt);
return -EINVAL;
}
- /* Get data_seg numbers */
- max_cnt = max(1U, cap->max_send_sge);
- if (hr_dev->hw_rev == HNS_ROCE_HW_VER1)
- hr_qp->sq.max_gs = roundup_pow_of_two(max_cnt);
- else
- hr_qp->sq.max_gs = max_cnt;
+ hr_qp->sq.wqe_shift = ilog2(hr_dev->caps.max_sq_desc_sz);
+ hr_qp->sq.wqe_cnt = cnt;
- ret = set_extend_sge_param(hr_dev, hr_qp);
- if (ret) {
- ibdev_err(&hr_dev->ib_dev, "set extend sge parameters fail\n");
+ ret = set_extend_sge_param(hr_dev, cnt, hr_qp, cap);
+ if (ret)
return ret;
- }
- /* Get buf size, SQ and RQ are aligned to PAGE_SIZE */
- page_size = 1 << (hr_dev->caps.mtt_buf_pg_sz + PAGE_SHIFT);
- hr_qp->sq.offset = 0;
- size = round_up(hr_qp->sq.wqe_cnt << hr_qp->sq.wqe_shift, page_size);
-
- if (hr_dev->hw_rev != HNS_ROCE_HW_VER1 && hr_qp->sge.sge_cnt) {
- hr_qp->sge.sge_cnt = max(page_size/(1 << hr_qp->sge.sge_shift),
- (u32)hr_qp->sge.sge_cnt);
- hr_qp->sge.offset = size;
- size += round_up(hr_qp->sge.sge_cnt << hr_qp->sge.sge_shift,
- page_size);
- }
-
- hr_qp->rq.offset = size;
- size += round_up((hr_qp->rq.wqe_cnt << hr_qp->rq.wqe_shift), page_size);
- hr_qp->buff_size = size;
-
- /* Get wr and sge number which send */
- cap->max_send_wr = hr_qp->sq.wqe_cnt;
+ /* sync the parameters of kernel QP to user's configuration */
+ cap->max_send_wr = cnt;
cap->max_send_sge = hr_qp->sq.max_gs;
/* We don't support inline sends for kernel QPs (yet) */
@@ -691,8 +616,8 @@ static int alloc_rq_inline_buf(struct hns_roce_qp *hr_qp,
struct ib_qp_init_attr *init_attr)
{
u32 max_recv_sge = init_attr->cap.max_recv_sge;
+ u32 wqe_cnt = hr_qp->rq_inl_buf.wqe_cnt;
struct hns_roce_rinl_wqe *wqe_list;
- u32 wqe_cnt = hr_qp->rq.wqe_cnt;
int i;
/* allocate recv inline buf */
@@ -714,7 +639,6 @@ static int alloc_rq_inline_buf(struct hns_roce_qp *hr_qp,
wqe_list[i].sg_list = &wqe_list[0].sg_list[i * max_recv_sge];
hr_qp->rq_inl_buf.wqe_list = wqe_list;
- hr_qp->rq_inl_buf.wqe_cnt = wqe_cnt;
return 0;
@@ -727,140 +651,55 @@ err:
static void free_rq_inline_buf(struct hns_roce_qp *hr_qp)
{
- kfree(hr_qp->rq_inl_buf.wqe_list[0].sg_list);
+ if (hr_qp->rq_inl_buf.wqe_list)
+ kfree(hr_qp->rq_inl_buf.wqe_list[0].sg_list);
kfree(hr_qp->rq_inl_buf.wqe_list);
}
-static int map_wqe_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
- u32 page_shift, bool is_user)
-{
-/* WQE buffer include 3 parts: SQ, extend SGE and RQ. */
-#define HNS_ROCE_WQE_REGION_MAX 3
- struct hns_roce_buf_region regions[HNS_ROCE_WQE_REGION_MAX] = {};
- dma_addr_t *buf_list[HNS_ROCE_WQE_REGION_MAX] = {};
- struct ib_device *ibdev = &hr_dev->ib_dev;
- struct hns_roce_buf_region *r;
- int region_count;
- int buf_count;
- int ret;
- int i;
-
- region_count = split_wqe_buf_region(hr_dev, hr_qp, regions,
- ARRAY_SIZE(regions), page_shift);
-
- /* alloc a tmp list to store WQE buffers address */
- ret = hns_roce_alloc_buf_list(regions, buf_list, region_count);
- if (ret) {
- ibdev_err(ibdev, "Failed to alloc WQE buffer list\n");
- return ret;
- }
-
- for (i = 0; i < region_count; i++) {
- r = &regions[i];
- if (is_user)
- buf_count = hns_roce_get_umem_bufs(hr_dev, buf_list[i],
- r->count, r->offset, hr_qp->umem,
- page_shift);
- else
- buf_count = hns_roce_get_kmem_bufs(hr_dev, buf_list[i],
- r->count, r->offset, &hr_qp->hr_buf);
-
- if (buf_count != r->count) {
- ibdev_err(ibdev, "Failed to get %s WQE buf, expect %d = %d.\n",
- is_user ? "user" : "kernel",
- r->count, buf_count);
- ret = -ENOBUFS;
- goto done;
- }
- }
-
- hr_qp->wqe_bt_pg_shift = hr_dev->caps.mtt_ba_pg_sz;
- hns_roce_mtr_init(&hr_qp->mtr, PAGE_SHIFT + hr_qp->wqe_bt_pg_shift,
- page_shift);
- ret = hns_roce_mtr_attach(hr_dev, &hr_qp->mtr, buf_list, regions,
- region_count);
- if (ret)
- ibdev_err(ibdev, "Failed to attach WQE's mtr\n");
-
- goto done;
-
- hns_roce_mtr_cleanup(hr_dev, &hr_qp->mtr);
-done:
- hns_roce_free_buf_list(buf_list, region_count);
-
- return ret;
-}
-
static int alloc_qp_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
struct ib_qp_init_attr *init_attr,
struct ib_udata *udata, unsigned long addr)
{
- u32 page_shift = PAGE_SHIFT + hr_dev->caps.mtt_buf_pg_sz;
struct ib_device *ibdev = &hr_dev->ib_dev;
- bool is_rq_buf_inline;
+ struct hns_roce_buf_attr buf_attr = {};
int ret;
- is_rq_buf_inline = (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) &&
- hns_roce_qp_has_rq(init_attr);
- if (is_rq_buf_inline) {
+ if (!udata && hr_qp->rq_inl_buf.wqe_cnt) {
ret = alloc_rq_inline_buf(hr_qp, init_attr);
if (ret) {
- ibdev_err(ibdev, "Failed to alloc inline RQ buffer\n");
+ ibdev_err(ibdev,
+ "failed to alloc inline buf, ret = %d.\n",
+ ret);
return ret;
}
- }
-
- if (udata) {
- hr_qp->umem = ib_umem_get(ibdev, addr, hr_qp->buff_size, 0);
- if (IS_ERR(hr_qp->umem)) {
- ret = PTR_ERR(hr_qp->umem);
- goto err_inline;
- }
} else {
- ret = hns_roce_buf_alloc(hr_dev, hr_qp->buff_size,
- (1 << page_shift) * 2,
- &hr_qp->hr_buf, page_shift);
- if (ret)
- goto err_inline;
+ hr_qp->rq_inl_buf.wqe_list = NULL;
}
- ret = map_wqe_buf(hr_dev, hr_qp, page_shift, udata);
- if (ret)
- goto err_alloc;
+ ret = set_wqe_buf_attr(hr_dev, hr_qp, &buf_attr);
+ if (ret) {
+ ibdev_err(ibdev, "failed to split WQE buf, ret = %d.\n", ret);
+ goto err_inline;
+ }
+ ret = hns_roce_mtr_create(hr_dev, &hr_qp->mtr, &buf_attr,
+ HNS_HW_PAGE_SHIFT + hr_dev->caps.mtt_ba_pg_sz,
+ udata, addr);
+ if (ret) {
+ ibdev_err(ibdev, "failed to create WQE mtr, ret = %d.\n", ret);
+ goto err_inline;
+ }
return 0;
-
err_inline:
- if (is_rq_buf_inline)
- free_rq_inline_buf(hr_qp);
-
-err_alloc:
- if (udata) {
- ib_umem_release(hr_qp->umem);
- hr_qp->umem = NULL;
- } else {
- hns_roce_buf_free(hr_dev, hr_qp->buff_size, &hr_qp->hr_buf);
- }
-
- ibdev_err(ibdev, "Failed to alloc WQE buffer, ret %d.\n", ret);
+ free_rq_inline_buf(hr_qp);
return ret;
}
static void free_qp_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
- hns_roce_mtr_cleanup(hr_dev, &hr_qp->mtr);
- if (hr_qp->umem) {
- ib_umem_release(hr_qp->umem);
- hr_qp->umem = NULL;
- }
-
- if (hr_qp->hr_buf.nbufs > 0)
- hns_roce_buf_free(hr_dev, hr_qp->buff_size, &hr_qp->hr_buf);
-
- if ((hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) &&
- hr_qp->rq.wqe_cnt)
- free_rq_inline_buf(hr_qp);
+ hns_roce_mtr_destroy(hr_dev, &hr_qp->mtr);
+ free_rq_inline_buf(hr_qp);
}
static inline bool user_qp_has_sdb(struct hns_roce_dev *hr_dev,
@@ -912,8 +751,8 @@ static int alloc_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
"Failed to map user SQ doorbell\n");
goto err_out;
}
- hr_qp->sdb_en = 1;
- resp->cap_flags |= HNS_ROCE_SUPPORT_SQ_RECORD_DB;
+ hr_qp->en_flags |= HNS_ROCE_QP_CAP_SQ_RECORD_DB;
+ resp->cap_flags |= HNS_ROCE_QP_CAP_SQ_RECORD_DB;
}
if (user_qp_has_rdb(hr_dev, init_attr, udata, resp)) {
@@ -924,8 +763,8 @@ static int alloc_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
"Failed to map user RQ doorbell\n");
goto err_sdb;
}
- hr_qp->rdb_en = 1;
- resp->cap_flags |= HNS_ROCE_SUPPORT_RQ_RECORD_DB;
+ hr_qp->en_flags |= HNS_ROCE_QP_CAP_RQ_RECORD_DB;
+ resp->cap_flags |= HNS_ROCE_QP_CAP_RQ_RECORD_DB;
}
} else {
/* QP doorbell register address */
@@ -942,13 +781,13 @@ static int alloc_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
goto err_out;
}
*hr_qp->rdb.db_record = 0;
- hr_qp->rdb_en = 1;
+ hr_qp->en_flags |= HNS_ROCE_QP_CAP_RQ_RECORD_DB;
}
}
return 0;
err_sdb:
- if (udata && hr_qp->sdb_en)
+ if (udata && hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB)
hns_roce_db_unmap_user(uctx, &hr_qp->sdb);
err_out:
return ret;
@@ -961,12 +800,12 @@ static void free_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
udata, struct hns_roce_ucontext, ibucontext);
if (udata) {
- if (hr_qp->rdb_en)
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
hns_roce_db_unmap_user(uctx, &hr_qp->rdb);
- if (hr_qp->sdb_en)
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB)
hns_roce_db_unmap_user(uctx, &hr_qp->sdb);
} else {
- if (hr_qp->rdb_en)
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
hns_roce_free_db(hr_dev, &hr_qp->rdb);
}
}
@@ -1003,8 +842,7 @@ err_sq:
return ret;
}
-static void free_kernel_wrid(struct hns_roce_dev *hr_dev,
- struct hns_roce_qp *hr_qp)
+static void free_kernel_wrid(struct hns_roce_qp *hr_qp)
{
kfree(hr_qp->rq.wrid);
kfree(hr_qp->sq.wrid);
@@ -1025,10 +863,11 @@ static int set_qp_param(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
else
hr_qp->sq_signal_bits = IB_SIGNAL_REQ_WR;
- ret = set_rq_size(hr_dev, &init_attr->cap, udata,
- hns_roce_qp_has_rq(init_attr), hr_qp);
+ ret = set_rq_size(hr_dev, &init_attr->cap, hr_qp,
+ hns_roce_qp_has_rq(init_attr));
if (ret) {
- ibdev_err(ibdev, "Failed to set user RQ size\n");
+ ibdev_err(ibdev, "failed to set user RQ size, ret = %d.\n",
+ ret);
return ret;
}
@@ -1156,7 +995,7 @@ err_buf:
err_db:
free_qp_db(hr_dev, hr_qp, udata);
err_wrid:
- free_kernel_wrid(hr_dev, hr_qp);
+ free_kernel_wrid(hr_qp);
return ret;
}
@@ -1170,7 +1009,7 @@ void hns_roce_qp_destroy(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
free_qpc(hr_dev, hr_qp);
free_qpn(hr_dev, hr_qp);
free_qp_buf(hr_dev, hr_qp);
- free_kernel_wrid(hr_dev, hr_qp);
+ free_kernel_wrid(hr_qp);
free_qp_db(hr_dev, hr_qp, udata);
kfree(hr_qp);
@@ -1339,10 +1178,10 @@ int hns_roce_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
if (ibqp->uobject &&
(attr_mask & IB_QP_STATE) && new_state == IB_QPS_ERR) {
- if (hr_qp->sdb_en == 1) {
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB) {
hr_qp->sq.head = *(int *)(hr_qp->sdb.virt_addr);
- if (hr_qp->rdb_en == 1)
+ if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
hr_qp->rq.head = *(int *)(hr_qp->rdb.virt_addr);
} else {
ibdev_warn(&hr_dev->ib_dev,
@@ -1431,10 +1270,9 @@ void hns_roce_unlock_cqs(struct hns_roce_cq *send_cq,
}
}
-static void *get_wqe(struct hns_roce_qp *hr_qp, int offset)
+static inline void *get_wqe(struct hns_roce_qp *hr_qp, int offset)
{
-
- return hns_roce_buf_offset(&hr_qp->hr_buf, offset);
+ return hns_roce_buf_offset(hr_qp->mtr.kmem, offset);
}
void *hns_roce_get_recv_wqe(struct hns_roce_qp *hr_qp, int n)
@@ -1449,8 +1287,7 @@ void *hns_roce_get_send_wqe(struct hns_roce_qp *hr_qp, int n)
void *hns_roce_get_extend_sge(struct hns_roce_qp *hr_qp, int n)
{
- return hns_roce_buf_offset(&hr_qp->hr_buf, hr_qp->sge.offset +
- (n << hr_qp->sge.sge_shift));
+ return get_wqe(hr_qp, hr_qp->sge.offset + (n << hr_qp->sge.sge_shift));
}
bool hns_roce_wq_overflow(struct hns_roce_wq *hr_wq, int nreq,
diff --git a/drivers/infiniband/hw/hns/hns_roce_srq.c b/drivers/infiniband/hw/hns/hns_roce_srq.c
index 5b3dd1a337d4..f40a000e94ee 100644
--- a/drivers/infiniband/hw/hns/hns_roce_srq.c
+++ b/drivers/infiniband/hw/hns/hns_roce_srq.c
@@ -77,56 +77,56 @@ static int hns_roce_hw_destroy_srq(struct hns_roce_dev *dev,
HNS_ROCE_CMD_TIMEOUT_MSECS);
}
-static int hns_roce_srq_alloc(struct hns_roce_dev *hr_dev, u32 pdn, u32 cqn,
- u16 xrcd, struct hns_roce_mtt *hr_mtt,
- u64 db_rec_addr, struct hns_roce_srq *srq)
+static int alloc_srqc(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq,
+ u32 pdn, u32 cqn, u16 xrcd, u64 db_rec_addr)
{
struct hns_roce_srq_table *srq_table = &hr_dev->srq_table;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
struct hns_roce_cmd_mailbox *mailbox;
- dma_addr_t dma_handle_wqe;
- dma_addr_t dma_handle_idx;
- u64 *mtts_wqe;
- u64 *mtts_idx;
+ u64 mtts_wqe[MTT_MIN_COUNT] = { 0 };
+ u64 mtts_idx[MTT_MIN_COUNT] = { 0 };
+ dma_addr_t dma_handle_wqe = 0;
+ dma_addr_t dma_handle_idx = 0;
int ret;
/* Get the physical address of srq buf */
- mtts_wqe = hns_roce_table_find(hr_dev,
- &hr_dev->mr_table.mtt_srqwqe_table,
- srq->mtt.first_seg,
- &dma_handle_wqe);
- if (!mtts_wqe) {
- dev_err(hr_dev->dev, "Failed to find mtt for srq buf.\n");
- return -EINVAL;
+ ret = hns_roce_mtr_find(hr_dev, &srq->buf_mtr, 0, mtts_wqe,
+ ARRAY_SIZE(mtts_wqe), &dma_handle_wqe);
+ if (ret < 1) {
+ ibdev_err(ibdev, "Failed to find mtr for SRQ WQE\n");
+ return -ENOBUFS;
}
/* Get physical address of idx que buf */
- mtts_idx = hns_roce_table_find(hr_dev, &hr_dev->mr_table.mtt_idx_table,
- srq->idx_que.mtt.first_seg,
- &dma_handle_idx);
- if (!mtts_idx) {
- dev_err(hr_dev->dev,
- "Failed to find mtt for srq idx queue buf.\n");
- return -EINVAL;
+ ret = hns_roce_mtr_find(hr_dev, &srq->idx_que.mtr, 0, mtts_idx,
+ ARRAY_SIZE(mtts_idx), &dma_handle_idx);
+ if (ret < 1) {
+ ibdev_err(ibdev, "Failed to find mtr for SRQ idx\n");
+ return -ENOBUFS;
}
ret = hns_roce_bitmap_alloc(&srq_table->bitmap, &srq->srqn);
if (ret) {
- dev_err(hr_dev->dev,
- "Failed to alloc a bit from srq bitmap.\n");
+ ibdev_err(ibdev, "Failed to alloc SRQ number, err %d\n", ret);
return -ENOMEM;
}
ret = hns_roce_table_get(hr_dev, &srq_table->table, srq->srqn);
- if (ret)
+ if (ret) {
+ ibdev_err(ibdev, "Failed to get SRQC table, err %d\n", ret);
goto err_out;
+ }
ret = xa_err(xa_store(&srq_table->xa, srq->srqn, srq, GFP_KERNEL));
- if (ret)
+ if (ret) {
+ ibdev_err(ibdev, "Failed to store SRQC, err %d\n", ret);
goto err_put;
+ }
mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
- if (IS_ERR(mailbox)) {
- ret = PTR_ERR(mailbox);
+ if (IS_ERR_OR_NULL(mailbox)) {
+ ret = -ENOMEM;
+ ibdev_err(ibdev, "Failed to alloc mailbox for SRQC\n");
goto err_xa;
}
@@ -136,8 +136,10 @@ static int hns_roce_srq_alloc(struct hns_roce_dev *hr_dev, u32 pdn, u32 cqn,
ret = hns_roce_hw_create_srq(hr_dev, mailbox, srq->srqn);
hns_roce_free_cmd_mailbox(hr_dev, mailbox);
- if (ret)
+ if (ret) {
+ ibdev_err(ibdev, "Failed to config SRQC, err %d\n", ret);
goto err_xa;
+ }
atomic_set(&srq->refcount, 1);
init_completion(&srq->free);
@@ -154,8 +156,7 @@ err_out:
return ret;
}
-static void hns_roce_srq_free(struct hns_roce_dev *hr_dev,
- struct hns_roce_srq *srq)
+static void free_srqc(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq)
{
struct hns_roce_srq_table *srq_table = &hr_dev->srq_table;
int ret;
@@ -175,187 +176,104 @@ static void hns_roce_srq_free(struct hns_roce_dev *hr_dev,
hns_roce_bitmap_free(&srq_table->bitmap, srq->srqn, BITMAP_NO_RR);
}
-static int create_user_srq(struct hns_roce_srq *srq, struct ib_udata *udata,
- int srq_buf_size)
+static int alloc_srq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq,
+ struct ib_udata *udata, unsigned long addr)
{
- struct hns_roce_dev *hr_dev = to_hr_dev(srq->ibsrq.device);
- struct hns_roce_ib_create_srq ucmd;
- struct hns_roce_buf *buf;
- int ret;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_buf_attr buf_attr = {};
+ int err;
+
+ srq->wqe_shift = ilog2(roundup_pow_of_two(max(HNS_ROCE_SGE_SIZE,
+ HNS_ROCE_SGE_SIZE *
+ srq->max_gs)));
+
+ buf_attr.page_shift = hr_dev->caps.srqwqe_buf_pg_sz + HNS_HW_PAGE_SHIFT;
+ buf_attr.region[0].size = to_hr_hem_entries_size(srq->wqe_cnt,
+ srq->wqe_shift);
+ buf_attr.region[0].hopnum = hr_dev->caps.srqwqe_hop_num;
+ buf_attr.region_count = 1;
+ buf_attr.fixed_page = true;
+
+ err = hns_roce_mtr_create(hr_dev, &srq->buf_mtr, &buf_attr,
+ hr_dev->caps.srqwqe_ba_pg_sz +
+ HNS_HW_PAGE_SHIFT, udata, addr);
+ if (err)
+ ibdev_err(ibdev, "Failed to alloc SRQ buf mtr, err %d\n", err);
+
+ return err;
+}
- if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd)))
- return -EFAULT;
-
- srq->umem =
- ib_umem_get(srq->ibsrq.device, ucmd.buf_addr, srq_buf_size, 0);
- if (IS_ERR(srq->umem))
- return PTR_ERR(srq->umem);
-
- buf = &srq->buf;
- buf->npages = (ib_umem_page_count(srq->umem) +
- (1 << hr_dev->caps.srqwqe_buf_pg_sz) - 1) /
- (1 << hr_dev->caps.srqwqe_buf_pg_sz);
- buf->page_shift = PAGE_SHIFT + hr_dev->caps.srqwqe_buf_pg_sz;
- ret = hns_roce_mtt_init(hr_dev, buf->npages, buf->page_shift,
- &srq->mtt);
- if (ret)
- goto err_user_buf;
+static void free_srq_buf(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq)
+{
+ hns_roce_mtr_destroy(hr_dev, &srq->buf_mtr);
+}
- ret = hns_roce_ib_umem_write_mtt(hr_dev, &srq->mtt, srq->umem);
- if (ret)
- goto err_user_srq_mtt;
-
- /* config index queue BA */
- srq->idx_que.umem = ib_umem_get(srq->ibsrq.device, ucmd.que_addr,
- srq->idx_que.buf_size, 0);
- if (IS_ERR(srq->idx_que.umem)) {
- dev_err(hr_dev->dev, "ib_umem_get error for index queue\n");
- ret = PTR_ERR(srq->idx_que.umem);
- goto err_user_srq_mtt;
+static int alloc_srq_idx(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq,
+ struct ib_udata *udata, unsigned long addr)
+{
+ struct hns_roce_idx_que *idx_que = &srq->idx_que;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_buf_attr buf_attr = {};
+ int err;
+
+ srq->idx_que.entry_shift = ilog2(HNS_ROCE_IDX_QUE_ENTRY_SZ);
+
+ buf_attr.page_shift = hr_dev->caps.idx_buf_pg_sz + HNS_HW_PAGE_SHIFT;
+ buf_attr.region[0].size = to_hr_hem_entries_size(srq->wqe_cnt,
+ srq->idx_que.entry_shift);
+ buf_attr.region[0].hopnum = hr_dev->caps.idx_hop_num;
+ buf_attr.region_count = 1;
+ buf_attr.fixed_page = true;
+
+ err = hns_roce_mtr_create(hr_dev, &idx_que->mtr, &buf_attr,
+ hr_dev->caps.idx_ba_pg_sz + HNS_HW_PAGE_SHIFT,
+ udata, addr);
+ if (err) {
+ ibdev_err(ibdev, "Failed to alloc SRQ idx mtr, err %d\n", err);
+ return err;
}
- buf = &srq->idx_que.idx_buf;
- buf->npages = DIV_ROUND_UP(ib_umem_page_count(srq->idx_que.umem),
- 1 << hr_dev->caps.idx_buf_pg_sz);
- buf->page_shift = PAGE_SHIFT + hr_dev->caps.idx_buf_pg_sz;
- ret = hns_roce_mtt_init(hr_dev, buf->npages, buf->page_shift,
- &srq->idx_que.mtt);
- if (ret) {
- dev_err(hr_dev->dev, "hns_roce_mtt_init error for idx que\n");
- goto err_user_idx_mtt;
- }
+ if (!udata) {
+ idx_que->bitmap = bitmap_zalloc(srq->wqe_cnt, GFP_KERNEL);
+ if (!idx_que->bitmap) {
+ ibdev_err(ibdev, "Failed to alloc SRQ idx bitmap\n");
+ err = -ENOMEM;
+ goto err_idx_mtr;
+ }
- ret = hns_roce_ib_umem_write_mtt(hr_dev, &srq->idx_que.mtt,
- srq->idx_que.umem);
- if (ret) {
- dev_err(hr_dev->dev,
- "hns_roce_ib_umem_write_mtt error for idx que\n");
- goto err_user_idx_buf;
}
return 0;
+err_idx_mtr:
+ hns_roce_mtr_destroy(hr_dev, &idx_que->mtr);
-err_user_idx_buf:
- hns_roce_mtt_cleanup(hr_dev, &srq->idx_que.mtt);
-
-err_user_idx_mtt:
- ib_umem_release(srq->idx_que.umem);
-
-err_user_srq_mtt:
- hns_roce_mtt_cleanup(hr_dev, &srq->mtt);
-
-err_user_buf:
- ib_umem_release(srq->umem);
-
- return ret;
+ return err;
}
-static int hns_roce_create_idx_que(struct ib_pd *pd, struct hns_roce_srq *srq,
- u32 page_shift)
+static void free_srq_idx(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq)
{
- struct hns_roce_dev *hr_dev = to_hr_dev(pd->device);
struct hns_roce_idx_que *idx_que = &srq->idx_que;
- idx_que->bitmap = bitmap_zalloc(srq->wqe_cnt, GFP_KERNEL);
- if (!idx_que->bitmap)
- return -ENOMEM;
-
- idx_que->buf_size = srq->idx_que.buf_size;
-
- if (hns_roce_buf_alloc(hr_dev, idx_que->buf_size, (1 << page_shift) * 2,
- &idx_que->idx_buf, page_shift)) {
- bitmap_free(idx_que->bitmap);
- return -ENOMEM;
- }
-
- return 0;
+ bitmap_free(idx_que->bitmap);
+ idx_que->bitmap = NULL;
+ hns_roce_mtr_destroy(hr_dev, &idx_que->mtr);
}
-static int create_kernel_srq(struct hns_roce_srq *srq, int srq_buf_size)
+static int alloc_srq_wrid(struct hns_roce_dev *hr_dev, struct hns_roce_srq *srq)
{
- struct hns_roce_dev *hr_dev = to_hr_dev(srq->ibsrq.device);
- u32 page_shift = PAGE_SHIFT + hr_dev->caps.srqwqe_buf_pg_sz;
- int ret;
-
- if (hns_roce_buf_alloc(hr_dev, srq_buf_size, (1 << page_shift) * 2,
- &srq->buf, page_shift))
- return -ENOMEM;
-
srq->head = 0;
srq->tail = srq->wqe_cnt - 1;
-
- ret = hns_roce_mtt_init(hr_dev, srq->buf.npages, srq->buf.page_shift,
- &srq->mtt);
- if (ret)
- goto err_kernel_buf;
-
- ret = hns_roce_buf_write_mtt(hr_dev, &srq->mtt, &srq->buf);
- if (ret)
- goto err_kernel_srq_mtt;
-
- page_shift = PAGE_SHIFT + hr_dev->caps.idx_buf_pg_sz;
- ret = hns_roce_create_idx_que(srq->ibsrq.pd, srq, page_shift);
- if (ret) {
- dev_err(hr_dev->dev, "Create idx queue fail(%d)!\n", ret);
- goto err_kernel_srq_mtt;
- }
-
- /* Init mtt table for idx_que */
- ret = hns_roce_mtt_init(hr_dev, srq->idx_que.idx_buf.npages,
- srq->idx_que.idx_buf.page_shift,
- &srq->idx_que.mtt);
- if (ret)
- goto err_kernel_create_idx;
-
- /* Write buffer address into the mtt table */
- ret = hns_roce_buf_write_mtt(hr_dev, &srq->idx_que.mtt,
- &srq->idx_que.idx_buf);
- if (ret)
- goto err_kernel_idx_buf;
-
srq->wrid = kvmalloc_array(srq->wqe_cnt, sizeof(u64), GFP_KERNEL);
- if (!srq->wrid) {
- ret = -ENOMEM;
- goto err_kernel_idx_buf;
- }
+ if (!srq->wrid)
+ return -ENOMEM;
return 0;
-
-err_kernel_idx_buf:
- hns_roce_mtt_cleanup(hr_dev, &srq->idx_que.mtt);
-
-err_kernel_create_idx:
- hns_roce_buf_free(hr_dev, srq->idx_que.buf_size,
- &srq->idx_que.idx_buf);
- kfree(srq->idx_que.bitmap);
-
-err_kernel_srq_mtt:
- hns_roce_mtt_cleanup(hr_dev, &srq->mtt);
-
-err_kernel_buf:
- hns_roce_buf_free(hr_dev, srq_buf_size, &srq->buf);
-
- return ret;
-}
-
-static void destroy_user_srq(struct hns_roce_dev *hr_dev,
- struct hns_roce_srq *srq)
-{
- hns_roce_mtt_cleanup(hr_dev, &srq->idx_que.mtt);
- ib_umem_release(srq->idx_que.umem);
- hns_roce_mtt_cleanup(hr_dev, &srq->mtt);
- ib_umem_release(srq->umem);
}
-static void destroy_kernel_srq(struct hns_roce_dev *hr_dev,
- struct hns_roce_srq *srq, int srq_buf_size)
+static void free_srq_wrid(struct hns_roce_srq *srq)
{
- kvfree(srq->wrid);
- hns_roce_mtt_cleanup(hr_dev, &srq->idx_que.mtt);
- hns_roce_buf_free(hr_dev, srq->idx_que.buf_size, &srq->idx_que.idx_buf);
- kfree(srq->idx_que.bitmap);
- hns_roce_mtt_cleanup(hr_dev, &srq->mtt);
- hns_roce_buf_free(hr_dev, srq_buf_size, &srq->buf);
+ kfree(srq->wrid);
+ srq->wrid = NULL;
}
int hns_roce_create_srq(struct ib_srq *ib_srq,
@@ -365,8 +283,8 @@ int hns_roce_create_srq(struct ib_srq *ib_srq,
struct hns_roce_dev *hr_dev = to_hr_dev(ib_srq->device);
struct hns_roce_ib_create_srq_resp resp = {};
struct hns_roce_srq *srq = to_hr_srq(ib_srq);
- int srq_desc_size;
- int srq_buf_size;
+ struct ib_device *ibdev = &hr_dev->ib_dev;
+ struct hns_roce_ib_create_srq ucmd = {};
int ret = 0;
u32 cqn;
@@ -379,43 +297,47 @@ int hns_roce_create_srq(struct ib_srq *ib_srq,
spin_lock_init(&srq->lock);
srq->wqe_cnt = roundup_pow_of_two(init_attr->attr.max_wr + 1);
- srq->max_gs = init_attr->attr.max_sge;
-
- srq_desc_size = roundup_pow_of_two(max(HNS_ROCE_SGE_SIZE,
- HNS_ROCE_SGE_SIZE * srq->max_gs));
-
- srq->wqe_shift = ilog2(srq_desc_size);
-
- srq_buf_size = srq->wqe_cnt * srq_desc_size;
-
- srq->idx_que.entry_sz = HNS_ROCE_IDX_QUE_ENTRY_SZ;
- srq->idx_que.buf_size = srq->wqe_cnt * srq->idx_que.entry_sz;
- srq->mtt.mtt_type = MTT_TYPE_SRQWQE;
- srq->idx_que.mtt.mtt_type = MTT_TYPE_IDX;
+ srq->max_gs = init_attr->attr.max_sge + HNS_ROCE_RESERVED_SGE;
if (udata) {
- ret = create_user_srq(srq, udata, srq_buf_size);
+ ret = ib_copy_from_udata(&ucmd, udata, sizeof(ucmd));
if (ret) {
- dev_err(hr_dev->dev, "Create user srq failed\n");
- goto err_srq;
+ ibdev_err(ibdev, "Failed to copy SRQ udata, err %d\n",
+ ret);
+ return ret;
}
- } else {
- ret = create_kernel_srq(srq, srq_buf_size);
+ }
+
+ ret = alloc_srq_buf(hr_dev, srq, udata, ucmd.buf_addr);
+ if (ret) {
+ ibdev_err(ibdev, "Failed to alloc SRQ buffer, err %d\n", ret);
+ return ret;
+ }
+
+ ret = alloc_srq_idx(hr_dev, srq, udata, ucmd.que_addr);
+ if (ret) {
+ ibdev_err(ibdev, "Failed to alloc SRQ idx, err %d\n", ret);
+ goto err_buf_alloc;
+ }
+
+ if (!udata) {
+ ret = alloc_srq_wrid(hr_dev, srq);
if (ret) {
- dev_err(hr_dev->dev, "Create kernel srq failed\n");
- goto err_srq;
+ ibdev_err(ibdev, "Failed to alloc SRQ wrid, err %d\n",
+ ret);
+ goto err_idx_alloc;
}
}
cqn = ib_srq_has_cq(init_attr->srq_type) ?
to_hr_cq(init_attr->ext.cq)->cqn : 0;
-
srq->db_reg_l = hr_dev->reg_base + SRQ_DB_REG;
- ret = hns_roce_srq_alloc(hr_dev, to_hr_pd(ib_srq->pd)->pdn, cqn, 0,
- &srq->mtt, 0, srq);
- if (ret)
- goto err_wrid;
+ ret = alloc_srqc(hr_dev, srq, to_hr_pd(ib_srq->pd)->pdn, cqn, 0, 0);
+ if (ret) {
+ ibdev_err(ibdev, "Failed to alloc SRQ context, err %d\n", ret);
+ goto err_wrid_alloc;
+ }
srq->event = hns_roce_ib_srq_event;
resp.srqn = srq->srqn;
@@ -431,15 +353,13 @@ int hns_roce_create_srq(struct ib_srq *ib_srq,
return 0;
err_srqc_alloc:
- hns_roce_srq_free(hr_dev, srq);
-
-err_wrid:
- if (udata)
- destroy_user_srq(hr_dev, srq);
- else
- destroy_kernel_srq(hr_dev, srq, srq_buf_size);
-
-err_srq:
+ free_srqc(hr_dev, srq);
+err_wrid_alloc:
+ free_srq_wrid(srq);
+err_idx_alloc:
+ free_srq_idx(hr_dev, srq);
+err_buf_alloc:
+ free_srq_buf(hr_dev, srq);
return ret;
}
@@ -448,18 +368,10 @@ void hns_roce_destroy_srq(struct ib_srq *ibsrq, struct ib_udata *udata)
struct hns_roce_dev *hr_dev = to_hr_dev(ibsrq->device);
struct hns_roce_srq *srq = to_hr_srq(ibsrq);
- hns_roce_srq_free(hr_dev, srq);
- hns_roce_mtt_cleanup(hr_dev, &srq->mtt);
-
- if (udata) {
- hns_roce_mtt_cleanup(hr_dev, &srq->idx_que.mtt);
- } else {
- kvfree(srq->wrid);
- hns_roce_buf_free(hr_dev, srq->wqe_cnt << srq->wqe_shift,
- &srq->buf);
- }
- ib_umem_release(srq->idx_que.umem);
- ib_umem_release(srq->umem);
+ free_srqc(hr_dev, srq);
+ free_srq_idx(hr_dev, srq);
+ free_srq_wrid(srq);
+ free_srq_buf(hr_dev, srq);
}
int hns_roce_init_srq_table(struct hns_roce_dev *hr_dev)
diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h
index 3c62c9327a9c..49d92638e0db 100644
--- a/drivers/infiniband/hw/i40iw/i40iw.h
+++ b/drivers/infiniband/hw/i40iw/i40iw.h
@@ -382,15 +382,6 @@ static inline struct i40iw_mr *to_iwmr(struct ib_mr *ibmr)
}
/**
- * to_iwmr_from_ibfmr - get device memory region
- * @ibfmr: ib fmr
- **/
-static inline struct i40iw_mr *to_iwmr_from_ibfmr(struct ib_fmr *ibfmr)
-{
- return container_of(ibfmr, struct i40iw_mr, ibfmr);
-}
-
-/**
* to_iwmw - get device memory window
* @ibmw: ib memory window
**/
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index 1b6fb1380961..19af29a48c55 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -83,7 +83,6 @@ static int i40iw_query_device(struct ib_device *ibdev,
props->max_qp_rd_atom = I40IW_MAX_IRD_SIZE;
props->max_qp_init_rd_atom = props->max_qp_rd_atom;
props->atomic_cap = IB_ATOMIC_NONE;
- props->max_map_per_fmr = 1;
props->max_fast_reg_page_list_len = I40IW_MAX_PAGES_PER_FMR;
return 0;
}
diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h
index 3a413752ccc3..331bc21cbcc7 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h
@@ -89,7 +89,6 @@ struct i40iw_mr {
union {
struct ib_mr ibmr;
struct ib_mw ibmw;
- struct ib_fmr ibfmr;
};
struct ib_umem *region;
u16 type;
diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c
index 02a169f8027b..5f8f8d5c0ce0 100644
--- a/drivers/infiniband/hw/mlx4/ah.c
+++ b/drivers/infiniband/hw/mlx4/ah.c
@@ -141,10 +141,11 @@ static int create_iboe_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr)
return 0;
}
-int mlx4_ib_create_ah(struct ib_ah *ib_ah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
-
+int mlx4_ib_create_ah(struct ib_ah *ib_ah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
+
if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
if (!(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH))
return -EINVAL;
@@ -167,12 +168,14 @@ int mlx4_ib_create_ah_slave(struct ib_ah *ah, struct rdma_ah_attr *ah_attr,
int slave_sgid_index, u8 *s_mac, u16 vlan_tag)
{
struct rdma_ah_attr slave_attr = *ah_attr;
+ struct rdma_ah_init_attr init_attr = {};
struct mlx4_ib_ah *mah = to_mah(ah);
int ret;
slave_attr.grh.sgid_attr = NULL;
slave_attr.grh.sgid_index = slave_sgid_index;
- ret = mlx4_ib_create_ah(ah, &slave_attr, 0, NULL);
+ init_attr.ah_attr = &slave_attr;
+ ret = mlx4_ib_create_ah(ah, &init_attr, NULL);
if (ret)
return ret;
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 275722cec8c6..816d28854a8e 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -558,7 +558,6 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
props->max_mcast_qp_attach = dev->dev->caps.num_qp_per_mgm;
props->max_total_mcast_qp_attach = props->max_mcast_qp_attach *
props->max_mcast_grp;
- props->max_map_per_fmr = dev->dev->caps.max_fmr_maps;
props->hca_core_clock = dev->dev->caps.hca_core_clock * 1000UL;
props->timestamp_mask = 0xFFFFFFFFFFFFULL;
props->max_ah = INT_MAX;
@@ -2600,13 +2599,6 @@ static const struct ib_device_ops mlx4_ib_dev_wq_ops = {
.modify_wq = mlx4_ib_modify_wq,
};
-static const struct ib_device_ops mlx4_ib_dev_fmr_ops = {
- .alloc_fmr = mlx4_ib_fmr_alloc,
- .dealloc_fmr = mlx4_ib_fmr_dealloc,
- .map_phys_fmr = mlx4_ib_map_phys_fmr,
- .unmap_fmr = mlx4_ib_unmap_fmr,
-};
-
static const struct ib_device_ops mlx4_ib_dev_mw_ops = {
.alloc_mw = mlx4_ib_alloc_mw,
.dealloc_mw = mlx4_ib_dealloc_mw,
@@ -2724,9 +2716,6 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
ib_set_device_ops(&ibdev->ib_dev, &mlx4_ib_dev_wq_ops);
}
- if (!mlx4_is_slave(ibdev->dev))
- ib_set_device_ops(&ibdev->ib_dev, &mlx4_ib_dev_fmr_ops);
-
if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN) {
ibdev->ib_dev.uverbs_cmd_mask |=
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index d188573187fa..6f4ea1067095 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -146,11 +146,6 @@ struct mlx4_ib_mw {
struct mlx4_mw mmw;
};
-struct mlx4_ib_fmr {
- struct ib_fmr ibfmr;
- struct mlx4_fmr mfmr;
-};
-
#define MAX_REGS_PER_FLOW 2
struct mlx4_flow_reg_id {
@@ -679,11 +674,6 @@ static inline struct mlx4_ib_mw *to_mmw(struct ib_mw *ibmw)
return container_of(ibmw, struct mlx4_ib_mw, ibmw);
}
-static inline struct mlx4_ib_fmr *to_mfmr(struct ib_fmr *ibfmr)
-{
- return container_of(ibfmr, struct mlx4_ib_fmr, ibfmr);
-}
-
static inline struct mlx4_ib_flow *to_mflow(struct ib_flow *ibflow)
{
return container_of(ibflow, struct mlx4_ib_flow, ibflow);
@@ -752,7 +742,7 @@ int mlx4_ib_arm_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags);
void __mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq);
void mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq);
-int mlx4_ib_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr, u32 flags,
+int mlx4_ib_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
int mlx4_ib_create_ah_slave(struct ib_ah *ah, struct rdma_ah_attr *ah_attr,
int slave_sgid_index, u8 *s_mac, u16 vlan_tag);
@@ -794,12 +784,6 @@ int mlx4_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
int mlx4_ib_mad_init(struct mlx4_ib_dev *dev);
void mlx4_ib_mad_cleanup(struct mlx4_ib_dev *dev);
-struct ib_fmr *mlx4_ib_fmr_alloc(struct ib_pd *pd, int mr_access_flags,
- struct ib_fmr_attr *fmr_attr);
-int mlx4_ib_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list, int npages,
- u64 iova);
-int mlx4_ib_unmap_fmr(struct list_head *fmr_list);
-int mlx4_ib_fmr_dealloc(struct ib_fmr *fmr);
int __mlx4_ib_query_port(struct ib_device *ibdev, u8 port,
struct ib_port_attr *props, int netw_view);
int __mlx4_ib_query_pkey(struct ib_device *ibdev, u8 port, u16 index,
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index b0121c90c561..e2fb71b23c80 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -698,99 +698,6 @@ err_free:
return ERR_PTR(err);
}
-struct ib_fmr *mlx4_ib_fmr_alloc(struct ib_pd *pd, int acc,
- struct ib_fmr_attr *fmr_attr)
-{
- struct mlx4_ib_dev *dev = to_mdev(pd->device);
- struct mlx4_ib_fmr *fmr;
- int err = -ENOMEM;
-
- fmr = kmalloc(sizeof *fmr, GFP_KERNEL);
- if (!fmr)
- return ERR_PTR(-ENOMEM);
-
- err = mlx4_fmr_alloc(dev->dev, to_mpd(pd)->pdn, convert_access(acc),
- fmr_attr->max_pages, fmr_attr->max_maps,
- fmr_attr->page_shift, &fmr->mfmr);
- if (err)
- goto err_free;
-
- err = mlx4_fmr_enable(to_mdev(pd->device)->dev, &fmr->mfmr);
- if (err)
- goto err_mr;
-
- fmr->ibfmr.rkey = fmr->ibfmr.lkey = fmr->mfmr.mr.key;
-
- return &fmr->ibfmr;
-
-err_mr:
- (void) mlx4_mr_free(to_mdev(pd->device)->dev, &fmr->mfmr.mr);
-
-err_free:
- kfree(fmr);
-
- return ERR_PTR(err);
-}
-
-int mlx4_ib_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
- int npages, u64 iova)
-{
- struct mlx4_ib_fmr *ifmr = to_mfmr(ibfmr);
- struct mlx4_ib_dev *dev = to_mdev(ifmr->ibfmr.device);
-
- return mlx4_map_phys_fmr(dev->dev, &ifmr->mfmr, page_list, npages, iova,
- &ifmr->ibfmr.lkey, &ifmr->ibfmr.rkey);
-}
-
-int mlx4_ib_unmap_fmr(struct list_head *fmr_list)
-{
- struct ib_fmr *ibfmr;
- int err;
- struct mlx4_dev *mdev = NULL;
-
- list_for_each_entry(ibfmr, fmr_list, list) {
- if (mdev && to_mdev(ibfmr->device)->dev != mdev)
- return -EINVAL;
- mdev = to_mdev(ibfmr->device)->dev;
- }
-
- if (!mdev)
- return 0;
-
- list_for_each_entry(ibfmr, fmr_list, list) {
- struct mlx4_ib_fmr *ifmr = to_mfmr(ibfmr);
-
- mlx4_fmr_unmap(mdev, &ifmr->mfmr, &ifmr->ibfmr.lkey, &ifmr->ibfmr.rkey);
- }
-
- /*
- * Make sure all MPT status updates are visible before issuing
- * SYNC_TPT firmware command.
- */
- wmb();
-
- err = mlx4_SYNC_TPT(mdev);
- if (err)
- pr_warn("SYNC_TPT error %d when "
- "unmapping FMRs\n", err);
-
- return 0;
-}
-
-int mlx4_ib_fmr_dealloc(struct ib_fmr *ibfmr)
-{
- struct mlx4_ib_fmr *ifmr = to_mfmr(ibfmr);
- struct mlx4_ib_dev *dev = to_mdev(ibfmr->device);
- int err;
-
- err = mlx4_fmr_free(dev->dev, &ifmr->mfmr);
-
- if (!err)
- kfree(ifmr);
-
- return err;
-}
-
static int mlx4_set_page(struct ib_mr *ibmr, u64 addr)
{
struct mlx4_ib_mr *mr = to_mmr(ibmr);
diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index 228be05fbaf8..8cca61c671f8 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -16,7 +16,8 @@ mlx5_ib-y := ah.o \
qpc.o \
restrack.o \
srq.o \
- srq_cmd.o
+ srq_cmd.o \
+ wr.o
mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o
mlx5_ib-$(CONFIG_MLX5_ESWITCH) += ib_rep.o
diff --git a/drivers/infiniband/hw/mlx5/ah.c b/drivers/infiniband/hw/mlx5/ah.c
index 80642dd359bc..59e5ec39b447 100644
--- a/drivers/infiniband/hw/mlx5/ah.c
+++ b/drivers/infiniband/hw/mlx5/ah.c
@@ -32,9 +32,28 @@
#include "mlx5_ib.h"
+static __be16 mlx5_ah_get_udp_sport(const struct mlx5_ib_dev *dev,
+ const struct rdma_ah_attr *ah_attr)
+{
+ enum ib_gid_type gid_type = ah_attr->grh.sgid_attr->gid_type;
+ __be16 sport;
+
+ if ((gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) &&
+ (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) &&
+ (ah_attr->grh.flow_label & IB_GRH_FLOWLABEL_MASK))
+ sport = cpu_to_be16(
+ rdma_flow_label_to_udp_sport(ah_attr->grh.flow_label));
+ else
+ sport = mlx5_get_roce_udp_sport_min(dev,
+ ah_attr->grh.sgid_attr);
+
+ return sport;
+}
+
static void create_ib_ah(struct mlx5_ib_dev *dev, struct mlx5_ib_ah *ah,
- struct rdma_ah_attr *ah_attr)
+ struct rdma_ah_init_attr *init_attr)
{
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
enum ib_gid_type gid_type;
if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) {
@@ -51,12 +70,15 @@ static void create_ib_ah(struct mlx5_ib_dev *dev, struct mlx5_ib_ah *ah,
ah->av.stat_rate_sl = (rdma_ah_get_static_rate(ah_attr) << 4);
if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
+ if (init_attr->xmit_slave)
+ ah->xmit_port =
+ mlx5_lag_get_slave_port(dev->mdev,
+ init_attr->xmit_slave);
gid_type = ah_attr->grh.sgid_attr->gid_type;
memcpy(ah->av.rmac, ah_attr->roce.dmac,
sizeof(ah_attr->roce.dmac));
- ah->av.udp_sport =
- mlx5_get_roce_udp_sport(dev, ah_attr->grh.sgid_attr);
+ ah->av.udp_sport = mlx5_ah_get_udp_sport(dev, ah_attr);
ah->av.stat_rate_sl |= (rdma_ah_get_sl(ah_attr) & 0x7) << 1;
if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
#define MLX5_ECN_ENABLED BIT(1)
@@ -68,10 +90,11 @@ static void create_ib_ah(struct mlx5_ib_dev *dev, struct mlx5_ib_ah *ah,
}
}
-int mlx5_ib_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
+int mlx5_ib_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
struct mlx5_ib_ah *ah = to_mah(ibah);
struct mlx5_ib_dev *dev = to_mdev(ibah->device);
enum rdma_ah_attr_type ah_type = ah_attr->type;
@@ -97,7 +120,7 @@ int mlx5_ib_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr,
return err;
}
- create_ib_ah(dev, ah, ah_attr);
+ create_ib_ah(dev, ah, init_attr);
return 0;
}
diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c
index a2fcbc49131e..cc24c711e92a 100644
--- a/drivers/infiniband/hw/mlx5/cmd.c
+++ b/drivers/infiniband/hw/mlx5/cmd.c
@@ -1,46 +1,19 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
/*
- * Copyright (c) 2017, Mellanox Technologies. All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses. You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- * Redistribution and use in source and binary forms, with or
- * without modification, are permitted provided that the following
- * conditions are met:
- *
- * - Redistributions of source code must retain the above
- * copyright notice, this list of conditions and the following
- * disclaimer.
- *
- * - Redistributions in binary form must reproduce the above
- * copyright notice, this list of conditions and the following
- * disclaimer in the documentation and/or other materials
- * provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
+ * Copyright (c) 2017-2020, Mellanox Technologies inc. All rights reserved.
*/
#include "cmd.h"
int mlx5_cmd_dump_fill_mkey(struct mlx5_core_dev *dev, u32 *mkey)
{
- u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {0};
- u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {};
int err;
MLX5_SET(query_special_contexts_in, in, opcode,
MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS);
- err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_inout(dev, query_special_contexts, in, out);
if (!err)
*mkey = MLX5_GET(query_special_contexts_out, out,
dump_fill_mkey);
@@ -50,12 +23,12 @@ int mlx5_cmd_dump_fill_mkey(struct mlx5_core_dev *dev, u32 *mkey)
int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey)
{
u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {};
- u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {};
int err;
MLX5_SET(query_special_contexts_in, in, opcode,
MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS);
- err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_inout(dev, query_special_contexts, in, out);
if (!err)
*null_mkey = MLX5_GET(query_special_contexts_out, out,
null_mkey);
@@ -63,23 +36,15 @@ int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey)
}
int mlx5_cmd_query_cong_params(struct mlx5_core_dev *dev, int cong_point,
- void *out, int out_size)
+ void *out)
{
- u32 in[MLX5_ST_SZ_DW(query_cong_params_in)] = { };
+ u32 in[MLX5_ST_SZ_DW(query_cong_params_in)] = {};
MLX5_SET(query_cong_params_in, in, opcode,
MLX5_CMD_OP_QUERY_CONG_PARAMS);
MLX5_SET(query_cong_params_in, in, cong_protocol, cong_point);
- return mlx5_cmd_exec(dev, in, sizeof(in), out, out_size);
-}
-
-int mlx5_cmd_modify_cong_params(struct mlx5_core_dev *dev,
- void *in, int in_size)
-{
- u32 out[MLX5_ST_SZ_DW(modify_cong_params_out)] = { };
-
- return mlx5_cmd_exec(dev, in, in_size, out, sizeof(out));
+ return mlx5_cmd_exec_inout(dev, query_cong_params, in, out);
}
int mlx5_cmd_alloc_memic(struct mlx5_dm *dm, phys_addr_t *addr,
@@ -133,7 +98,7 @@ int mlx5_cmd_alloc_memic(struct mlx5_dm *dm, phys_addr_t *addr,
MLX5_SET64(alloc_memic_in, in, range_start_addr,
hw_start_addr + (page_idx * PAGE_SIZE));
- ret = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ ret = mlx5_cmd_exec_inout(dev, alloc_memic, in, out);
if (ret) {
spin_lock(&dm->lock);
bitmap_clear(dm->memic_alloc_pages,
@@ -162,8 +127,7 @@ void mlx5_cmd_dealloc_memic(struct mlx5_dm *dm, phys_addr_t addr, u64 length)
struct mlx5_core_dev *dev = dm->dev;
u64 hw_start_addr = MLX5_CAP64_DEV_MEM(dev, memic_bar_start_addr);
u32 num_pages = DIV_ROUND_UP(length, PAGE_SIZE);
- u32 out[MLX5_ST_SZ_DW(dealloc_memic_out)] = {0};
- u32 in[MLX5_ST_SZ_DW(dealloc_memic_in)] = {0};
+ u32 in[MLX5_ST_SZ_DW(dealloc_memic_in)] = {};
u64 start_page_idx;
int err;
@@ -174,7 +138,7 @@ void mlx5_cmd_dealloc_memic(struct mlx5_dm *dm, phys_addr_t addr, u64 length)
MLX5_SET64(dealloc_memic_in, in, memic_start_addr, addr);
MLX5_SET(dealloc_memic_in, in, memic_size, length);
- err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_in(dev, dealloc_memic, in);
if (err)
return;
@@ -198,49 +162,46 @@ int mlx5_cmd_query_ext_ppcnt_counters(struct mlx5_core_dev *dev, void *out)
void mlx5_cmd_destroy_tir(struct mlx5_core_dev *dev, u32 tirn, u16 uid)
{
- u32 in[MLX5_ST_SZ_DW(destroy_tir_in)] = {};
- u32 out[MLX5_ST_SZ_DW(destroy_tir_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(destroy_tir_in)] = {};
MLX5_SET(destroy_tir_in, in, opcode, MLX5_CMD_OP_DESTROY_TIR);
MLX5_SET(destroy_tir_in, in, tirn, tirn);
MLX5_SET(destroy_tir_in, in, uid, uid);
- mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ mlx5_cmd_exec_in(dev, destroy_tir, in);
}
void mlx5_cmd_destroy_tis(struct mlx5_core_dev *dev, u32 tisn, u16 uid)
{
- u32 in[MLX5_ST_SZ_DW(destroy_tis_in)] = {0};
- u32 out[MLX5_ST_SZ_DW(destroy_tis_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(destroy_tis_in)] = {};
MLX5_SET(destroy_tis_in, in, opcode, MLX5_CMD_OP_DESTROY_TIS);
MLX5_SET(destroy_tis_in, in, tisn, tisn);
MLX5_SET(destroy_tis_in, in, uid, uid);
- mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ mlx5_cmd_exec_in(dev, destroy_tis, in);
}
void mlx5_cmd_destroy_rqt(struct mlx5_core_dev *dev, u32 rqtn, u16 uid)
{
- u32 in[MLX5_ST_SZ_DW(destroy_rqt_in)] = {};
- u32 out[MLX5_ST_SZ_DW(destroy_rqt_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(destroy_rqt_in)] = {};
MLX5_SET(destroy_rqt_in, in, opcode, MLX5_CMD_OP_DESTROY_RQT);
MLX5_SET(destroy_rqt_in, in, rqtn, rqtn);
MLX5_SET(destroy_rqt_in, in, uid, uid);
- mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ mlx5_cmd_exec_in(dev, destroy_rqt, in);
}
int mlx5_cmd_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn,
u16 uid)
{
- u32 in[MLX5_ST_SZ_DW(alloc_transport_domain_in)] = {0};
- u32 out[MLX5_ST_SZ_DW(alloc_transport_domain_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(alloc_transport_domain_in)] = {};
+ u32 out[MLX5_ST_SZ_DW(alloc_transport_domain_out)] = {};
int err;
MLX5_SET(alloc_transport_domain_in, in, opcode,
MLX5_CMD_OP_ALLOC_TRANSPORT_DOMAIN);
MLX5_SET(alloc_transport_domain_in, in, uid, uid);
- err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_inout(dev, alloc_transport_domain, in, out);
if (!err)
*tdn = MLX5_GET(alloc_transport_domain_out, out,
transport_domain);
@@ -251,32 +212,29 @@ int mlx5_cmd_alloc_transport_domain(struct mlx5_core_dev *dev, u32 *tdn,
void mlx5_cmd_dealloc_transport_domain(struct mlx5_core_dev *dev, u32 tdn,
u16 uid)
{
- u32 in[MLX5_ST_SZ_DW(dealloc_transport_domain_in)] = {0};
- u32 out[MLX5_ST_SZ_DW(dealloc_transport_domain_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(dealloc_transport_domain_in)] = {};
MLX5_SET(dealloc_transport_domain_in, in, opcode,
MLX5_CMD_OP_DEALLOC_TRANSPORT_DOMAIN);
MLX5_SET(dealloc_transport_domain_in, in, uid, uid);
MLX5_SET(dealloc_transport_domain_in, in, transport_domain, tdn);
- mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ mlx5_cmd_exec_in(dev, dealloc_transport_domain, in);
}
void mlx5_cmd_dealloc_pd(struct mlx5_core_dev *dev, u32 pdn, u16 uid)
{
- u32 out[MLX5_ST_SZ_DW(dealloc_pd_out)] = {};
- u32 in[MLX5_ST_SZ_DW(dealloc_pd_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(dealloc_pd_in)] = {};
MLX5_SET(dealloc_pd_in, in, opcode, MLX5_CMD_OP_DEALLOC_PD);
MLX5_SET(dealloc_pd_in, in, pd, pdn);
MLX5_SET(dealloc_pd_in, in, uid, uid);
- mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ mlx5_cmd_exec_in(dev, dealloc_pd, in);
}
int mlx5_cmd_attach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid,
u32 qpn, u16 uid)
{
- u32 out[MLX5_ST_SZ_DW(attach_to_mcg_out)] = {};
- u32 in[MLX5_ST_SZ_DW(attach_to_mcg_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(attach_to_mcg_in)] = {};
void *gid;
MLX5_SET(attach_to_mcg_in, in, opcode, MLX5_CMD_OP_ATTACH_TO_MCG);
@@ -284,14 +242,13 @@ int mlx5_cmd_attach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid,
MLX5_SET(attach_to_mcg_in, in, uid, uid);
gid = MLX5_ADDR_OF(attach_to_mcg_in, in, multicast_gid);
memcpy(gid, mgid, sizeof(*mgid));
- return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev, attach_to_mcg, in);
}
int mlx5_cmd_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid,
u32 qpn, u16 uid)
{
- u32 out[MLX5_ST_SZ_DW(detach_from_mcg_out)] = {};
- u32 in[MLX5_ST_SZ_DW(detach_from_mcg_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(detach_from_mcg_in)] = {};
void *gid;
MLX5_SET(detach_from_mcg_in, in, opcode, MLX5_CMD_OP_DETACH_FROM_MCG);
@@ -299,18 +256,18 @@ int mlx5_cmd_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid,
MLX5_SET(detach_from_mcg_in, in, uid, uid);
gid = MLX5_ADDR_OF(detach_from_mcg_in, in, multicast_gid);
memcpy(gid, mgid, sizeof(*mgid));
- return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev, detach_from_mcg, in);
}
int mlx5_cmd_xrcd_alloc(struct mlx5_core_dev *dev, u32 *xrcdn, u16 uid)
{
u32 out[MLX5_ST_SZ_DW(alloc_xrcd_out)] = {};
- u32 in[MLX5_ST_SZ_DW(alloc_xrcd_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(alloc_xrcd_in)] = {};
int err;
MLX5_SET(alloc_xrcd_in, in, opcode, MLX5_CMD_OP_ALLOC_XRCD);
MLX5_SET(alloc_xrcd_in, in, uid, uid);
- err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_inout(dev, alloc_xrcd, in, out);
if (!err)
*xrcdn = MLX5_GET(alloc_xrcd_out, out, xrcd);
return err;
@@ -318,13 +275,12 @@ int mlx5_cmd_xrcd_alloc(struct mlx5_core_dev *dev, u32 *xrcdn, u16 uid)
int mlx5_cmd_xrcd_dealloc(struct mlx5_core_dev *dev, u32 xrcdn, u16 uid)
{
- u32 out[MLX5_ST_SZ_DW(dealloc_xrcd_out)] = {};
- u32 in[MLX5_ST_SZ_DW(dealloc_xrcd_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(dealloc_xrcd_in)] = {};
MLX5_SET(dealloc_xrcd_in, in, opcode, MLX5_CMD_OP_DEALLOC_XRCD);
MLX5_SET(dealloc_xrcd_in, in, xrcd, xrcdn);
MLX5_SET(dealloc_xrcd_in, in, uid, uid);
- return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev, dealloc_xrcd, in);
}
int mlx5_cmd_mad_ifc(struct mlx5_core_dev *dev, const void *inb, void *outb,
@@ -350,7 +306,7 @@ int mlx5_cmd_mad_ifc(struct mlx5_core_dev *dev, const void *inb, void *outb,
data = MLX5_ADDR_OF(mad_ifc_in, in, mad);
memcpy(data, inb, MLX5_FLD_SZ_BYTES(mad_ifc_in, mad));
- err = mlx5_cmd_exec(dev, in, inlen, out, outlen);
+ err = mlx5_cmd_exec_inout(dev, mad_ifc, in, out);
if (err)
goto out;
diff --git a/drivers/infiniband/hw/mlx5/cmd.h b/drivers/infiniband/hw/mlx5/cmd.h
index 43079b18d9b4..f4d8558db434 100644
--- a/drivers/infiniband/hw/mlx5/cmd.h
+++ b/drivers/infiniband/hw/mlx5/cmd.h
@@ -40,10 +40,8 @@
int mlx5_cmd_dump_fill_mkey(struct mlx5_core_dev *dev, u32 *mkey);
int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey);
int mlx5_cmd_query_cong_params(struct mlx5_core_dev *dev, int cong_point,
- void *out, int out_size);
+ void *out);
int mlx5_cmd_query_ext_ppcnt_counters(struct mlx5_core_dev *dev, void *out);
-int mlx5_cmd_modify_cong_params(struct mlx5_core_dev *mdev,
- void *in, int in_size);
int mlx5_cmd_alloc_memic(struct mlx5_dm *dm, phys_addr_t *addr,
u64 length, u32 alignment);
void mlx5_cmd_dealloc_memic(struct mlx5_dm *dm, phys_addr_t addr, u64 length);
diff --git a/drivers/infiniband/hw/mlx5/cong.c b/drivers/infiniband/hw/mlx5/cong.c
index de4da92b81a6..b9291e482428 100644
--- a/drivers/infiniband/hw/mlx5/cong.c
+++ b/drivers/infiniband/hw/mlx5/cong.c
@@ -290,7 +290,7 @@ static int mlx5_ib_get_cc_params(struct mlx5_ib_dev *dev, u8 port_num,
node = mlx5_ib_param_to_node(offset);
- err = mlx5_cmd_query_cong_params(mdev, node, out, outlen);
+ err = mlx5_cmd_query_cong_params(mdev, node, out);
if (err)
goto free;
@@ -339,7 +339,7 @@ static int mlx5_ib_set_cc_params(struct mlx5_ib_dev *dev, u8 port_num,
MLX5_SET(field_select_r_roce_rp, field, field_select_r_roce_rp,
attr_mask);
- err = mlx5_cmd_modify_cong_params(mdev, in, inlen);
+ err = mlx5_cmd_exec_in(dev->mdev, modify_cong_params, in);
kvfree(in);
alloc_err:
mlx5_ib_put_native_port_mdev(dev, port_num + 1);
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index 35b98c2d64d5..9454a66c12cc 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -495,6 +495,10 @@ static u64 devx_get_obj_id(const void *in)
obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
MLX5_GET(rst2init_qp_in, in, qpn));
break;
+ case MLX5_CMD_OP_INIT2INIT_QP:
+ obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
+ MLX5_GET(init2init_qp_in, in, qpn));
+ break;
case MLX5_CMD_OP_INIT2RTR_QP:
obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
MLX5_GET(init2rtr_qp_in, in, qpn));
@@ -615,7 +619,7 @@ static bool devx_is_valid_obj_id(struct uverbs_attr_bundle *attrs,
enum ib_qp_type qp_type = qp->ibqp.qp_type;
if (qp_type == IB_QPT_RAW_PACKET ||
- (qp->flags & MLX5_IB_QP_UNDERLAY)) {
+ (qp->flags & IB_QP_CREATE_SOURCE_QPN)) {
struct mlx5_ib_raw_packet_qp *raw_packet_qp =
&qp->raw_packet_qp;
struct mlx5_ib_rq *rq = &raw_packet_qp->rq;
@@ -820,6 +824,7 @@ static bool devx_is_obj_modify_cmd(const void *in)
case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
case MLX5_CMD_OP_RST2INIT_QP:
case MLX5_CMD_OP_INIT2RTR_QP:
+ case MLX5_CMD_OP_INIT2INIT_QP:
case MLX5_CMD_OP_RTR2RTS_QP:
case MLX5_CMD_OP_RTS2RTS_QP:
case MLX5_CMD_OP_SQERR2RTS_QP:
@@ -2217,14 +2222,12 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_UMEM_REG)(
obj->mdev = dev->mdev;
uobj->object = obj;
devx_obj_build_destroy_cmd(cmd.in, cmd.out, obj->dinbox, &obj->dinlen, &obj_id);
- err = uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_OUT_ID, &obj_id, sizeof(obj_id));
- if (err)
- goto err_umem_destroy;
+ uverbs_finalize_uobj_create(attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_HANDLE);
- return 0;
+ err = uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_OUT_ID, &obj_id,
+ sizeof(obj_id));
+ return err;
-err_umem_destroy:
- mlx5_cmd_exec(obj->mdev, obj->dinbox, obj->dinlen, cmd.out, sizeof(cmd.out));
err_umem_release:
ib_umem_release(obj->umem);
err_obj_free:
diff --git a/drivers/infiniband/hw/mlx5/flow.c b/drivers/infiniband/hw/mlx5/flow.c
index 3a0601c2052c..216a1108ad34 100644
--- a/drivers/infiniband/hw/mlx5/flow.c
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -67,46 +67,41 @@ static const struct uverbs_attr_spec mlx5_ib_flow_type[] = {
},
};
-#define MLX5_IB_CREATE_FLOW_MAX_FLOW_ACTIONS 2
-static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
- struct uverbs_attr_bundle *attrs)
+static int get_dests(struct uverbs_attr_bundle *attrs,
+ struct mlx5_ib_flow_matcher *fs_matcher, int *dest_id,
+ int *dest_type, struct ib_qp **qp, u32 *flags)
{
- struct mlx5_flow_context flow_context = {.flow_tag = MLX5_FS_DEFAULT_FLOW_TAG};
- struct mlx5_ib_flow_handler *flow_handler;
- struct mlx5_ib_flow_matcher *fs_matcher;
- struct ib_uobject **arr_flow_actions;
- struct ib_uflow_resources *uflow_res;
- struct mlx5_flow_act flow_act = {};
- void *devx_obj;
- int dest_id, dest_type;
- void *cmd_in;
- int inlen;
bool dest_devx, dest_qp;
- struct ib_qp *qp = NULL;
- struct ib_uobject *uobj =
- uverbs_attr_get_uobject(attrs, MLX5_IB_ATTR_CREATE_FLOW_HANDLE);
- struct mlx5_ib_dev *dev = mlx5_udata_to_mdev(&attrs->driver_udata);
- int len, ret, i;
- u32 counter_id = 0;
- u32 *offset_attr;
- u32 offset = 0;
-
- if (!capable(CAP_NET_RAW))
- return -EPERM;
+ void *devx_obj;
+ int err;
- dest_devx =
- uverbs_attr_is_valid(attrs, MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
+ dest_devx = uverbs_attr_is_valid(attrs,
+ MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
dest_qp = uverbs_attr_is_valid(attrs,
MLX5_IB_ATTR_CREATE_FLOW_DEST_QP);
- fs_matcher = uverbs_attr_get_obj(attrs,
- MLX5_IB_ATTR_CREATE_FLOW_MATCHER);
- if (fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_BYPASS &&
- ((dest_devx && dest_qp) || (!dest_devx && !dest_qp)))
+ *flags = 0;
+ err = uverbs_get_flags32(flags, attrs, MLX5_IB_ATTR_CREATE_FLOW_FLAGS,
+ MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DEFAULT_MISS |
+ MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DROP);
+ if (err)
+ return err;
+
+ /* Both flags are not allowed */
+ if (*flags & MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DEFAULT_MISS &&
+ *flags & MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DROP)
return -EINVAL;
- /* Allow only DEVX object as dest when inserting to FDB */
- if (fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_FDB && !dest_devx)
+ if (fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_BYPASS) {
+ if (dest_devx && (dest_qp || *flags))
+ return -EINVAL;
+ else if (dest_qp && *flags)
+ return -EINVAL;
+ }
+
+ /* Allow only DEVX object, drop as dest for FDB */
+ if (fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_FDB && !(dest_devx ||
+ (*flags & MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DROP)))
return -EINVAL;
/* Allow only DEVX object or QP as dest when inserting to RDMA_RX */
@@ -114,43 +109,86 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
((!dest_devx && !dest_qp) || (dest_devx && dest_qp)))
return -EINVAL;
+ *qp = NULL;
if (dest_devx) {
- devx_obj = uverbs_attr_get_obj(
- attrs, MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
- if (IS_ERR(devx_obj))
- return PTR_ERR(devx_obj);
+ devx_obj =
+ uverbs_attr_get_obj(attrs,
+ MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
/* Verify that the given DEVX object is a flow
* steering destination.
*/
- if (!mlx5_ib_devx_is_flow_dest(devx_obj, &dest_id, &dest_type))
+ if (!mlx5_ib_devx_is_flow_dest(devx_obj, dest_id, dest_type))
return -EINVAL;
/* Allow only flow table as dest when inserting to FDB or RDMA_RX */
if ((fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_FDB ||
fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_RDMA_RX) &&
- dest_type != MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE)
+ *dest_type != MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE)
return -EINVAL;
} else if (dest_qp) {
struct mlx5_ib_qp *mqp;
- qp = uverbs_attr_get_obj(attrs,
- MLX5_IB_ATTR_CREATE_FLOW_DEST_QP);
- if (IS_ERR(qp))
- return PTR_ERR(qp);
+ *qp = uverbs_attr_get_obj(attrs,
+ MLX5_IB_ATTR_CREATE_FLOW_DEST_QP);
+ if (IS_ERR(*qp))
+ return PTR_ERR(*qp);
- if (qp->qp_type != IB_QPT_RAW_PACKET)
+ if ((*qp)->qp_type != IB_QPT_RAW_PACKET)
return -EINVAL;
- mqp = to_mqp(qp);
- if (mqp->flags & MLX5_IB_QP_RSS)
- dest_id = mqp->rss_qp.tirn;
+ mqp = to_mqp(*qp);
+ if (mqp->is_rss)
+ *dest_id = mqp->rss_qp.tirn;
else
- dest_id = mqp->raw_packet_qp.rq.tirn;
- dest_type = MLX5_FLOW_DESTINATION_TYPE_TIR;
- } else {
- dest_type = MLX5_FLOW_DESTINATION_TYPE_PORT;
+ *dest_id = mqp->raw_packet_qp.rq.tirn;
+ *dest_type = MLX5_FLOW_DESTINATION_TYPE_TIR;
+ } else if (fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_EGRESS) {
+ *dest_type = MLX5_FLOW_DESTINATION_TYPE_PORT;
}
+ if (*dest_type == MLX5_FLOW_DESTINATION_TYPE_TIR &&
+ fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_EGRESS)
+ return -EINVAL;
+
+ return 0;
+}
+
+#define MLX5_IB_CREATE_FLOW_MAX_FLOW_ACTIONS 2
+static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
+ struct uverbs_attr_bundle *attrs)
+{
+ struct mlx5_flow_context flow_context = {.flow_tag =
+ MLX5_FS_DEFAULT_FLOW_TAG};
+ u32 *offset_attr, offset = 0, counter_id = 0;
+ int dest_id, dest_type, inlen, len, ret, i;
+ struct mlx5_ib_flow_handler *flow_handler;
+ struct mlx5_ib_flow_matcher *fs_matcher;
+ struct ib_uobject **arr_flow_actions;
+ struct ib_uflow_resources *uflow_res;
+ struct mlx5_flow_act flow_act = {};
+ struct ib_qp *qp = NULL;
+ void *devx_obj, *cmd_in;
+ struct ib_uobject *uobj;
+ struct mlx5_ib_dev *dev;
+ u32 flags;
+
+ if (!capable(CAP_NET_RAW))
+ return -EPERM;
+
+ fs_matcher = uverbs_attr_get_obj(attrs,
+ MLX5_IB_ATTR_CREATE_FLOW_MATCHER);
+ uobj = uverbs_attr_get_uobject(attrs, MLX5_IB_ATTR_CREATE_FLOW_HANDLE);
+ dev = mlx5_udata_to_mdev(&attrs->driver_udata);
+
+ if (get_dests(attrs, fs_matcher, &dest_id, &dest_type, &qp, &flags))
+ return -EINVAL;
+
+ if (flags & MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DEFAULT_MISS)
+ flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_NS;
+
+ if (flags & MLX5_IB_ATTR_CREATE_FLOW_FLAGS_DROP)
+ flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
+
len = uverbs_attr_get_uobjs_arr(attrs,
MLX5_IB_ATTR_CREATE_FLOW_ARR_COUNTERS_DEVX, &arr_flow_actions);
if (len) {
@@ -180,10 +218,6 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT;
}
- if (dest_type == MLX5_FLOW_DESTINATION_TYPE_TIR &&
- fs_matcher->ns_type == MLX5_FLOW_NAMESPACE_EGRESS)
- return -EINVAL;
-
cmd_in = uverbs_attr_get_alloced_ptr(
attrs, MLX5_IB_ATTR_CREATE_FLOW_MATCH_VALUE);
inlen = uverbs_attr_get_len(attrs,
@@ -629,7 +663,10 @@ DECLARE_UVERBS_NAMED_METHOD(
UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_CREATE_FLOW_ARR_COUNTERS_DEVX_OFFSET,
UVERBS_ATTR_MIN_SIZE(sizeof(u32)),
UA_OPTIONAL,
- UA_ALLOC_AND_COPY));
+ UA_ALLOC_AND_COPY),
+ UVERBS_ATTR_FLAGS_IN(MLX5_IB_ATTR_CREATE_FLOW_FLAGS,
+ enum mlx5_ib_create_flow_flags,
+ UA_OPTIONAL));
DECLARE_UVERBS_NAMED_METHOD_DESTROY(
MLX5_IB_METHOD_DESTROY_FLOW,
diff --git a/drivers/infiniband/hw/mlx5/gsi.c b/drivers/infiniband/hw/mlx5/gsi.c
index 1ae6fd95acaa..40d418153891 100644
--- a/drivers/infiniband/hw/mlx5/gsi.c
+++ b/drivers/infiniband/hw/mlx5/gsi.c
@@ -119,17 +119,15 @@ struct ib_qp *mlx5_ib_gsi_create_qp(struct ib_pd *pd,
struct mlx5_ib_gsi_qp *gsi;
struct ib_qp_init_attr hw_init_attr = *init_attr;
const u8 port_num = init_attr->port_num;
- const int num_pkeys = pd->device->attrs.max_pkeys;
- const int num_qps = mlx5_ib_deth_sqpn_cap(dev) ? num_pkeys : 0;
+ int num_qps = 0;
int ret;
- mlx5_ib_dbg(dev, "creating GSI QP\n");
-
- if (port_num > ARRAY_SIZE(dev->devr.ports) || port_num < 1) {
- mlx5_ib_warn(dev,
- "invalid port number %d during GSI QP creation\n",
- port_num);
- return ERR_PTR(-EINVAL);
+ if (mlx5_ib_deth_sqpn_cap(dev)) {
+ if (MLX5_CAP_GEN(dev->mdev,
+ port_type) == MLX5_CAP_PORT_TYPE_IB)
+ num_qps = pd->device->attrs.max_pkeys;
+ else if (dev->lag_active)
+ num_qps = MLX5_MAX_PORTS;
}
gsi = kzalloc(sizeof(*gsi), GFP_KERNEL);
@@ -270,7 +268,7 @@ static struct ib_qp *create_gsi_ud_qp(struct mlx5_ib_gsi_qp *gsi)
}
static int modify_to_rts(struct mlx5_ib_gsi_qp *gsi, struct ib_qp *qp,
- u16 qp_index)
+ u16 pkey_index)
{
struct mlx5_ib_dev *dev = to_mdev(qp->device);
struct ib_qp_attr attr;
@@ -279,7 +277,7 @@ static int modify_to_rts(struct mlx5_ib_gsi_qp *gsi, struct ib_qp *qp,
mask = IB_QP_STATE | IB_QP_PKEY_INDEX | IB_QP_QKEY | IB_QP_PORT;
attr.qp_state = IB_QPS_INIT;
- attr.pkey_index = qp_index;
+ attr.pkey_index = pkey_index;
attr.qkey = IB_QP1_QKEY;
attr.port_num = gsi->port_num;
ret = ib_modify_qp(qp, &attr, mask);
@@ -313,12 +311,17 @@ static void setup_qp(struct mlx5_ib_gsi_qp *gsi, u16 qp_index)
{
struct ib_device *device = gsi->rx_qp->device;
struct mlx5_ib_dev *dev = to_mdev(device);
+ int pkey_index = qp_index;
+ struct mlx5_ib_qp *mqp;
struct ib_qp *qp;
unsigned long flags;
u16 pkey;
int ret;
- ret = ib_query_pkey(device, gsi->port_num, qp_index, &pkey);
+ if (MLX5_CAP_GEN(dev->mdev, port_type) != MLX5_CAP_PORT_TYPE_IB)
+ pkey_index = 0;
+
+ ret = ib_query_pkey(device, gsi->port_num, pkey_index, &pkey);
if (ret) {
mlx5_ib_warn(dev, "unable to read P_Key at port %d, index %d\n",
gsi->port_num, qp_index);
@@ -347,7 +350,10 @@ static void setup_qp(struct mlx5_ib_gsi_qp *gsi, u16 qp_index)
return;
}
- ret = modify_to_rts(gsi, qp, qp_index);
+ mqp = to_mqp(qp);
+ if (dev->lag_active)
+ mqp->gsi_lag_port = qp_index + 1;
+ ret = modify_to_rts(gsi, qp, pkey_index);
if (ret)
goto err_destroy_qp;
@@ -466,11 +472,15 @@ static int mlx5_ib_gsi_silent_drop(struct mlx5_ib_gsi_qp *gsi,
static struct ib_qp *get_tx_qp(struct mlx5_ib_gsi_qp *gsi, struct ib_ud_wr *wr)
{
struct mlx5_ib_dev *dev = to_mdev(gsi->rx_qp->device);
+ struct mlx5_ib_ah *ah = to_mah(wr->ah);
int qp_index = wr->pkey_index;
- if (!mlx5_ib_deth_sqpn_cap(dev))
+ if (!gsi->num_qps)
return gsi->rx_qp;
+ if (dev->lag_active && ah->xmit_port)
+ qp_index = ah->xmit_port - 1;
+
if (qp_index >= gsi->num_qps)
return NULL;
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.h b/drivers/infiniband/hw/mlx5/ib_rep.h
index 3b6750cba796..5b30d3fa8f8d 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.h
+++ b/drivers/infiniband/hw/mlx5/ib_rep.h
@@ -9,9 +9,9 @@
#include <linux/mlx5/eswitch.h>
#include "mlx5_ib.h"
-#ifdef CONFIG_MLX5_ESWITCH
extern const struct mlx5_ib_profile raw_eth_profile;
+#ifdef CONFIG_MLX5_ESWITCH
u8 mlx5_ib_eswitch_mode(struct mlx5_eswitch *esw);
struct mlx5_ib_dev *mlx5_ib_get_rep_ibdev(struct mlx5_eswitch *esw,
u16 vport_num);
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 566b42f3fb18..343a8b8361e7 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -53,6 +53,7 @@
#include <linux/list.h>
#include <rdma/ib_smi.h>
#include <rdma/ib_umem.h>
+#include <rdma/lag.h>
#include <linux/in.h>
#include <linux/etherdevice.h>
#include "mlx5_ib.h"
@@ -60,6 +61,7 @@
#include "cmd.h"
#include "srq.h"
#include "qp.h"
+#include "wr.h"
#include <linux/mlx5/fs_helpers.h>
#include <linux/mlx5/accel.h>
#include <rdma/uverbs_std_types.h>
@@ -70,17 +72,10 @@
#define UVERBS_MODULE_NAME mlx5_ib
#include <rdma/uverbs_named_ioctl.h>
-#define DRIVER_NAME "mlx5_ib"
-#define DRIVER_VERSION "5.0-0"
-
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
-MODULE_DESCRIPTION("Mellanox Connect-IB HCA IB driver");
+MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) IB driver");
MODULE_LICENSE("Dual BSD/GPL");
-static char mlx5_version[] =
- DRIVER_NAME ": Mellanox Connect-IB Infiniband driver v"
- DRIVER_VERSION "\n";
-
struct mlx5_ib_event_work {
struct work_struct work;
union {
@@ -628,8 +623,8 @@ static int mlx5_ib_del_gid(const struct ib_gid_attr *attr,
attr->index, NULL, NULL);
}
-__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev,
- const struct ib_gid_attr *attr)
+__be16 mlx5_get_roce_udp_sport_min(const struct mlx5_ib_dev *dev,
+ const struct ib_gid_attr *attr)
{
if (attr->gid_type != IB_GID_TYPE_ROCE_UDP_ENCAP)
return 0;
@@ -1004,7 +999,6 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
props->max_mcast_qp_attach = MLX5_CAP_GEN(mdev, max_qp_mcg);
props->max_total_mcast_qp_attach = props->max_mcast_qp_attach *
props->max_mcast_grp;
- props->max_map_per_fmr = INT_MAX; /* no limit in ConnectIB */
props->max_ah = INT_MAX;
props->hca_core_clock = MLX5_CAP_GEN(mdev, device_frequency_khz);
props->timestamp_mask = 0x7FFFFFFFFFFFFFFFULL;
@@ -1964,6 +1958,9 @@ uar_done:
resp.response_length += sizeof(resp.dump_fill_mkey);
}
+ if (MLX5_CAP_GEN(dev->mdev, ece_support))
+ resp.comp_mask |= MLX5_IB_ALLOC_UCONTEXT_RESP_MASK_ECE;
+
err = ib_copy_to_udata(udata, &resp, resp.response_length);
if (err)
goto out_mdev;
@@ -1974,7 +1971,7 @@ uar_done:
context->lib_caps = req.lib_caps;
print_lib_caps(dev, context->lib_caps);
- if (dev->lag_active) {
+ if (mlx5_ib_lag_should_assign_affinity(dev)) {
u8 port = mlx5_core_native_port_num(dev->mdev) - 1;
atomic_set(&context->tx_port_affinity,
@@ -2561,7 +2558,7 @@ static int mlx5_ib_alloc_pd(struct ib_pd *ibpd, struct ib_udata *udata)
struct mlx5_ib_alloc_pd_resp resp;
int err;
u32 out[MLX5_ST_SZ_DW(alloc_pd_out)] = {};
- u32 in[MLX5_ST_SZ_DW(alloc_pd_in)] = {};
+ u32 in[MLX5_ST_SZ_DW(alloc_pd_in)] = {};
u16 uid = 0;
struct mlx5_ib_ucontext *context = rdma_udata_to_drv_context(
udata, struct mlx5_ib_ucontext, ibucontext);
@@ -2569,8 +2566,7 @@ static int mlx5_ib_alloc_pd(struct ib_pd *ibpd, struct ib_udata *udata)
uid = context ? context->devx_uid : 0;
MLX5_SET(alloc_pd_in, in, opcode, MLX5_CMD_OP_ALLOC_PD);
MLX5_SET(alloc_pd_in, in, uid, uid);
- err = mlx5_cmd_exec(to_mdev(ibdev)->mdev, in, sizeof(in),
- out, sizeof(out));
+ err = mlx5_cmd_exec_inout(to_mdev(ibdev)->mdev, alloc_pd, in, out);
if (err)
return err;
@@ -3944,7 +3940,7 @@ static struct ib_flow *mlx5_ib_create_flow(struct ib_qp *qp,
dst->type = MLX5_FLOW_DESTINATION_TYPE_PORT;
} else {
dst->type = MLX5_FLOW_DESTINATION_TYPE_TIR;
- if (mqp->flags & MLX5_IB_QP_RSS)
+ if (mqp->is_rss)
dst->tir_num = mqp->rss_qp.tirn;
else
dst->tir_num = mqp->raw_packet_qp.rq.tirn;
@@ -4199,18 +4195,17 @@ mlx5_ib_raw_fs_rule_add(struct mlx5_ib_dev *dev,
if (dest_type == MLX5_FLOW_DESTINATION_TYPE_TIR) {
dst[dst_num].type = dest_type;
- dst[dst_num].tir_num = dest_id;
+ dst[dst_num++].tir_num = dest_id;
flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
} else if (dest_type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE) {
dst[dst_num].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE_NUM;
- dst[dst_num].ft_num = dest_id;
+ dst[dst_num++].ft_num = dest_id;
flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
- } else {
- dst[dst_num].type = MLX5_FLOW_DESTINATION_TYPE_PORT;
+ } else if (dest_type == MLX5_FLOW_DESTINATION_TYPE_PORT) {
+ dst[dst_num++].type = MLX5_FLOW_DESTINATION_TYPE_PORT;
flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_ALLOW;
}
- dst_num++;
if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) {
dst[dst_num].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
@@ -4420,7 +4415,7 @@ static int mlx5_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
uid = ibqp->pd ?
to_mpd(ibqp->pd)->uid : 0;
- if (mqp->flags & MLX5_IB_QP_UNDERLAY) {
+ if (mqp->flags & IB_QP_CREATE_SOURCE_QPN) {
mlx5_ib_dbg(dev, "Attaching a multi cast group to underlay QP is not supported\n");
return -EOPNOTSUPP;
}
@@ -6194,26 +6189,20 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_VAR_OBJ_ALLOC)(
mmap_offset = mlx5_entry_to_mmap_offset(entry);
length = entry->rdma_entry.npages * PAGE_SIZE;
uobj->object = entry;
+ uverbs_finalize_uobj_create(attrs, MLX5_IB_ATTR_VAR_OBJ_ALLOC_HANDLE);
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_VAR_OBJ_ALLOC_MMAP_OFFSET,
&mmap_offset, sizeof(mmap_offset));
if (err)
- goto err;
+ return err;
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_VAR_OBJ_ALLOC_PAGE_ID,
&entry->page_idx, sizeof(entry->page_idx));
if (err)
- goto err;
+ return err;
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_VAR_OBJ_ALLOC_MMAP_LENGTH,
&length, sizeof(length));
- if (err)
- goto err;
-
- return 0;
-
-err:
- rdma_user_mmap_entry_remove(&entry->rdma_entry);
return err;
}
@@ -6327,26 +6316,20 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_UAR_OBJ_ALLOC)(
mmap_offset = mlx5_entry_to_mmap_offset(entry);
length = entry->rdma_entry.npages * PAGE_SIZE;
uobj->object = entry;
+ uverbs_finalize_uobj_create(attrs, MLX5_IB_ATTR_UAR_OBJ_ALLOC_HANDLE);
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_UAR_OBJ_ALLOC_MMAP_OFFSET,
&mmap_offset, sizeof(mmap_offset));
if (err)
- goto err;
+ return err;
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_UAR_OBJ_ALLOC_PAGE_ID,
&entry->page_idx, sizeof(entry->page_idx));
if (err)
- goto err;
+ return err;
err = uverbs_copy_to(attrs, MLX5_IB_ATTR_UAR_OBJ_ALLOC_MMAP_LENGTH,
&length, sizeof(length));
- if (err)
- goto err;
-
- return 0;
-
-err:
- rdma_user_mmap_entry_remove(&entry->rdma_entry);
return err;
}
@@ -6540,6 +6523,7 @@ static int mlx5_ib_stage_init_init(struct mlx5_ib_dev *dev)
dev->ib_dev.phys_port_cnt = dev->num_ports;
dev->ib_dev.num_comp_vectors = mlx5_comp_vectors_count(mdev);
dev->ib_dev.dev.parent = mdev->device;
+ dev->ib_dev.lag_flags = RDMA_LAG_FLAGS_HASH_ALL_SLAVES;
mutex_init(&dev->cap_mask_mutex);
INIT_LIST_HEAD(&dev->qp_list);
@@ -6629,8 +6613,8 @@ static const struct ib_device_ops mlx5_ib_dev_ops = {
.modify_qp = mlx5_ib_modify_qp,
.modify_srq = mlx5_ib_modify_srq,
.poll_cq = mlx5_ib_poll_cq,
- .post_recv = mlx5_ib_post_recv,
- .post_send = mlx5_ib_post_send,
+ .post_recv = mlx5_ib_post_recv_nodrain,
+ .post_send = mlx5_ib_post_send_nodrain,
.post_srq_recv = mlx5_ib_post_srq_recv,
.process_mad = mlx5_ib_process_mad,
.query_ah = mlx5_ib_query_ah,
@@ -7131,6 +7115,8 @@ void *__mlx5_ib_add(struct mlx5_ib_dev *dev,
int err;
int i;
+ dev->profile = profile;
+
for (i = 0; i < MLX5_IB_STAGE_MAX; i++) {
if (profile->stage[i].init) {
err = profile->stage[i].init(dev);
@@ -7139,7 +7125,6 @@ void *__mlx5_ib_add(struct mlx5_ib_dev *dev,
}
}
- dev->profile = profile;
dev->ib_active = true;
return dev;
@@ -7313,8 +7298,6 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
int port_type_cap;
int num_ports;
- printk_once(KERN_INFO "%s", mlx5_version);
-
if (MLX5_ESWITCH_MANAGER(mdev) &&
mlx5_ib_eswitch_mode(mdev->priv.eswitch) == MLX5_ESWITCH_OFFLOADS) {
if (!mlx5_core_mp_enabled(mdev))
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index aaabb8a98eed..5dbe3eb0d9cb 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -337,7 +337,6 @@ struct mlx5_ib_rwq {
struct ib_umem *umem;
size_t buf_size;
unsigned int page_shift;
- int create_type;
struct mlx5_db db;
u32 user_index;
u32 wqe_count;
@@ -346,17 +345,6 @@ struct mlx5_ib_rwq {
u32 create_flags; /* Use enum mlx5_ib_wq_flags */
};
-enum {
- MLX5_QP_USER,
- MLX5_QP_KERNEL,
- MLX5_QP_EMPTY
-};
-
-enum {
- MLX5_WQ_USER,
- MLX5_WQ_KERNEL
-};
-
struct mlx5_ib_rwq_ind_table {
struct ib_rwq_ind_table ib_rwq_ind_tbl;
u32 rqtn;
@@ -443,34 +431,37 @@ struct mlx5_ib_qp {
/* serialize qp state modifications
*/
struct mutex mutex;
+ /* cached variant of create_flags from struct ib_qp_init_attr */
u32 flags;
u8 port;
u8 state;
- int wq_sig;
- int scat_cqe;
int max_inline_data;
struct mlx5_bf bf;
- int has_rq;
+ u8 has_rq:1;
+ u8 is_rss:1;
/* only for user space QPs. For kernel
* we have it from the bf object
*/
int bfregn;
- int create_type;
-
struct list_head qps_list;
struct list_head cq_recv_list;
struct list_head cq_send_list;
struct mlx5_rate_limit rl;
u32 underlay_qpn;
u32 flags_en;
- /* storage for qp sub type when core qp type is IB_QPT_DRIVER */
- enum ib_qp_type qp_sub_type;
+ /*
+ * IB/core doesn't store low-level QP types, so
+ * store both MLX and IBTA types in the field below.
+ * IB_QPT_DRIVER will be break to DCI/DCT subtypes.
+ */
+ enum ib_qp_type type;
/* A flag to indicate if there's a new counter is configured
* but not take effective
*/
u32 counter_pending;
+ u16 gsi_lag_port;
};
struct mlx5_ib_cq_buf {
@@ -481,24 +472,6 @@ struct mlx5_ib_cq_buf {
int nent;
};
-enum mlx5_ib_qp_flags {
- MLX5_IB_QP_LSO = IB_QP_CREATE_IPOIB_UD_LSO,
- MLX5_IB_QP_BLOCK_MULTICAST_LOOPBACK = IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK,
- MLX5_IB_QP_CROSS_CHANNEL = IB_QP_CREATE_CROSS_CHANNEL,
- MLX5_IB_QP_MANAGED_SEND = IB_QP_CREATE_MANAGED_SEND,
- MLX5_IB_QP_MANAGED_RECV = IB_QP_CREATE_MANAGED_RECV,
- MLX5_IB_QP_SIGNATURE_HANDLING = 1 << 5,
- /* QP uses 1 as its source QP number */
- MLX5_IB_QP_SQPN_QP1 = 1 << 6,
- MLX5_IB_QP_CAP_SCATTER_FCS = 1 << 7,
- MLX5_IB_QP_RSS = 1 << 8,
- MLX5_IB_QP_CVLAN_STRIPPING = 1 << 9,
- MLX5_IB_QP_UNDERLAY = 1 << 10,
- MLX5_IB_QP_PCI_WRITE_END_PADDING = 1 << 11,
- MLX5_IB_QP_TUNNEL_OFFLOAD = 1 << 12,
- MLX5_IB_QP_PACKET_BASED_CREDIT = 1 << 13,
-};
-
struct mlx5_umr_wr {
struct ib_send_wr wr;
u64 virt_addr;
@@ -702,12 +675,6 @@ struct umr_common {
struct semaphore sem;
};
-enum {
- MLX5_FMR_INVALID,
- MLX5_FMR_VALID,
- MLX5_FMR_BUSY,
-};
-
struct mlx5_cache_ent {
struct list_head head;
/* sync access to the cahce entry
@@ -1181,7 +1148,7 @@ void mlx5_ib_db_unmap_user(struct mlx5_ib_ucontext *context, struct mlx5_db *db)
void __mlx5_ib_cq_clean(struct mlx5_ib_cq *cq, u32 qpn, struct mlx5_ib_srq *srq);
void mlx5_ib_cq_clean(struct mlx5_ib_cq *cq, u32 qpn, struct mlx5_ib_srq *srq);
void mlx5_ib_free_srq_wqe(struct mlx5_ib_srq *srq, int wqe_index);
-int mlx5_ib_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr, u32 flags,
+int mlx5_ib_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
int mlx5_ib_query_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr);
void mlx5_ib_destroy_ah(struct ib_ah *ah, u32 flags);
@@ -1205,10 +1172,6 @@ int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr
int mlx5_ib_destroy_qp(struct ib_qp *qp, struct ib_udata *udata);
void mlx5_ib_drain_sq(struct ib_qp *qp);
void mlx5_ib_drain_rq(struct ib_qp *qp);
-int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
- const struct ib_send_wr **bad_wr);
-int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
- const struct ib_recv_wr **bad_wr);
int mlx5_ib_read_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
size_t buflen, size_t *bc);
int mlx5_ib_read_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
@@ -1284,8 +1247,6 @@ int mlx5_query_mad_ifc_port(struct ib_device *ibdev, u8 port,
struct ib_port_attr *props);
int mlx5_ib_query_port(struct ib_device *ibdev, u8 port,
struct ib_port_attr *props);
-int mlx5_ib_init_fmr(struct mlx5_ib_dev *dev);
-void mlx5_ib_cleanup_fmr(struct mlx5_ib_dev *dev);
void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr,
unsigned long max_page_shift,
int *count, int *shift,
@@ -1383,8 +1344,8 @@ int mlx5_ib_get_vf_guid(struct ib_device *device, int vf, u8 port,
int mlx5_ib_set_vf_guid(struct ib_device *device, int vf, u8 port,
u64 guid, int type);
-__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev,
- const struct ib_gid_attr *attr);
+__be16 mlx5_get_roce_udp_sport_min(const struct mlx5_ib_dev *dev,
+ const struct ib_gid_attr *attr);
void mlx5_ib_cleanup_cong_debugfs(struct mlx5_ib_dev *dev, u8 port_num);
void mlx5_ib_init_cong_debugfs(struct mlx5_ib_dev *dev, u8 port_num);
@@ -1581,4 +1542,11 @@ static inline bool mlx5_ib_can_use_umr(struct mlx5_ib_dev *dev,
int mlx5_ib_enable_driver(struct ib_device *dev);
int mlx5_ib_test_wc(struct mlx5_ib_dev *dev);
+
+static inline bool mlx5_ib_lag_should_assign_affinity(struct mlx5_ib_dev *dev)
+{
+ return dev->lag_active ||
+ (MLX5_CAP_GEN(dev->mdev, num_lag_ports) > 1 &&
+ MLX5_CAP_GEN(dev->mdev, lag_tx_port_affinity));
+}
#endif /* MLX5_IB_H */
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index 16af1105cfcf..7d2ec9ee5097 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -447,8 +447,7 @@ static void mlx5_ib_page_fault_resume(struct mlx5_ib_dev *dev,
{
int wq_num = pfault->event_subtype == MLX5_PFAULT_SUBTYPE_WQE ?
pfault->wqe.wq_num : pfault->token;
- u32 out[MLX5_ST_SZ_DW(page_fault_resume_out)] = { };
- u32 in[MLX5_ST_SZ_DW(page_fault_resume_in)] = { };
+ u32 in[MLX5_ST_SZ_DW(page_fault_resume_in)] = {};
int err;
MLX5_SET(page_fault_resume_in, in, opcode, MLX5_CMD_OP_PAGE_FAULT_RESUME);
@@ -457,7 +456,7 @@ static void mlx5_ib_page_fault_resume(struct mlx5_ib_dev *dev,
MLX5_SET(page_fault_resume_in, in, wq_number, wq_num);
MLX5_SET(page_fault_resume_in, in, error, !!error);
- err = mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+ err = mlx5_cmd_exec_in(dev->mdev, page_fault_resume, in);
if (err)
mlx5_ib_err(dev, "Failed to resolve the page fault on WQ 0x%x err %d\n",
wq_num, err);
@@ -1136,8 +1135,7 @@ static int mlx5_ib_mr_initiator_pfault_handler(
if (qp->ibqp.qp_type == IB_QPT_XRC_INI)
*wqe += sizeof(struct mlx5_wqe_xrc_seg);
- if (qp->ibqp.qp_type == IB_QPT_UD ||
- qp->qp_sub_type == MLX5_IB_QPT_DCI) {
+ if (qp->type == IB_QPT_UD || qp->type == MLX5_IB_QPT_DCI) {
av = *wqe;
if (av->dqp_dct & cpu_to_be32(MLX5_EXTENDED_UD_AV))
*wqe += sizeof(struct mlx5_av);
@@ -1190,7 +1188,7 @@ static int mlx5_ib_mr_responder_pfault_handler_rq(struct mlx5_ib_dev *dev,
struct mlx5_ib_wq *wq = &qp->rq;
int wqe_size = 1 << wq->wqe_shift;
- if (qp->wq_sig) {
+ if (qp->flags_en & MLX5_QP_FLAG_SIGNATURE) {
mlx5_ib_err(dev, "ODP fault with WQE signatures is not supported\n");
return -EFAULT;
}
diff --git a/drivers/infiniband/hw/mlx5/qos.c b/drivers/infiniband/hw/mlx5/qos.c
index cac878a70edb..dce92554142a 100644
--- a/drivers/infiniband/hw/mlx5/qos.c
+++ b/drivers/infiniband/hw/mlx5/qos.c
@@ -69,17 +69,14 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_PP_OBJ_ALLOC)(
if (err)
goto err;
- err = uverbs_copy_to(attrs, MLX5_IB_ATTR_PP_OBJ_ALLOC_INDEX,
- &pp_entry->index, sizeof(pp_entry->index));
- if (err)
- goto clean;
-
pp_entry->mdev = dev->mdev;
uobj->object = pp_entry;
- return 0;
+ uverbs_finalize_uobj_create(attrs, MLX5_IB_ATTR_PP_OBJ_ALLOC_HANDLE);
+
+ err = uverbs_copy_to(attrs, MLX5_IB_ATTR_PP_OBJ_ALLOC_INDEX,
+ &pp_entry->index, sizeof(pp_entry->index));
+ return err;
-clean:
- mlx5_rl_remove_rate_raw(dev->mdev, pp_entry->index);
err:
kfree(pp_entry);
return err;
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index d93eec5d3277..81bf6b975e0e 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -40,9 +40,7 @@
#include "ib_rep.h"
#include "cmd.h"
#include "qp.h"
-
-/* not supported currently */
-static int wq_signature;
+#include "wr.h"
enum {
MLX5_IB_ACK_REQ_FREQ = 8,
@@ -55,32 +53,6 @@ enum {
MLX5_IB_LINK_TYPE_ETH = 1
};
-enum {
- MLX5_IB_SQ_STRIDE = 6,
- MLX5_IB_SQ_UMR_INLINE_THRESHOLD = 64,
-};
-
-static const u32 mlx5_ib_opcode[] = {
- [IB_WR_SEND] = MLX5_OPCODE_SEND,
- [IB_WR_LSO] = MLX5_OPCODE_LSO,
- [IB_WR_SEND_WITH_IMM] = MLX5_OPCODE_SEND_IMM,
- [IB_WR_RDMA_WRITE] = MLX5_OPCODE_RDMA_WRITE,
- [IB_WR_RDMA_WRITE_WITH_IMM] = MLX5_OPCODE_RDMA_WRITE_IMM,
- [IB_WR_RDMA_READ] = MLX5_OPCODE_RDMA_READ,
- [IB_WR_ATOMIC_CMP_AND_SWP] = MLX5_OPCODE_ATOMIC_CS,
- [IB_WR_ATOMIC_FETCH_AND_ADD] = MLX5_OPCODE_ATOMIC_FA,
- [IB_WR_SEND_WITH_INV] = MLX5_OPCODE_SEND_INVAL,
- [IB_WR_LOCAL_INV] = MLX5_OPCODE_UMR,
- [IB_WR_REG_MR] = MLX5_OPCODE_UMR,
- [IB_WR_MASKED_ATOMIC_CMP_AND_SWP] = MLX5_OPCODE_ATOMIC_MASKED_CS,
- [IB_WR_MASKED_ATOMIC_FETCH_AND_ADD] = MLX5_OPCODE_ATOMIC_MASKED_FA,
- [MLX5_IB_WR_UMR] = MLX5_OPCODE_UMR,
-};
-
-struct mlx5_wqe_eth_pad {
- u8 rsvd0[16];
-};
-
enum raw_qp_set_mask_map {
MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID = 1UL << 0,
MLX5_RAW_QP_RATE_LIMIT = 1UL << 1,
@@ -392,17 +364,26 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap,
cap->max_recv_wr = 0;
cap->max_recv_sge = 0;
} else {
+ int wq_sig = !!(qp->flags_en & MLX5_QP_FLAG_SIGNATURE);
+
if (ucmd) {
qp->rq.wqe_cnt = ucmd->rq_wqe_count;
if (ucmd->rq_wqe_shift > BITS_PER_BYTE * sizeof(ucmd->rq_wqe_shift))
return -EINVAL;
qp->rq.wqe_shift = ucmd->rq_wqe_shift;
- if ((1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) < qp->wq_sig)
+ if ((1 << qp->rq.wqe_shift) /
+ sizeof(struct mlx5_wqe_data_seg) <
+ wq_sig)
return -EINVAL;
- qp->rq.max_gs = (1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig;
+ qp->rq.max_gs =
+ (1 << qp->rq.wqe_shift) /
+ sizeof(struct mlx5_wqe_data_seg) -
+ wq_sig;
qp->rq.max_post = qp->rq.wqe_cnt;
} else {
- wqe_size = qp->wq_sig ? sizeof(struct mlx5_wqe_signature_seg) : 0;
+ wqe_size =
+ wq_sig ? sizeof(struct mlx5_wqe_signature_seg) :
+ 0;
wqe_size += cap->max_recv_sge * sizeof(struct mlx5_wqe_data_seg);
wqe_size = roundup_pow_of_two(wqe_size);
wq_size = roundup_pow_of_two(cap->max_recv_wr) * wqe_size;
@@ -416,7 +397,10 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap,
return -EINVAL;
}
qp->rq.wqe_shift = ilog2(wqe_size);
- qp->rq.max_gs = (1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig;
+ qp->rq.max_gs =
+ (1 << qp->rq.wqe_shift) /
+ sizeof(struct mlx5_wqe_data_seg) -
+ wq_sig;
qp->rq.max_post = qp->rq.wqe_cnt;
}
}
@@ -596,7 +580,7 @@ static int set_user_buf_size(struct mlx5_ib_dev *dev,
}
if (attr->qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) {
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) {
base->ubuffer.buf_size = qp->rq.wqe_cnt << qp->rq.wqe_shift;
qp->raw_packet_qp.sq.ubuffer.buf_size = qp->sq.wqe_cnt << 6;
} else {
@@ -751,10 +735,7 @@ static int to_mlx5_st(enum ib_qp_type type)
case IB_QPT_SMI: return MLX5_QP_ST_QP0;
case MLX5_IB_QPT_HW_GSI: return MLX5_QP_ST_QP1;
case MLX5_IB_QPT_DCI: return MLX5_QP_ST_DCI;
- case IB_QPT_RAW_IPV6: return MLX5_QP_ST_RAW_IPV6;
- case IB_QPT_RAW_PACKET:
- case IB_QPT_RAW_ETHERTYPE: return MLX5_QP_ST_RAW_ETHERTYPE;
- case IB_QPT_MAX:
+ case IB_QPT_RAW_PACKET: return MLX5_QP_ST_RAW_ETHERTYPE;
default: return -EINVAL;
}
}
@@ -891,7 +872,6 @@ static int create_user_rq(struct mlx5_ib_dev *dev, struct ib_pd *pd,
goto err_umem;
}
- rwq->create_type = MLX5_WQ_USER;
return 0;
err_umem:
@@ -906,15 +886,14 @@ static int adjust_bfregn(struct mlx5_ib_dev *dev,
bfregn % MLX5_NON_FP_BFREGS_PER_UAR;
}
-static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
- struct mlx5_ib_qp *qp, struct ib_udata *udata,
- struct ib_qp_init_attr *attr,
- u32 **in,
- struct mlx5_ib_create_qp_resp *resp, int *inlen,
- struct mlx5_ib_qp_base *base)
+static int _create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp, struct ib_udata *udata,
+ struct ib_qp_init_attr *attr, u32 **in,
+ struct mlx5_ib_create_qp_resp *resp, int *inlen,
+ struct mlx5_ib_qp_base *base,
+ struct mlx5_ib_create_qp *ucmd)
{
struct mlx5_ib_ucontext *context;
- struct mlx5_ib_create_qp ucmd;
struct mlx5_ib_ubuffer *ubuffer = &base->ubuffer;
int page_shift = 0;
int uar_index = 0;
@@ -928,30 +907,24 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
u16 uid;
u32 uar_flags;
- err = ib_copy_from_udata(&ucmd, udata, sizeof(ucmd));
- if (err) {
- mlx5_ib_dbg(dev, "copy failed\n");
- return err;
- }
-
context = rdma_udata_to_drv_context(udata, struct mlx5_ib_ucontext,
ibucontext);
- uar_flags = ucmd.flags & (MLX5_QP_FLAG_UAR_PAGE_INDEX |
- MLX5_QP_FLAG_BFREG_INDEX);
+ uar_flags = qp->flags_en &
+ (MLX5_QP_FLAG_UAR_PAGE_INDEX | MLX5_QP_FLAG_BFREG_INDEX);
switch (uar_flags) {
case MLX5_QP_FLAG_UAR_PAGE_INDEX:
- uar_index = ucmd.bfreg_index;
+ uar_index = ucmd->bfreg_index;
bfregn = MLX5_IB_INVALID_BFREG;
break;
case MLX5_QP_FLAG_BFREG_INDEX:
uar_index = bfregn_to_uar_index(dev, &context->bfregi,
- ucmd.bfreg_index, true);
+ ucmd->bfreg_index, true);
if (uar_index < 0)
return uar_index;
bfregn = MLX5_IB_INVALID_BFREG;
break;
case 0:
- if (qp->flags & MLX5_IB_QP_CROSS_CHANNEL)
+ if (qp->flags & IB_QP_CREATE_CROSS_CHANNEL)
return -EINVAL;
bfregn = alloc_bfreg(dev, &context->bfregi);
if (bfregn < 0)
@@ -970,12 +943,12 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
qp->sq.wqe_shift = ilog2(MLX5_SEND_WQE_BB);
qp->sq.offset = qp->rq.wqe_cnt << qp->rq.wqe_shift;
- err = set_user_buf_size(dev, qp, &ucmd, base, attr);
+ err = set_user_buf_size(dev, qp, ucmd, base, attr);
if (err)
goto err_bfreg;
- if (ucmd.buf_addr && ubuffer->buf_size) {
- ubuffer->buf_addr = ucmd.buf_addr;
+ if (ucmd->buf_addr && ubuffer->buf_size) {
+ ubuffer->buf_addr = ucmd->buf_addr;
err = mlx5_ib_umem_get(dev, udata, ubuffer->buf_addr,
ubuffer->buf_size, &ubuffer->umem,
&npages, &page_shift, &ncont, &offset);
@@ -993,8 +966,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
goto err_umem;
}
- uid = (attr->qp_type != IB_QPT_XRC_TGT &&
- attr->qp_type != IB_QPT_XRC_INI) ? to_mpd(pd)->uid : 0;
+ uid = (attr->qp_type != IB_QPT_XRC_INI) ? to_mpd(pd)->uid : 0;
MLX5_SET(create_qp_in, *in, uid, uid);
pas = (__be64 *)MLX5_ADDR_OF(create_qp_in, *in, pas);
if (ubuffer->umem)
@@ -1012,24 +984,14 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
resp->bfreg_index = MLX5_IB_INVALID_BFREG;
qp->bfregn = bfregn;
- err = mlx5_ib_db_map_user(context, udata, ucmd.db_addr, &qp->db);
+ err = mlx5_ib_db_map_user(context, udata, ucmd->db_addr, &qp->db);
if (err) {
mlx5_ib_dbg(dev, "map failed\n");
goto err_free;
}
- err = ib_copy_to_udata(udata, resp, min(udata->outlen, sizeof(*resp)));
- if (err) {
- mlx5_ib_dbg(dev, "copy failed\n");
- goto err_unmap;
- }
- qp->create_type = MLX5_QP_USER;
-
return 0;
-err_unmap:
- mlx5_ib_db_unmap_user(context, &qp->db);
-
err_free:
kvfree(*in);
@@ -1042,72 +1004,50 @@ err_bfreg:
return err;
}
-static void destroy_qp_user(struct mlx5_ib_dev *dev, struct ib_pd *pd,
- struct mlx5_ib_qp *qp, struct mlx5_ib_qp_base *base,
- struct ib_udata *udata)
+static void destroy_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ struct mlx5_ib_qp_base *base, struct ib_udata *udata)
{
- struct mlx5_ib_ucontext *context =
- rdma_udata_to_drv_context(
- udata,
- struct mlx5_ib_ucontext,
- ibucontext);
-
- mlx5_ib_db_unmap_user(context, &qp->db);
- ib_umem_release(base->ubuffer.umem);
-
- /*
- * Free only the BFREGs which are handled by the kernel.
- * BFREGs of UARs allocated dynamically are handled by user.
- */
- if (qp->bfregn != MLX5_IB_INVALID_BFREG)
- mlx5_ib_free_bfreg(dev, &context->bfregi, qp->bfregn);
-}
+ struct mlx5_ib_ucontext *context = rdma_udata_to_drv_context(
+ udata, struct mlx5_ib_ucontext, ibucontext);
-/* get_sq_edge - Get the next nearby edge.
- *
- * An 'edge' is defined as the first following address after the end
- * of the fragment or the SQ. Accordingly, during the WQE construction
- * which repetitively increases the pointer to write the next data, it
- * simply should check if it gets to an edge.
- *
- * @sq - SQ buffer.
- * @idx - Stride index in the SQ buffer.
- *
- * Return:
- * The new edge.
- */
-static void *get_sq_edge(struct mlx5_ib_wq *sq, u32 idx)
-{
- void *fragment_end;
+ if (udata) {
+ /* User QP */
+ mlx5_ib_db_unmap_user(context, &qp->db);
+ ib_umem_release(base->ubuffer.umem);
- fragment_end = mlx5_frag_buf_get_wqe
- (&sq->fbc,
- mlx5_frag_buf_get_idx_last_contig_stride(&sq->fbc, idx));
+ /*
+ * Free only the BFREGs which are handled by the kernel.
+ * BFREGs of UARs allocated dynamically are handled by user.
+ */
+ if (qp->bfregn != MLX5_IB_INVALID_BFREG)
+ mlx5_ib_free_bfreg(dev, &context->bfregi, qp->bfregn);
+ return;
+ }
- return fragment_end + MLX5_SEND_WQE_BB;
+ /* Kernel QP */
+ kvfree(qp->sq.wqe_head);
+ kvfree(qp->sq.w_list);
+ kvfree(qp->sq.wrid);
+ kvfree(qp->sq.wr_data);
+ kvfree(qp->rq.wrid);
+ if (qp->db.db)
+ mlx5_db_free(dev->mdev, &qp->db);
+ if (qp->buf.frags)
+ mlx5_frag_buf_free(dev->mdev, &qp->buf);
}
-static int create_kernel_qp(struct mlx5_ib_dev *dev,
- struct ib_qp_init_attr *init_attr,
- struct mlx5_ib_qp *qp,
- u32 **in, int *inlen,
- struct mlx5_ib_qp_base *base)
+static int _create_kernel_qp(struct mlx5_ib_dev *dev,
+ struct ib_qp_init_attr *init_attr,
+ struct mlx5_ib_qp *qp, u32 **in, int *inlen,
+ struct mlx5_ib_qp_base *base)
{
int uar_index;
void *qpc;
int err;
- if (init_attr->create_flags & ~(IB_QP_CREATE_INTEGRITY_EN |
- IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK |
- IB_QP_CREATE_IPOIB_UD_LSO |
- IB_QP_CREATE_NETIF_QP |
- MLX5_IB_QP_CREATE_SQPN_QP1 |
- MLX5_IB_QP_CREATE_WC_TEST))
- return -EINVAL;
-
if (init_attr->qp_type == MLX5_IB_QPT_REG_UMR)
qp->bf.bfreg = &dev->fp_bfreg;
- else if (init_attr->create_flags & MLX5_IB_QP_CREATE_WC_TEST)
+ else if (qp->flags & MLX5_IB_QP_CREATE_WC_TEST)
qp->bf.bfreg = &dev->wc_bfreg;
else
qp->bf.bfreg = &dev->bfreg;
@@ -1167,10 +1107,8 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev,
MLX5_SET(qpc, qpc, fre, 1);
MLX5_SET(qpc, qpc, rlky, 1);
- if (init_attr->create_flags & MLX5_IB_QP_CREATE_SQPN_QP1) {
+ if (qp->flags & MLX5_IB_QP_CREATE_SQPN_QP1)
MLX5_SET(qpc, qpc, deth_sqpn, 1);
- qp->flags |= MLX5_IB_QP_SQPN_QP1;
- }
mlx5_fill_page_frag_array(&qp->buf,
(__be64 *)MLX5_ADDR_OF(create_qp_in,
@@ -1198,7 +1136,6 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev,
err = -ENOMEM;
goto err_wrid;
}
- qp->create_type = MLX5_QP_KERNEL;
return 0;
@@ -1218,36 +1155,15 @@ err_buf:
return err;
}
-static void destroy_qp_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp)
-{
- kvfree(qp->sq.wqe_head);
- kvfree(qp->sq.w_list);
- kvfree(qp->sq.wrid);
- kvfree(qp->sq.wr_data);
- kvfree(qp->rq.wrid);
- mlx5_db_free(dev->mdev, &qp->db);
- mlx5_frag_buf_free(dev->mdev, &qp->buf);
-}
-
static u32 get_rx_type(struct mlx5_ib_qp *qp, struct ib_qp_init_attr *attr)
{
- if (attr->srq || (attr->qp_type == IB_QPT_XRC_TGT) ||
- (attr->qp_type == MLX5_IB_QPT_DCI) ||
- (attr->qp_type == IB_QPT_XRC_INI))
+ if (attr->srq || (qp->type == IB_QPT_XRC_TGT) ||
+ (qp->type == MLX5_IB_QPT_DCI) || (qp->type == IB_QPT_XRC_INI))
return MLX5_SRQ_RQ;
else if (!qp->has_rq)
return MLX5_ZERO_LEN_RQ;
- else
- return MLX5_NON_ZERO_RQ;
-}
-
-static int is_connected(enum ib_qp_type qp_type)
-{
- if (qp_type == IB_QPT_RC || qp_type == IB_QPT_UC ||
- qp_type == MLX5_IB_QPT_DCI)
- return 1;
- return 0;
+ return MLX5_NON_ZERO_RQ;
}
static int create_raw_packet_qp_tis(struct mlx5_ib_dev *dev,
@@ -1260,7 +1176,7 @@ static int create_raw_packet_qp_tis(struct mlx5_ib_dev *dev,
MLX5_SET(create_tis_in, in, uid, to_mpd(pd)->uid);
MLX5_SET(tisc, tisc, transport_domain, tdn);
- if (qp->flags & MLX5_IB_QP_UNDERLAY)
+ if (qp->flags & IB_QP_CREATE_SOURCE_QPN)
MLX5_SET(tisc, tisc, underlay_qpn, qp->underlay_qpn);
return mlx5_core_create_tis(dev->mdev, in, &sq->tisn);
@@ -1409,7 +1325,7 @@ static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev,
MLX5_SET(rqc, rqc, user_index, MLX5_GET(qpc, qpc, user_index));
MLX5_SET(rqc, rqc, cqn, MLX5_GET(qpc, qpc, cqn_rcv));
- if (mqp->flags & MLX5_IB_QP_CAP_SCATTER_FCS)
+ if (mqp->flags & IB_QP_CREATE_SCATTER_FCS)
MLX5_SET(rqc, rqc, scatter_fcs, 1);
wq = MLX5_ADDR_OF(rqc, rqc, wq);
@@ -1440,13 +1356,6 @@ static void destroy_raw_packet_qp_rq(struct mlx5_ib_dev *dev,
mlx5_core_destroy_rq_tracked(dev, &rq->base.mqp);
}
-static bool tunnel_offload_supported(struct mlx5_core_dev *dev)
-{
- return (MLX5_CAP_ETH(dev, tunnel_stateless_vxlan) ||
- MLX5_CAP_ETH(dev, tunnel_stateless_gre) ||
- MLX5_CAP_ETH(dev, tunnel_stateless_geneve_rx));
-}
-
static void destroy_raw_packet_qp_tir(struct mlx5_ib_dev *dev,
struct mlx5_ib_rq *rq,
u32 qp_flags_en,
@@ -1524,6 +1433,8 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
u16 uid = to_mpd(pd)->uid;
u32 out[MLX5_ST_SZ_DW(create_tir_out)] = {};
+ if (!qp->sq.wqe_cnt && !qp->rq.wqe_cnt)
+ return -EINVAL;
if (qp->sq.wqe_cnt) {
err = create_raw_packet_qp_tis(dev, qp, sq, tdn, pd);
if (err)
@@ -1547,9 +1458,9 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
if (qp->rq.wqe_cnt) {
rq->base.container_mibqp = qp;
- if (qp->flags & MLX5_IB_QP_CVLAN_STRIPPING)
+ if (qp->flags & IB_QP_CREATE_CVLAN_STRIPPING)
rq->flags |= MLX5_IB_RQ_CVLAN_STRIPPING;
- if (qp->flags & MLX5_IB_QP_PCI_WRITE_END_PADDING)
+ if (qp->flags & IB_QP_CREATE_PCI_WRITE_END_PADDING)
rq->flags |= MLX5_IB_RQ_PCI_WRITE_END_PADDING;
err = create_raw_packet_qp_rq(dev, rq, in, inlen, pd);
if (err)
@@ -1584,14 +1495,8 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
qp->trans_qp.base.mqp.qpn = qp->sq.wqe_cnt ? sq->base.mqp.qpn :
rq->base.mqp.qpn;
- err = ib_copy_to_udata(udata, resp, min(udata->outlen, sizeof(*resp)));
- if (err)
- goto err_destroy_tir;
-
return 0;
-err_destroy_tir:
- destroy_raw_packet_qp_tir(dev, rq, qp->flags_en, pd);
err_destroy_rq:
destroy_raw_packet_qp_rq(dev, rq);
err_destroy_sq:
@@ -1643,14 +1548,27 @@ static void destroy_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *q
to_mpd(qp->ibqp.pd)->uid);
}
-static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
- struct ib_pd *pd,
- struct ib_qp_init_attr *init_attr,
- struct ib_udata *udata)
+struct mlx5_create_qp_params {
+ struct ib_udata *udata;
+ size_t inlen;
+ size_t outlen;
+ size_t ucmd_size;
+ void *ucmd;
+ u8 is_rss_raw : 1;
+ struct ib_qp_init_attr *attr;
+ u32 uidx;
+ struct mlx5_ib_create_qp_resp resp;
+};
+
+static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
{
+ struct ib_qp_init_attr *init_attr = params->attr;
+ struct mlx5_ib_create_qp_rss *ucmd = params->ucmd;
+ struct ib_udata *udata = params->udata;
struct mlx5_ib_ucontext *mucontext = rdma_udata_to_drv_context(
udata, struct mlx5_ib_ucontext, ibucontext);
- struct mlx5_ib_create_qp_resp resp = {};
int inlen;
int outlen;
int err;
@@ -1660,79 +1578,28 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
void *hfso;
u32 selected_fields = 0;
u32 outer_l4;
- size_t min_resp_len;
u32 tdn = mucontext->tdn;
- struct mlx5_ib_create_qp_rss ucmd = {};
- size_t required_cmd_sz;
u8 lb_flag = 0;
- if (init_attr->qp_type != IB_QPT_RAW_PACKET)
- return -EOPNOTSUPP;
-
- if (init_attr->create_flags || init_attr->send_cq)
- return -EINVAL;
-
- min_resp_len = offsetof(typeof(resp), bfreg_index) + sizeof(resp.bfreg_index);
- if (udata->outlen < min_resp_len)
- return -EINVAL;
-
- required_cmd_sz = offsetof(typeof(ucmd), flags) + sizeof(ucmd.flags);
- if (udata->inlen < required_cmd_sz) {
- mlx5_ib_dbg(dev, "invalid inlen\n");
- return -EINVAL;
- }
-
- if (udata->inlen > sizeof(ucmd) &&
- !ib_is_udata_cleared(udata, sizeof(ucmd),
- udata->inlen - sizeof(ucmd))) {
- mlx5_ib_dbg(dev, "inlen is not supported\n");
- return -EOPNOTSUPP;
- }
-
- if (ib_copy_from_udata(&ucmd, udata, min(sizeof(ucmd), udata->inlen))) {
- mlx5_ib_dbg(dev, "copy failed\n");
- return -EFAULT;
- }
-
- if (ucmd.comp_mask) {
+ if (ucmd->comp_mask) {
mlx5_ib_dbg(dev, "invalid comp mask\n");
return -EOPNOTSUPP;
}
- if (ucmd.flags & ~(MLX5_QP_FLAG_TUNNEL_OFFLOADS |
- MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC |
- MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC)) {
- mlx5_ib_dbg(dev, "invalid flags\n");
- return -EOPNOTSUPP;
- }
-
- if (ucmd.flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS &&
- !tunnel_offload_supported(dev->mdev)) {
- mlx5_ib_dbg(dev, "tunnel offloads isn't supported\n");
- return -EOPNOTSUPP;
- }
-
- if (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_INNER &&
- !(ucmd.flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS)) {
+ if (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_INNER &&
+ !(ucmd->flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS)) {
mlx5_ib_dbg(dev, "Tunnel offloads must be set for inner RSS\n");
return -EOPNOTSUPP;
}
- if (ucmd.flags & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC || dev->is_rep) {
- lb_flag |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST;
+ if (dev->is_rep)
qp->flags_en |= MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC;
- }
- if (ucmd.flags & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC) {
- lb_flag |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST;
- qp->flags_en |= MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC;
- }
+ if (qp->flags_en & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC)
+ lb_flag |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST;
- err = ib_copy_to_udata(udata, &resp, min(udata->outlen, sizeof(resp)));
- if (err) {
- mlx5_ib_dbg(dev, "copy failed\n");
- return -EINVAL;
- }
+ if (qp->flags_en & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC)
+ lb_flag |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST;
inlen = MLX5_ST_SZ_BYTES(create_tir_in);
outlen = MLX5_ST_SZ_BYTES(create_tir_out);
@@ -1751,29 +1618,29 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer);
- if (ucmd.flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS)
+ if (ucmd->flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS)
MLX5_SET(tirc, tirc, tunneled_offload_en, 1);
MLX5_SET(tirc, tirc, self_lb_block, lb_flag);
- if (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_INNER)
+ if (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_INNER)
hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_inner);
else
hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer);
- switch (ucmd.rx_hash_function) {
+ switch (ucmd->rx_hash_function) {
case MLX5_RX_HASH_FUNC_TOEPLITZ:
{
void *rss_key = MLX5_ADDR_OF(tirc, tirc, rx_hash_toeplitz_key);
size_t len = MLX5_FLD_SZ_BYTES(tirc, rx_hash_toeplitz_key);
- if (len != ucmd.rx_key_len) {
+ if (len != ucmd->rx_key_len) {
err = -EINVAL;
goto err;
}
MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_TOEPLITZ);
- memcpy(rss_key, ucmd.rx_hash_key, len);
+ memcpy(rss_key, ucmd->rx_hash_key, len);
break;
}
default:
@@ -1781,7 +1648,7 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
goto err;
}
- if (!ucmd.rx_hash_fields_mask) {
+ if (!ucmd->rx_hash_fields_mask) {
/* special case when this TIR serves as steering entry without hashing */
if (!init_attr->rwq_ind_tbl->log_ind_tbl_size)
goto create_tir;
@@ -1789,29 +1656,31 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
goto err;
}
- if (((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4)) &&
- ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))) {
+ if (((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4)) &&
+ ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))) {
err = -EINVAL;
goto err;
}
/* If none of IPV4 & IPV6 SRC/DST was set - this bit field is ignored */
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4))
MLX5_SET(rx_hash_field_select, hfso, l3_prot_type,
MLX5_L3_PROT_TYPE_IPV4);
- else if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))
+ else if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))
MLX5_SET(rx_hash_field_select, hfso, l3_prot_type,
MLX5_L3_PROT_TYPE_IPV6);
- outer_l4 = ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP)) << 0 |
- ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP)) << 1 |
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_IPSEC_SPI) << 2;
+ outer_l4 = ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP))
+ << 0 |
+ ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP))
+ << 1 |
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_IPSEC_SPI) << 2;
/* Check that only one l4 protocol is set */
if (outer_l4 & (outer_l4 - 1)) {
@@ -1820,32 +1689,32 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
}
/* If none of TCP & UDP SRC/DST was set - this bit field is ignored */
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP))
MLX5_SET(rx_hash_field_select, hfso, l4_prot_type,
MLX5_L4_PROT_TYPE_TCP);
- else if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP))
+ else if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP))
MLX5_SET(rx_hash_field_select, hfso, l4_prot_type,
MLX5_L4_PROT_TYPE_UDP);
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV4) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_IPV6))
selected_fields |= MLX5_HASH_FIELD_SEL_SRC_IP;
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV4) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_IPV6))
selected_fields |= MLX5_HASH_FIELD_SEL_DST_IP;
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_TCP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_SRC_PORT_UDP))
selected_fields |= MLX5_HASH_FIELD_SEL_L4_SPORT;
- if ((ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP) ||
- (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP))
+ if ((ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_TCP) ||
+ (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_DST_PORT_UDP))
selected_fields |= MLX5_HASH_FIELD_SEL_L4_DPORT;
- if (ucmd.rx_hash_fields_mask & MLX5_RX_HASH_IPSEC_SPI)
+ if (ucmd->rx_hash_fields_mask & MLX5_RX_HASH_IPSEC_SPI)
selected_fields |= MLX5_HASH_FIELD_SEL_IPSEC_SPI;
MLX5_SET(rx_hash_field_select, hfso, selected_fields, selected_fields);
@@ -1867,73 +1736,43 @@ create_tir:
goto err;
if (mucontext->devx_uid) {
- resp.comp_mask |= MLX5_IB_CREATE_QP_RESP_MASK_TIRN;
- resp.tirn = qp->rss_qp.tirn;
+ params->resp.comp_mask |= MLX5_IB_CREATE_QP_RESP_MASK_TIRN;
+ params->resp.tirn = qp->rss_qp.tirn;
if (MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev, sw_owner)) {
- resp.tir_icm_addr =
+ params->resp.tir_icm_addr =
MLX5_GET(create_tir_out, out, icm_address_31_0);
- resp.tir_icm_addr |= (u64)MLX5_GET(create_tir_out, out,
- icm_address_39_32)
- << 32;
- resp.tir_icm_addr |= (u64)MLX5_GET(create_tir_out, out,
- icm_address_63_40)
- << 40;
- resp.comp_mask |=
+ params->resp.tir_icm_addr |=
+ (u64)MLX5_GET(create_tir_out, out,
+ icm_address_39_32)
+ << 32;
+ params->resp.tir_icm_addr |=
+ (u64)MLX5_GET(create_tir_out, out,
+ icm_address_63_40)
+ << 40;
+ params->resp.comp_mask |=
MLX5_IB_CREATE_QP_RESP_MASK_TIR_ICM_ADDR;
}
}
- err = ib_copy_to_udata(udata, &resp, min(udata->outlen, sizeof(resp)));
- if (err)
- goto err_copy;
-
kvfree(in);
/* qpn is reserved for that QP */
qp->trans_qp.base.mqp.qpn = 0;
- qp->flags |= MLX5_IB_QP_RSS;
+ qp->is_rss = true;
return 0;
-err_copy:
- mlx5_cmd_destroy_tir(dev->mdev, qp->rss_qp.tirn, mucontext->devx_uid);
err:
kvfree(in);
return err;
}
-static void configure_responder_scat_cqe(struct ib_qp_init_attr *init_attr,
- void *qpc)
-{
- int rcqe_sz;
-
- if (init_attr->qp_type == MLX5_IB_QPT_DCI)
- return;
-
- rcqe_sz = mlx5_ib_get_cqe_size(init_attr->recv_cq);
-
- if (init_attr->qp_type == MLX5_IB_QPT_DCT) {
- if (rcqe_sz == 128)
- MLX5_SET(dctc, qpc, cs_res, MLX5_RES_SCAT_DATA64_CQE);
-
- return;
- }
-
- MLX5_SET(qpc, qpc, cs_res,
- rcqe_sz == 128 ? MLX5_RES_SCAT_DATA64_CQE :
- MLX5_RES_SCAT_DATA32_CQE);
-}
-
static void configure_requester_scat_cqe(struct mlx5_ib_dev *dev,
struct ib_qp_init_attr *init_attr,
struct mlx5_ib_create_qp *ucmd,
void *qpc)
{
- enum ib_qp_type qpt = init_attr->qp_type;
int scqe_sz;
bool allow_scat_cqe = false;
- if (qpt == IB_QPT_UC || qpt == IB_QPT_UD)
- return;
-
if (ucmd)
allow_scat_cqe = ucmd->flags & MLX5_QP_FLAG_ALLOW_SCATTER_CQE;
@@ -1998,269 +1837,182 @@ static int get_atomic_mode(struct mlx5_ib_dev *dev,
return atomic_mode;
}
-static inline bool check_flags_mask(uint64_t input, uint64_t supported)
-{
- return (input & ~supported) == 0;
-}
-
-static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
- struct ib_qp_init_attr *init_attr,
- struct ib_udata *udata, struct mlx5_ib_qp *qp)
+static int create_xrc_tgt_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
{
+ struct mlx5_ib_create_qp *ucmd = params->ucmd;
+ struct ib_qp_init_attr *attr = params->attr;
+ u32 uidx = params->uidx;
struct mlx5_ib_resources *devr = &dev->devr;
+ u32 out[MLX5_ST_SZ_DW(create_qp_out)] = {};
int inlen = MLX5_ST_SZ_BYTES(create_qp_in);
struct mlx5_core_dev *mdev = dev->mdev;
- struct mlx5_ib_create_qp_resp resp = {};
- struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
- udata, struct mlx5_ib_ucontext, ibucontext);
- struct mlx5_ib_cq *send_cq;
- struct mlx5_ib_cq *recv_cq;
- unsigned long flags;
- u32 uidx = MLX5_IB_DEFAULT_UIDX;
- struct mlx5_ib_create_qp ucmd;
struct mlx5_ib_qp_base *base;
- int mlx5_st;
+ unsigned long flags;
void *qpc;
u32 *in;
int err;
mutex_init(&qp->mutex);
- spin_lock_init(&qp->sq.lock);
- spin_lock_init(&qp->rq.lock);
- mlx5_st = to_mlx5_st(init_attr->qp_type);
- if (mlx5_st < 0)
- return -EINVAL;
+ if (attr->sq_sig_type == IB_SIGNAL_ALL_WR)
+ qp->sq_signal_bits = MLX5_WQE_CTRL_CQ_UPDATE;
- if (init_attr->rwq_ind_tbl) {
- if (!udata)
- return -ENOSYS;
+ in = kvzalloc(inlen, GFP_KERNEL);
+ if (!in)
+ return -ENOMEM;
- err = create_rss_raw_qp_tir(dev, qp, pd, init_attr, udata);
- return err;
- }
+ if (MLX5_CAP_GEN(mdev, ece_support))
+ MLX5_SET(create_qp_in, in, ece, ucmd->ece_options);
+ qpc = MLX5_ADDR_OF(create_qp_in, in, qpc);
- if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) {
- if (!MLX5_CAP_GEN(mdev, block_lb_mc)) {
- mlx5_ib_dbg(dev, "block multicast loopback isn't supported\n");
- return -EINVAL;
- } else {
- qp->flags |= MLX5_IB_QP_BLOCK_MULTICAST_LOOPBACK;
- }
- }
+ MLX5_SET(qpc, qpc, st, MLX5_QP_ST_XRC);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
+ MLX5_SET(qpc, qpc, pd, to_mpd(devr->p0)->pdn);
- if (init_attr->create_flags &
- (IB_QP_CREATE_CROSS_CHANNEL |
- IB_QP_CREATE_MANAGED_SEND |
- IB_QP_CREATE_MANAGED_RECV)) {
- if (!MLX5_CAP_GEN(mdev, cd)) {
- mlx5_ib_dbg(dev, "cross-channel isn't supported\n");
- return -EINVAL;
- }
- if (init_attr->create_flags & IB_QP_CREATE_CROSS_CHANNEL)
- qp->flags |= MLX5_IB_QP_CROSS_CHANNEL;
- if (init_attr->create_flags & IB_QP_CREATE_MANAGED_SEND)
- qp->flags |= MLX5_IB_QP_MANAGED_SEND;
- if (init_attr->create_flags & IB_QP_CREATE_MANAGED_RECV)
- qp->flags |= MLX5_IB_QP_MANAGED_RECV;
- }
-
- if (init_attr->qp_type == IB_QPT_UD &&
- (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO))
- if (!MLX5_CAP_GEN(mdev, ipoib_basic_offloads)) {
- mlx5_ib_dbg(dev, "ipoib UD lso qp isn't supported\n");
- return -EOPNOTSUPP;
- }
+ if (qp->flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK)
+ MLX5_SET(qpc, qpc, block_lb_mc, 1);
+ if (qp->flags & IB_QP_CREATE_CROSS_CHANNEL)
+ MLX5_SET(qpc, qpc, cd_master, 1);
+ if (qp->flags & IB_QP_CREATE_MANAGED_SEND)
+ MLX5_SET(qpc, qpc, cd_slave_send, 1);
+ if (qp->flags & IB_QP_CREATE_MANAGED_RECV)
+ MLX5_SET(qpc, qpc, cd_slave_receive, 1);
- if (init_attr->create_flags & IB_QP_CREATE_SCATTER_FCS) {
- if (init_attr->qp_type != IB_QPT_RAW_PACKET) {
- mlx5_ib_dbg(dev, "Scatter FCS is supported only for Raw Packet QPs");
- return -EOPNOTSUPP;
- }
- if (!MLX5_CAP_GEN(dev->mdev, eth_net_offloads) ||
- !MLX5_CAP_ETH(dev->mdev, scatter_fcs)) {
- mlx5_ib_dbg(dev, "Scatter FCS isn't supported\n");
- return -EOPNOTSUPP;
- }
- qp->flags |= MLX5_IB_QP_CAP_SCATTER_FCS;
- }
+ MLX5_SET(qpc, qpc, rq_type, MLX5_SRQ_RQ);
+ MLX5_SET(qpc, qpc, no_sq, 1);
+ MLX5_SET(qpc, qpc, cqn_rcv, to_mcq(devr->c0)->mcq.cqn);
+ MLX5_SET(qpc, qpc, cqn_snd, to_mcq(devr->c0)->mcq.cqn);
+ MLX5_SET(qpc, qpc, srqn_rmpn_xrqn, to_msrq(devr->s0)->msrq.srqn);
+ MLX5_SET(qpc, qpc, xrcd, to_mxrcd(attr->xrcd)->xrcdn);
+ MLX5_SET64(qpc, qpc, dbr_addr, qp->db.dma);
- if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR)
- qp->sq_signal_bits = MLX5_WQE_CTRL_CQ_UPDATE;
+ /* 0xffffff means we ask to work with cqe version 0 */
+ if (MLX5_CAP_GEN(mdev, cqe_version) == MLX5_CQE_VERSION_V1)
+ MLX5_SET(qpc, qpc, user_index, uidx);
- if (init_attr->create_flags & IB_QP_CREATE_CVLAN_STRIPPING) {
- if (!(MLX5_CAP_GEN(dev->mdev, eth_net_offloads) &&
- MLX5_CAP_ETH(dev->mdev, vlan_cap)) ||
- (init_attr->qp_type != IB_QPT_RAW_PACKET))
- return -EOPNOTSUPP;
- qp->flags |= MLX5_IB_QP_CVLAN_STRIPPING;
+ if (qp->flags & IB_QP_CREATE_PCI_WRITE_END_PADDING) {
+ MLX5_SET(qpc, qpc, end_padding_mode,
+ MLX5_WQ_END_PAD_MODE_ALIGN);
+ /* Special case to clean flag */
+ qp->flags &= ~IB_QP_CREATE_PCI_WRITE_END_PADDING;
}
- if (udata) {
- if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) {
- mlx5_ib_dbg(dev, "copy failed\n");
- return -EFAULT;
- }
+ base = &qp->trans_qp.base;
+ err = mlx5_qpc_create_qp(dev, &base->mqp, in, inlen, out);
+ kvfree(in);
+ if (err)
+ return err;
- if (!check_flags_mask(ucmd.flags,
- MLX5_QP_FLAG_ALLOW_SCATTER_CQE |
- MLX5_QP_FLAG_BFREG_INDEX |
- MLX5_QP_FLAG_PACKET_BASED_CREDIT_MODE |
- MLX5_QP_FLAG_SCATTER_CQE |
- MLX5_QP_FLAG_SIGNATURE |
- MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC |
- MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC |
- MLX5_QP_FLAG_TUNNEL_OFFLOADS |
- MLX5_QP_FLAG_UAR_PAGE_INDEX |
- MLX5_QP_FLAG_TYPE_DCI |
- MLX5_QP_FLAG_TYPE_DCT))
- return -EINVAL;
+ base->container_mibqp = qp;
+ base->mqp.event = mlx5_ib_qp_event;
+ if (MLX5_CAP_GEN(mdev, ece_support))
+ params->resp.ece_options = MLX5_GET(create_qp_out, out, ece);
- err = get_qp_user_index(ucontext, &ucmd, udata->inlen, &uidx);
- if (err)
- return err;
+ spin_lock_irqsave(&dev->reset_flow_resource_lock, flags);
+ list_add_tail(&qp->qps_list, &dev->qp_list);
+ spin_unlock_irqrestore(&dev->reset_flow_resource_lock, flags);
- qp->wq_sig = !!(ucmd.flags & MLX5_QP_FLAG_SIGNATURE);
- if (MLX5_CAP_GEN(dev->mdev, sctr_data_cqe))
- qp->scat_cqe = !!(ucmd.flags & MLX5_QP_FLAG_SCATTER_CQE);
- if (ucmd.flags & MLX5_QP_FLAG_TUNNEL_OFFLOADS) {
- if (init_attr->qp_type != IB_QPT_RAW_PACKET ||
- !tunnel_offload_supported(mdev)) {
- mlx5_ib_dbg(dev, "Tunnel offload isn't supported\n");
- return -EOPNOTSUPP;
- }
- qp->flags_en |= MLX5_QP_FLAG_TUNNEL_OFFLOADS;
- }
+ qp->trans_qp.xrcdn = to_mxrcd(attr->xrcd)->xrcdn;
+ return 0;
+}
- if (ucmd.flags & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC) {
- if (init_attr->qp_type != IB_QPT_RAW_PACKET) {
- mlx5_ib_dbg(dev, "Self-LB UC isn't supported\n");
- return -EOPNOTSUPP;
- }
- qp->flags_en |= MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC;
- }
+static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
+{
+ struct ib_qp_init_attr *init_attr = params->attr;
+ struct mlx5_ib_create_qp *ucmd = params->ucmd;
+ u32 out[MLX5_ST_SZ_DW(create_qp_out)] = {};
+ struct ib_udata *udata = params->udata;
+ u32 uidx = params->uidx;
+ struct mlx5_ib_resources *devr = &dev->devr;
+ int inlen = MLX5_ST_SZ_BYTES(create_qp_in);
+ struct mlx5_core_dev *mdev = dev->mdev;
+ struct mlx5_ib_cq *send_cq;
+ struct mlx5_ib_cq *recv_cq;
+ unsigned long flags;
+ struct mlx5_ib_qp_base *base;
+ int mlx5_st;
+ void *qpc;
+ u32 *in;
+ int err;
- if (ucmd.flags & MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC) {
- if (init_attr->qp_type != IB_QPT_RAW_PACKET) {
- mlx5_ib_dbg(dev, "Self-LB UM isn't supported\n");
- return -EOPNOTSUPP;
- }
- qp->flags_en |= MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC;
- }
+ mutex_init(&qp->mutex);
+ spin_lock_init(&qp->sq.lock);
+ spin_lock_init(&qp->rq.lock);
- if (ucmd.flags & MLX5_QP_FLAG_PACKET_BASED_CREDIT_MODE) {
- if (init_attr->qp_type != IB_QPT_RC ||
- !MLX5_CAP_GEN(dev->mdev, qp_packet_based)) {
- mlx5_ib_dbg(dev, "packet based credit mode isn't supported\n");
- return -EOPNOTSUPP;
- }
- qp->flags |= MLX5_IB_QP_PACKET_BASED_CREDIT;
- }
+ mlx5_st = to_mlx5_st(qp->type);
+ if (mlx5_st < 0)
+ return -EINVAL;
- if (init_attr->create_flags & IB_QP_CREATE_SOURCE_QPN) {
- if (init_attr->qp_type != IB_QPT_UD ||
- (MLX5_CAP_GEN(dev->mdev, port_type) !=
- MLX5_CAP_PORT_TYPE_IB) ||
- !mlx5_get_flow_namespace(dev->mdev, MLX5_FLOW_NAMESPACE_BYPASS)) {
- mlx5_ib_dbg(dev, "Source QP option isn't supported\n");
- return -EOPNOTSUPP;
- }
+ if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR)
+ qp->sq_signal_bits = MLX5_WQE_CTRL_CQ_UPDATE;
- qp->flags |= MLX5_IB_QP_UNDERLAY;
- qp->underlay_qpn = init_attr->source_qpn;
- }
- } else {
- qp->wq_sig = !!wq_signature;
- }
+ if (qp->flags & IB_QP_CREATE_SOURCE_QPN)
+ qp->underlay_qpn = init_attr->source_qpn;
base = (init_attr->qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) ?
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) ?
&qp->raw_packet_qp.rq.base :
&qp->trans_qp.base;
qp->has_rq = qp_has_rq(init_attr);
- err = set_rq_size(dev, &init_attr->cap, qp->has_rq,
- qp, udata ? &ucmd : NULL);
+ err = set_rq_size(dev, &init_attr->cap, qp->has_rq, qp, ucmd);
if (err) {
mlx5_ib_dbg(dev, "err %d\n", err);
return err;
}
- if (pd) {
- if (udata) {
- __u32 max_wqes =
- 1 << MLX5_CAP_GEN(mdev, log_max_qp_sz);
- mlx5_ib_dbg(dev, "requested sq_wqe_count (%d)\n", ucmd.sq_wqe_count);
- if (ucmd.rq_wqe_shift != qp->rq.wqe_shift ||
- ucmd.rq_wqe_count != qp->rq.wqe_cnt) {
- mlx5_ib_dbg(dev, "invalid rq params\n");
- return -EINVAL;
- }
- if (ucmd.sq_wqe_count > max_wqes) {
- mlx5_ib_dbg(dev, "requested sq_wqe_count (%d) > max allowed (%d)\n",
- ucmd.sq_wqe_count, max_wqes);
- return -EINVAL;
- }
- if (init_attr->create_flags &
- MLX5_IB_QP_CREATE_SQPN_QP1) {
- mlx5_ib_dbg(dev, "user-space is not allowed to create UD QPs spoofing as QP1\n");
- return -EINVAL;
- }
- err = create_user_qp(dev, pd, qp, udata, init_attr, &in,
- &resp, &inlen, base);
- if (err)
- mlx5_ib_dbg(dev, "err %d\n", err);
- } else {
- err = create_kernel_qp(dev, init_attr, qp, &in, &inlen,
- base);
- if (err)
- mlx5_ib_dbg(dev, "err %d\n", err);
- }
+ if (ucmd->rq_wqe_shift != qp->rq.wqe_shift ||
+ ucmd->rq_wqe_count != qp->rq.wqe_cnt)
+ return -EINVAL;
- if (err)
- return err;
- } else {
- in = kvzalloc(inlen, GFP_KERNEL);
- if (!in)
- return -ENOMEM;
+ if (ucmd->sq_wqe_count > (1 << MLX5_CAP_GEN(mdev, log_max_qp_sz)))
+ return -EINVAL;
- qp->create_type = MLX5_QP_EMPTY;
- }
+ err = _create_user_qp(dev, pd, qp, udata, init_attr, &in, &params->resp,
+ &inlen, base, ucmd);
+ if (err)
+ return err;
if (is_sqp(init_attr->qp_type))
qp->port = init_attr->port_num;
+ if (MLX5_CAP_GEN(mdev, ece_support))
+ MLX5_SET(create_qp_in, in, ece, ucmd->ece_options);
qpc = MLX5_ADDR_OF(create_qp_in, in, qpc);
MLX5_SET(qpc, qpc, st, mlx5_st);
MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
+ MLX5_SET(qpc, qpc, pd, to_mpd(pd)->pdn);
- if (init_attr->qp_type != MLX5_IB_QPT_REG_UMR)
- MLX5_SET(qpc, qpc, pd, to_mpd(pd ? pd : devr->p0)->pdn);
- else
- MLX5_SET(qpc, qpc, latency_sensitive, 1);
-
-
- if (qp->wq_sig)
+ if (qp->flags_en & MLX5_QP_FLAG_SIGNATURE)
MLX5_SET(qpc, qpc, wq_signature, 1);
- if (qp->flags & MLX5_IB_QP_BLOCK_MULTICAST_LOOPBACK)
+ if (qp->flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK)
MLX5_SET(qpc, qpc, block_lb_mc, 1);
- if (qp->flags & MLX5_IB_QP_CROSS_CHANNEL)
+ if (qp->flags & IB_QP_CREATE_CROSS_CHANNEL)
MLX5_SET(qpc, qpc, cd_master, 1);
- if (qp->flags & MLX5_IB_QP_MANAGED_SEND)
+ if (qp->flags & IB_QP_CREATE_MANAGED_SEND)
MLX5_SET(qpc, qpc, cd_slave_send, 1);
- if (qp->flags & MLX5_IB_QP_MANAGED_RECV)
+ if (qp->flags & IB_QP_CREATE_MANAGED_RECV)
MLX5_SET(qpc, qpc, cd_slave_receive, 1);
- if (qp->flags & MLX5_IB_QP_PACKET_BASED_CREDIT)
+ if (qp->flags_en & MLX5_QP_FLAG_PACKET_BASED_CREDIT_MODE)
MLX5_SET(qpc, qpc, req_e2e_credit_mode, 1);
- if (qp->scat_cqe && is_connected(init_attr->qp_type)) {
- configure_responder_scat_cqe(init_attr, qpc);
- configure_requester_scat_cqe(dev, init_attr,
- udata ? &ucmd : NULL,
- qpc);
+ if ((qp->flags_en & MLX5_QP_FLAG_SCATTER_CQE) &&
+ (init_attr->qp_type == IB_QPT_RC ||
+ init_attr->qp_type == IB_QPT_UC)) {
+ int rcqe_sz = mlx5_ib_get_cqe_size(init_attr->recv_cq);
+
+ MLX5_SET(qpc, qpc, cs_res,
+ rcqe_sz == 128 ? MLX5_RES_SCAT_DATA64_CQE :
+ MLX5_RES_SCAT_DATA32_CQE);
}
+ if ((qp->flags_en & MLX5_QP_FLAG_SCATTER_CQE) &&
+ (qp->type == MLX5_IB_QPT_DCI || qp->type == IB_QPT_RC))
+ configure_requester_scat_cqe(dev, init_attr, ucmd, qpc);
if (qp->rq.wqe_cnt) {
MLX5_SET(qpc, qpc, log_rq_stride, qp->rq.wqe_shift - 4);
@@ -2281,12 +2033,6 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
/* Set default resources */
switch (init_attr->qp_type) {
- case IB_QPT_XRC_TGT:
- MLX5_SET(qpc, qpc, cqn_rcv, to_mcq(devr->c0)->mcq.cqn);
- MLX5_SET(qpc, qpc, cqn_snd, to_mcq(devr->c0)->mcq.cqn);
- MLX5_SET(qpc, qpc, srqn_rmpn_xrqn, to_msrq(devr->s0)->msrq.srqn);
- MLX5_SET(qpc, qpc, xrcd, to_mxrcd(init_attr->xrcd)->xrcdn);
- break;
case IB_QPT_XRC_INI:
MLX5_SET(qpc, qpc, cqn_rcv, to_mcq(devr->c0)->mcq.cqn);
MLX5_SET(qpc, qpc, xrcd, to_mxrcd(devr->x1)->xrcdn);
@@ -2314,52 +2060,163 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
if (MLX5_CAP_GEN(mdev, cqe_version) == MLX5_CQE_VERSION_V1)
MLX5_SET(qpc, qpc, user_index, uidx);
- /* we use IB_QP_CREATE_IPOIB_UD_LSO to indicates ipoib qp */
- if (init_attr->qp_type == IB_QPT_UD &&
- (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO)) {
- MLX5_SET(qpc, qpc, ulp_stateless_offload_mode, 1);
- qp->flags |= MLX5_IB_QP_LSO;
+ if (qp->flags & IB_QP_CREATE_PCI_WRITE_END_PADDING &&
+ init_attr->qp_type != IB_QPT_RAW_PACKET) {
+ MLX5_SET(qpc, qpc, end_padding_mode,
+ MLX5_WQ_END_PAD_MODE_ALIGN);
+ /* Special case to clean flag */
+ qp->flags &= ~IB_QP_CREATE_PCI_WRITE_END_PADDING;
}
- if (init_attr->create_flags & IB_QP_CREATE_PCI_WRITE_END_PADDING) {
- if (!MLX5_CAP_GEN(dev->mdev, end_pad)) {
- mlx5_ib_dbg(dev, "scatter end padding is not supported\n");
- err = -EOPNOTSUPP;
- goto err;
- } else if (init_attr->qp_type != IB_QPT_RAW_PACKET) {
- MLX5_SET(qpc, qpc, end_padding_mode,
- MLX5_WQ_END_PAD_MODE_ALIGN);
- } else {
- qp->flags |= MLX5_IB_QP_PCI_WRITE_END_PADDING;
- }
+ if (init_attr->qp_type == IB_QPT_RAW_PACKET ||
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) {
+ qp->raw_packet_qp.sq.ubuffer.buf_addr = ucmd->sq_buf_addr;
+ raw_packet_qp_copy_info(qp, &qp->raw_packet_qp);
+ err = create_raw_packet_qp(dev, qp, in, inlen, pd, udata,
+ &params->resp);
+ } else
+ err = mlx5_qpc_create_qp(dev, &base->mqp, in, inlen, out);
+
+ kvfree(in);
+ if (err)
+ goto err_create;
+
+ base->container_mibqp = qp;
+ base->mqp.event = mlx5_ib_qp_event;
+ if (MLX5_CAP_GEN(mdev, ece_support))
+ params->resp.ece_options = MLX5_GET(create_qp_out, out, ece);
+
+ get_cqs(qp->type, init_attr->send_cq, init_attr->recv_cq,
+ &send_cq, &recv_cq);
+ spin_lock_irqsave(&dev->reset_flow_resource_lock, flags);
+ mlx5_ib_lock_cqs(send_cq, recv_cq);
+ /* Maintain device to QPs access, needed for further handling via reset
+ * flow
+ */
+ list_add_tail(&qp->qps_list, &dev->qp_list);
+ /* Maintain CQ to QPs access, needed for further handling via reset flow
+ */
+ if (send_cq)
+ list_add_tail(&qp->cq_send_list, &send_cq->list_send_qp);
+ if (recv_cq)
+ list_add_tail(&qp->cq_recv_list, &recv_cq->list_recv_qp);
+ mlx5_ib_unlock_cqs(send_cq, recv_cq);
+ spin_unlock_irqrestore(&dev->reset_flow_resource_lock, flags);
+
+ return 0;
+
+err_create:
+ destroy_qp(dev, qp, base, udata);
+ return err;
+}
+
+static int create_kernel_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
+{
+ struct ib_qp_init_attr *attr = params->attr;
+ u32 uidx = params->uidx;
+ struct mlx5_ib_resources *devr = &dev->devr;
+ u32 out[MLX5_ST_SZ_DW(create_qp_out)] = {};
+ int inlen = MLX5_ST_SZ_BYTES(create_qp_in);
+ struct mlx5_core_dev *mdev = dev->mdev;
+ struct mlx5_ib_cq *send_cq;
+ struct mlx5_ib_cq *recv_cq;
+ unsigned long flags;
+ struct mlx5_ib_qp_base *base;
+ int mlx5_st;
+ void *qpc;
+ u32 *in;
+ int err;
+
+ mutex_init(&qp->mutex);
+ spin_lock_init(&qp->sq.lock);
+ spin_lock_init(&qp->rq.lock);
+
+ mlx5_st = to_mlx5_st(qp->type);
+ if (mlx5_st < 0)
+ return -EINVAL;
+
+ if (attr->sq_sig_type == IB_SIGNAL_ALL_WR)
+ qp->sq_signal_bits = MLX5_WQE_CTRL_CQ_UPDATE;
+
+ base = &qp->trans_qp.base;
+
+ qp->has_rq = qp_has_rq(attr);
+ err = set_rq_size(dev, &attr->cap, qp->has_rq, qp, NULL);
+ if (err) {
+ mlx5_ib_dbg(dev, "err %d\n", err);
+ return err;
}
- if (inlen < 0) {
- err = -EINVAL;
- goto err;
+ err = _create_kernel_qp(dev, attr, qp, &in, &inlen, base);
+ if (err)
+ return err;
+
+ if (is_sqp(attr->qp_type))
+ qp->port = attr->port_num;
+
+ qpc = MLX5_ADDR_OF(create_qp_in, in, qpc);
+
+ MLX5_SET(qpc, qpc, st, mlx5_st);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
+
+ if (attr->qp_type != MLX5_IB_QPT_REG_UMR)
+ MLX5_SET(qpc, qpc, pd, to_mpd(pd ? pd : devr->p0)->pdn);
+ else
+ MLX5_SET(qpc, qpc, latency_sensitive, 1);
+
+
+ if (qp->flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK)
+ MLX5_SET(qpc, qpc, block_lb_mc, 1);
+
+ if (qp->rq.wqe_cnt) {
+ MLX5_SET(qpc, qpc, log_rq_stride, qp->rq.wqe_shift - 4);
+ MLX5_SET(qpc, qpc, log_rq_size, ilog2(qp->rq.wqe_cnt));
}
- if (init_attr->qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) {
- qp->raw_packet_qp.sq.ubuffer.buf_addr = ucmd.sq_buf_addr;
- raw_packet_qp_copy_info(qp, &qp->raw_packet_qp);
- err = create_raw_packet_qp(dev, qp, in, inlen, pd, udata,
- &resp);
+ MLX5_SET(qpc, qpc, rq_type, get_rx_type(qp, attr));
+
+ if (qp->sq.wqe_cnt)
+ MLX5_SET(qpc, qpc, log_sq_size, ilog2(qp->sq.wqe_cnt));
+ else
+ MLX5_SET(qpc, qpc, no_sq, 1);
+
+ if (attr->srq) {
+ MLX5_SET(qpc, qpc, xrcd, to_mxrcd(devr->x0)->xrcdn);
+ MLX5_SET(qpc, qpc, srqn_rmpn_xrqn,
+ to_msrq(attr->srq)->msrq.srqn);
} else {
- err = mlx5_core_create_qp(dev, &base->mqp, in, inlen);
+ MLX5_SET(qpc, qpc, xrcd, to_mxrcd(devr->x1)->xrcdn);
+ MLX5_SET(qpc, qpc, srqn_rmpn_xrqn,
+ to_msrq(devr->s1)->msrq.srqn);
}
- if (err) {
- mlx5_ib_dbg(dev, "create qp failed\n");
- goto err_create;
- }
+ if (attr->send_cq)
+ MLX5_SET(qpc, qpc, cqn_snd, to_mcq(attr->send_cq)->mcq.cqn);
+
+ if (attr->recv_cq)
+ MLX5_SET(qpc, qpc, cqn_rcv, to_mcq(attr->recv_cq)->mcq.cqn);
+
+ MLX5_SET64(qpc, qpc, dbr_addr, qp->db.dma);
+
+ /* 0xffffff means we ask to work with cqe version 0 */
+ if (MLX5_CAP_GEN(mdev, cqe_version) == MLX5_CQE_VERSION_V1)
+ MLX5_SET(qpc, qpc, user_index, uidx);
+
+ /* we use IB_QP_CREATE_IPOIB_UD_LSO to indicates ipoib qp */
+ if (qp->flags & IB_QP_CREATE_IPOIB_UD_LSO)
+ MLX5_SET(qpc, qpc, ulp_stateless_offload_mode, 1);
+ err = mlx5_qpc_create_qp(dev, &base->mqp, in, inlen, out);
kvfree(in);
+ if (err)
+ goto err_create;
base->container_mibqp = qp;
base->mqp.event = mlx5_ib_qp_event;
- get_cqs(init_attr->qp_type, init_attr->send_cq, init_attr->recv_cq,
+ get_cqs(qp->type, attr->send_cq, attr->recv_cq,
&send_cq, &recv_cq);
spin_lock_irqsave(&dev->reset_flow_resource_lock, flags);
mlx5_ib_lock_cqs(send_cq, recv_cq);
@@ -2379,13 +2236,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
return 0;
err_create:
- if (qp->create_type == MLX5_QP_USER)
- destroy_qp_user(dev, pd, qp, base, udata);
- else if (qp->create_type == MLX5_QP_KERNEL)
- destroy_qp_kernel(dev, qp);
-
-err:
- kvfree(in);
+ destroy_qp(dev, qp, base, NULL);
return err;
}
@@ -2447,11 +2298,6 @@ static void mlx5_ib_unlock_cqs(struct mlx5_ib_cq *send_cq, struct mlx5_ib_cq *re
}
}
-static struct mlx5_ib_pd *get_pd(struct mlx5_ib_qp *qp)
-{
- return to_mpd(qp->ibqp.pd);
-}
-
static void get_cqs(enum ib_qp_type qp_type,
struct ib_cq *ib_send_cq, struct ib_cq *ib_recv_cq,
struct mlx5_ib_cq **send_cq, struct mlx5_ib_cq **recv_cq)
@@ -2472,14 +2318,10 @@ static void get_cqs(enum ib_qp_type qp_type,
case IB_QPT_RC:
case IB_QPT_UC:
case IB_QPT_UD:
- case IB_QPT_RAW_IPV6:
- case IB_QPT_RAW_ETHERTYPE:
case IB_QPT_RAW_PACKET:
*send_cq = ib_send_cq ? to_mcq(ib_send_cq) : NULL;
*recv_cq = ib_recv_cq ? to_mcq(ib_recv_cq) : NULL;
break;
-
- case IB_QPT_MAX:
default:
*send_cq = NULL;
*recv_cq = NULL;
@@ -2505,15 +2347,15 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
}
base = (qp->ibqp.qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) ?
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) ?
&qp->raw_packet_qp.rq.base :
&qp->trans_qp.base;
if (qp->state != IB_QPS_RESET) {
if (qp->ibqp.qp_type != IB_QPT_RAW_PACKET &&
- !(qp->flags & MLX5_IB_QP_UNDERLAY)) {
+ !(qp->flags & IB_QP_CREATE_SOURCE_QPN)) {
err = mlx5_core_qp_modify(dev, MLX5_CMD_OP_2RST_QP, 0,
- NULL, &base->mqp);
+ NULL, &base->mqp, NULL);
} else {
struct mlx5_modify_raw_qp_param raw_qp_param = {
.operation = MLX5_CMD_OP_2RST_QP
@@ -2539,7 +2381,7 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
if (recv_cq)
list_del(&qp->cq_recv_list);
- if (qp->create_type == MLX5_QP_KERNEL) {
+ if (!udata) {
__mlx5_ib_cq_clean(recv_cq, base->mqp.qpn,
qp->ibqp.srq ? to_msrq(qp->ibqp.srq) : NULL);
if (send_cq != recv_cq)
@@ -2550,7 +2392,7 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
spin_unlock_irqrestore(&dev->reset_flow_resource_lock, flags);
if (qp->ibqp.qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) {
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) {
destroy_raw_packet_qp(dev, qp);
} else {
err = mlx5_core_destroy_qp(dev, &base->mqp);
@@ -2559,254 +2401,454 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
base->mqp.qpn);
}
- if (qp->create_type == MLX5_QP_KERNEL)
- destroy_qp_kernel(dev, qp);
- else if (qp->create_type == MLX5_QP_USER)
- destroy_qp_user(dev, &get_pd(qp)->ibpd, qp, base, udata);
+ destroy_qp(dev, qp, base, udata);
}
-static const char *ib_qp_type_str(enum ib_qp_type type)
+static int create_dct(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
{
- switch (type) {
- case IB_QPT_SMI:
- return "IB_QPT_SMI";
- case IB_QPT_GSI:
- return "IB_QPT_GSI";
+ struct ib_qp_init_attr *attr = params->attr;
+ struct mlx5_ib_create_qp *ucmd = params->ucmd;
+ u32 uidx = params->uidx;
+ void *dctc;
+
+ qp->dct.in = kzalloc(MLX5_ST_SZ_BYTES(create_dct_in), GFP_KERNEL);
+ if (!qp->dct.in)
+ return -ENOMEM;
+
+ MLX5_SET(create_dct_in, qp->dct.in, uid, to_mpd(pd)->uid);
+ dctc = MLX5_ADDR_OF(create_dct_in, qp->dct.in, dct_context_entry);
+ MLX5_SET(dctc, dctc, pd, to_mpd(pd)->pdn);
+ MLX5_SET(dctc, dctc, srqn_xrqn, to_msrq(attr->srq)->msrq.srqn);
+ MLX5_SET(dctc, dctc, cqn, to_mcq(attr->recv_cq)->mcq.cqn);
+ MLX5_SET64(dctc, dctc, dc_access_key, ucmd->access_key);
+ MLX5_SET(dctc, dctc, user_index, uidx);
+ if (MLX5_CAP_GEN(dev->mdev, ece_support))
+ MLX5_SET(dctc, dctc, ece, ucmd->ece_options);
+
+ if (qp->flags_en & MLX5_QP_FLAG_SCATTER_CQE) {
+ int rcqe_sz = mlx5_ib_get_cqe_size(attr->recv_cq);
+
+ if (rcqe_sz == 128)
+ MLX5_SET(dctc, dctc, cs_res, MLX5_RES_SCAT_DATA64_CQE);
+ }
+
+ qp->state = IB_QPS_RESET;
+
+ return 0;
+}
+
+static int check_qp_type(struct mlx5_ib_dev *dev, struct ib_qp_init_attr *attr,
+ enum ib_qp_type *type)
+{
+ if (attr->qp_type == IB_QPT_DRIVER && !MLX5_CAP_GEN(dev->mdev, dct))
+ goto out;
+
+ switch (attr->qp_type) {
+ case IB_QPT_XRC_TGT:
+ case IB_QPT_XRC_INI:
+ if (!MLX5_CAP_GEN(dev->mdev, xrc))
+ goto out;
+ fallthrough;
case IB_QPT_RC:
- return "IB_QPT_RC";
case IB_QPT_UC:
- return "IB_QPT_UC";
- case IB_QPT_UD:
- return "IB_QPT_UD";
- case IB_QPT_RAW_IPV6:
- return "IB_QPT_RAW_IPV6";
- case IB_QPT_RAW_ETHERTYPE:
- return "IB_QPT_RAW_ETHERTYPE";
- case IB_QPT_XRC_INI:
- return "IB_QPT_XRC_INI";
- case IB_QPT_XRC_TGT:
- return "IB_QPT_XRC_TGT";
+ case IB_QPT_SMI:
+ case MLX5_IB_QPT_HW_GSI:
+ case IB_QPT_DRIVER:
+ case IB_QPT_GSI:
+ if (dev->profile == &raw_eth_profile)
+ goto out;
case IB_QPT_RAW_PACKET:
- return "IB_QPT_RAW_PACKET";
+ case IB_QPT_UD:
case MLX5_IB_QPT_REG_UMR:
- return "MLX5_IB_QPT_REG_UMR";
- case IB_QPT_DRIVER:
- return "IB_QPT_DRIVER";
- case IB_QPT_MAX:
+ break;
default:
- return "Invalid QP type";
+ goto out;
}
+
+ *type = attr->qp_type;
+ return 0;
+
+out:
+ mlx5_ib_dbg(dev, "Unsupported QP type %d\n", attr->qp_type);
+ return -EOPNOTSUPP;
}
-static struct ib_qp *mlx5_ib_create_dct(struct ib_pd *pd,
- struct ib_qp_init_attr *attr,
- struct mlx5_ib_create_qp *ucmd,
- struct ib_udata *udata)
+static int check_valid_flow(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct ib_qp_init_attr *attr,
+ struct ib_udata *udata)
{
struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
udata, struct mlx5_ib_ucontext, ibucontext);
- struct mlx5_ib_qp *qp;
- int err = 0;
- u32 uidx = MLX5_IB_DEFAULT_UIDX;
- void *dctc;
- if (!attr->srq || !attr->recv_cq)
- return ERR_PTR(-EINVAL);
+ if (!udata) {
+ /* Kernel create_qp callers */
+ if (attr->rwq_ind_tbl)
+ return -EOPNOTSUPP;
- err = get_qp_user_index(ucontext, ucmd, sizeof(*ucmd), &uidx);
- if (err)
- return ERR_PTR(err);
+ switch (attr->qp_type) {
+ case IB_QPT_RAW_PACKET:
+ case IB_QPT_DRIVER:
+ return -EOPNOTSUPP;
+ default:
+ return 0;
+ }
+ }
- qp = kzalloc(sizeof(*qp), GFP_KERNEL);
- if (!qp)
- return ERR_PTR(-ENOMEM);
+ /* Userspace create_qp callers */
+ if (attr->qp_type == IB_QPT_RAW_PACKET && !ucontext->cqe_version) {
+ mlx5_ib_dbg(dev,
+ "Raw Packet QP is only supported for CQE version > 0\n");
+ return -EINVAL;
+ }
- qp->dct.in = kzalloc(MLX5_ST_SZ_BYTES(create_dct_in), GFP_KERNEL);
- if (!qp->dct.in) {
- err = -ENOMEM;
- goto err_free;
+ if (attr->qp_type != IB_QPT_RAW_PACKET && attr->rwq_ind_tbl) {
+ mlx5_ib_dbg(dev,
+ "Wrong QP type %d for the RWQ indirect table\n",
+ attr->qp_type);
+ return -EINVAL;
}
- MLX5_SET(create_dct_in, qp->dct.in, uid, to_mpd(pd)->uid);
- dctc = MLX5_ADDR_OF(create_dct_in, qp->dct.in, dct_context_entry);
- qp->qp_sub_type = MLX5_IB_QPT_DCT;
- MLX5_SET(dctc, dctc, pd, to_mpd(pd)->pdn);
- MLX5_SET(dctc, dctc, srqn_xrqn, to_msrq(attr->srq)->msrq.srqn);
- MLX5_SET(dctc, dctc, cqn, to_mcq(attr->recv_cq)->mcq.cqn);
- MLX5_SET64(dctc, dctc, dc_access_key, ucmd->access_key);
- MLX5_SET(dctc, dctc, user_index, uidx);
+ switch (attr->qp_type) {
+ case IB_QPT_SMI:
+ case MLX5_IB_QPT_HW_GSI:
+ case MLX5_IB_QPT_REG_UMR:
+ case IB_QPT_GSI:
+ mlx5_ib_dbg(dev, "Kernel doesn't support QP type %d\n",
+ attr->qp_type);
+ return -EINVAL;
+ default:
+ break;
+ }
- if (ucmd->flags & MLX5_QP_FLAG_SCATTER_CQE)
- configure_responder_scat_cqe(attr, dctc);
+ /*
+ * We don't need to see this warning, it means that kernel code
+ * missing ib_pd. Placed here to catch developer's mistakes.
+ */
+ WARN_ONCE(!pd && attr->qp_type != IB_QPT_XRC_TGT,
+ "There is a missing PD pointer assignment\n");
+ return 0;
+}
- qp->state = IB_QPS_RESET;
+static void process_vendor_flag(struct mlx5_ib_dev *dev, int *flags, int flag,
+ bool cond, struct mlx5_ib_qp *qp)
+{
+ if (!(*flags & flag))
+ return;
- return &qp->ibqp;
-err_free:
- kfree(qp);
- return ERR_PTR(err);
+ if (cond) {
+ qp->flags_en |= flag;
+ *flags &= ~flag;
+ return;
+ }
+
+ if (flag == MLX5_QP_FLAG_SCATTER_CQE) {
+ /*
+ * We don't return error if this flag was provided,
+ * and mlx5 doesn't have right capability.
+ */
+ *flags &= ~MLX5_QP_FLAG_SCATTER_CQE;
+ return;
+ }
+ mlx5_ib_dbg(dev, "Vendor create QP flag 0x%X is not supported\n", flag);
}
-static int set_mlx_qp_type(struct mlx5_ib_dev *dev,
- struct ib_qp_init_attr *init_attr,
- struct mlx5_ib_create_qp *ucmd,
- struct ib_udata *udata)
+static int process_vendor_flags(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ void *ucmd, struct ib_qp_init_attr *attr)
{
- enum { MLX_QP_FLAGS = MLX5_QP_FLAG_TYPE_DCT | MLX5_QP_FLAG_TYPE_DCI };
- int err;
+ struct mlx5_core_dev *mdev = dev->mdev;
+ bool cond;
+ int flags;
- if (!udata)
+ if (attr->rwq_ind_tbl)
+ flags = ((struct mlx5_ib_create_qp_rss *)ucmd)->flags;
+ else
+ flags = ((struct mlx5_ib_create_qp *)ucmd)->flags;
+
+ switch (flags & (MLX5_QP_FLAG_TYPE_DCT | MLX5_QP_FLAG_TYPE_DCI)) {
+ case MLX5_QP_FLAG_TYPE_DCI:
+ qp->type = MLX5_IB_QPT_DCI;
+ break;
+ case MLX5_QP_FLAG_TYPE_DCT:
+ qp->type = MLX5_IB_QPT_DCT;
+ break;
+ default:
+ if (qp->type != IB_QPT_DRIVER)
+ break;
+ /*
+ * It is IB_QPT_DRIVER and or no subtype or
+ * wrong subtype were provided.
+ */
return -EINVAL;
+ }
- if (udata->inlen < sizeof(*ucmd)) {
- mlx5_ib_dbg(dev, "create_qp user command is smaller than expected\n");
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_TYPE_DCI, true, qp);
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_TYPE_DCT, true, qp);
+
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_SIGNATURE, true, qp);
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_SCATTER_CQE,
+ MLX5_CAP_GEN(mdev, sctr_data_cqe), qp);
+
+ if (qp->type == IB_QPT_RAW_PACKET) {
+ cond = MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan) ||
+ MLX5_CAP_ETH(mdev, tunnel_stateless_gre) ||
+ MLX5_CAP_ETH(mdev, tunnel_stateless_geneve_rx);
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_TUNNEL_OFFLOADS,
+ cond, qp);
+ process_vendor_flag(dev, &flags,
+ MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC, true,
+ qp);
+ process_vendor_flag(dev, &flags,
+ MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC, true,
+ qp);
+ }
+
+ if (qp->type == IB_QPT_RC)
+ process_vendor_flag(dev, &flags,
+ MLX5_QP_FLAG_PACKET_BASED_CREDIT_MODE,
+ MLX5_CAP_GEN(mdev, qp_packet_based), qp);
+
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_BFREG_INDEX, true, qp);
+ process_vendor_flag(dev, &flags, MLX5_QP_FLAG_UAR_PAGE_INDEX, true, qp);
+
+ cond = qp->flags_en & ~(MLX5_QP_FLAG_TUNNEL_OFFLOADS |
+ MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_UC |
+ MLX5_QP_FLAG_TIR_ALLOW_SELF_LB_MC);
+ if (attr->rwq_ind_tbl && cond) {
+ mlx5_ib_dbg(dev, "RSS RAW QP has unsupported flags 0x%X\n",
+ cond);
return -EINVAL;
}
- err = ib_copy_from_udata(ucmd, udata, sizeof(*ucmd));
- if (err)
- return err;
- if ((ucmd->flags & MLX_QP_FLAGS) == MLX5_QP_FLAG_TYPE_DCI) {
- init_attr->qp_type = MLX5_IB_QPT_DCI;
- } else {
- if ((ucmd->flags & MLX_QP_FLAGS) == MLX5_QP_FLAG_TYPE_DCT) {
- init_attr->qp_type = MLX5_IB_QPT_DCT;
- } else {
- mlx5_ib_dbg(dev, "Invalid QP flags\n");
- return -EINVAL;
- }
+ if (flags)
+ mlx5_ib_dbg(dev, "udata has unsupported flags 0x%X\n", flags);
+
+ return (flags) ? -EINVAL : 0;
}
- if (!MLX5_CAP_GEN(dev->mdev, dct)) {
- mlx5_ib_dbg(dev, "DC transport is not supported\n");
- return -EOPNOTSUPP;
+static void process_create_flag(struct mlx5_ib_dev *dev, int *flags, int flag,
+ bool cond, struct mlx5_ib_qp *qp)
+{
+ if (!(*flags & flag))
+ return;
+
+ if (cond) {
+ qp->flags |= flag;
+ *flags &= ~flag;
+ return;
}
- return 0;
+ if (flag == MLX5_IB_QP_CREATE_WC_TEST) {
+ /*
+ * Special case, if condition didn't meet, it won't be error,
+ * just different in-kernel flow.
+ */
+ *flags &= ~MLX5_IB_QP_CREATE_WC_TEST;
+ return;
+ }
+ mlx5_ib_dbg(dev, "Verbs create QP flag 0x%X is not supported\n", flag);
}
-struct ib_qp *mlx5_ib_create_qp(struct ib_pd *pd,
- struct ib_qp_init_attr *verbs_init_attr,
- struct ib_udata *udata)
+static int process_create_flags(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ struct ib_qp_init_attr *attr)
{
- struct mlx5_ib_dev *dev;
- struct mlx5_ib_qp *qp;
- u16 xrcdn = 0;
- int err;
- struct ib_qp_init_attr mlx_init_attr;
- struct ib_qp_init_attr *init_attr = verbs_init_attr;
- struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
- udata, struct mlx5_ib_ucontext, ibucontext);
+ enum ib_qp_type qp_type = qp->type;
+ struct mlx5_core_dev *mdev = dev->mdev;
+ int create_flags = attr->create_flags;
+ bool cond;
- if (pd) {
- dev = to_mdev(pd->device);
+ if (qp->type == IB_QPT_UD && dev->profile == &raw_eth_profile)
+ if (create_flags & ~MLX5_IB_QP_CREATE_WC_TEST)
+ return -EINVAL;
- if (init_attr->qp_type == IB_QPT_RAW_PACKET) {
- if (!ucontext) {
- mlx5_ib_dbg(dev, "Raw Packet QP is not supported for kernel consumers\n");
- return ERR_PTR(-EINVAL);
- } else if (!ucontext->cqe_version) {
- mlx5_ib_dbg(dev, "Raw Packet QP is only supported for CQE version > 0\n");
- return ERR_PTR(-EINVAL);
- }
- }
- } else {
- /* being cautious here */
- if (init_attr->qp_type != IB_QPT_XRC_TGT &&
- init_attr->qp_type != MLX5_IB_QPT_REG_UMR) {
- pr_warn("%s: no PD for transport %s\n", __func__,
- ib_qp_type_str(init_attr->qp_type));
- return ERR_PTR(-EINVAL);
- }
- dev = to_mdev(to_mxrcd(init_attr->xrcd)->ibxrcd.device);
+ if (qp_type == MLX5_IB_QPT_DCT)
+ return (create_flags) ? -EINVAL : 0;
+
+ if (qp_type == IB_QPT_RAW_PACKET && attr->rwq_ind_tbl)
+ return (create_flags) ? -EINVAL : 0;
+
+ process_create_flag(dev, &create_flags,
+ IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK,
+ MLX5_CAP_GEN(mdev, block_lb_mc), qp);
+ process_create_flag(dev, &create_flags, IB_QP_CREATE_CROSS_CHANNEL,
+ MLX5_CAP_GEN(mdev, cd), qp);
+ process_create_flag(dev, &create_flags, IB_QP_CREATE_MANAGED_SEND,
+ MLX5_CAP_GEN(mdev, cd), qp);
+ process_create_flag(dev, &create_flags, IB_QP_CREATE_MANAGED_RECV,
+ MLX5_CAP_GEN(mdev, cd), qp);
+
+ if (qp_type == IB_QPT_UD) {
+ process_create_flag(dev, &create_flags,
+ IB_QP_CREATE_IPOIB_UD_LSO,
+ MLX5_CAP_GEN(mdev, ipoib_basic_offloads),
+ qp);
+ cond = MLX5_CAP_GEN(mdev, port_type) == MLX5_CAP_PORT_TYPE_IB;
+ process_create_flag(dev, &create_flags, IB_QP_CREATE_SOURCE_QPN,
+ cond, qp);
+ }
+
+ if (qp_type == IB_QPT_RAW_PACKET) {
+ cond = MLX5_CAP_GEN(mdev, eth_net_offloads) &&
+ MLX5_CAP_ETH(mdev, scatter_fcs);
+ process_create_flag(dev, &create_flags,
+ IB_QP_CREATE_SCATTER_FCS, cond, qp);
+
+ cond = MLX5_CAP_GEN(mdev, eth_net_offloads) &&
+ MLX5_CAP_ETH(mdev, vlan_cap);
+ process_create_flag(dev, &create_flags,
+ IB_QP_CREATE_CVLAN_STRIPPING, cond, qp);
+ }
+
+ process_create_flag(dev, &create_flags,
+ IB_QP_CREATE_PCI_WRITE_END_PADDING,
+ MLX5_CAP_GEN(mdev, end_pad), qp);
+
+ process_create_flag(dev, &create_flags, MLX5_IB_QP_CREATE_WC_TEST,
+ qp_type != MLX5_IB_QPT_REG_UMR, qp);
+ process_create_flag(dev, &create_flags, MLX5_IB_QP_CREATE_SQPN_QP1,
+ true, qp);
+
+ if (create_flags)
+ mlx5_ib_dbg(dev, "Create QP has unsupported flags 0x%X\n",
+ create_flags);
+
+ return (create_flags) ? -EINVAL : 0;
+}
+
+static int process_udata_size(struct mlx5_ib_dev *dev,
+ struct mlx5_create_qp_params *params)
+{
+ size_t ucmd = sizeof(struct mlx5_ib_create_qp);
+ struct ib_udata *udata = params->udata;
+ size_t outlen = udata->outlen;
+ size_t inlen = udata->inlen;
+
+ params->outlen = min(outlen, sizeof(struct mlx5_ib_create_qp_resp));
+ params->ucmd_size = ucmd;
+ if (!params->is_rss_raw) {
+ /* User has old rdma-core, which doesn't support ECE */
+ size_t min_inlen =
+ offsetof(struct mlx5_ib_create_qp, ece_options);
+
+ /*
+ * We will check in check_ucmd_data() that user
+ * cleared everything after inlen.
+ */
+ params->inlen = (inlen < min_inlen) ? 0 : min(inlen, ucmd);
+ goto out;
}
- if (init_attr->qp_type == IB_QPT_DRIVER) {
- struct mlx5_ib_create_qp ucmd;
+ /* RSS RAW QP */
+ if (inlen < offsetofend(struct mlx5_ib_create_qp_rss, flags))
+ return -EINVAL;
- init_attr = &mlx_init_attr;
- memcpy(init_attr, verbs_init_attr, sizeof(*verbs_init_attr));
- err = set_mlx_qp_type(dev, init_attr, &ucmd, udata);
- if (err)
- return ERR_PTR(err);
+ if (outlen < offsetofend(struct mlx5_ib_create_qp_resp, bfreg_index))
+ return -EINVAL;
- if (init_attr->qp_type == MLX5_IB_QPT_DCI) {
- if (init_attr->cap.max_recv_wr ||
- init_attr->cap.max_recv_sge) {
- mlx5_ib_dbg(dev, "DCI QP requires zero size receive queue\n");
- return ERR_PTR(-EINVAL);
- }
- } else {
- return mlx5_ib_create_dct(pd, init_attr, &ucmd, udata);
- }
+ ucmd = sizeof(struct mlx5_ib_create_qp_rss);
+ params->ucmd_size = ucmd;
+ if (inlen > ucmd && !ib_is_udata_cleared(udata, ucmd, inlen - ucmd))
+ return -EINVAL;
+
+ params->inlen = min(ucmd, inlen);
+out:
+ if (!params->inlen)
+ mlx5_ib_dbg(dev, "udata is too small\n");
+
+ return (params->inlen) ? 0 : -EINVAL;
+}
+
+static int create_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
+ struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
+{
+ int err;
+
+ if (params->is_rss_raw) {
+ err = create_rss_raw_qp_tir(dev, pd, qp, params);
+ goto out;
}
- switch (init_attr->qp_type) {
- case IB_QPT_XRC_TGT:
- case IB_QPT_XRC_INI:
- if (!MLX5_CAP_GEN(dev->mdev, xrc)) {
- mlx5_ib_dbg(dev, "XRC not supported\n");
- return ERR_PTR(-ENOSYS);
- }
- init_attr->recv_cq = NULL;
- if (init_attr->qp_type == IB_QPT_XRC_TGT) {
- xrcdn = to_mxrcd(init_attr->xrcd)->xrcdn;
- init_attr->send_cq = NULL;
- }
+ if (qp->type == MLX5_IB_QPT_DCT) {
+ err = create_dct(dev, pd, qp, params);
+ goto out;
+ }
- /* fall through */
- case IB_QPT_RAW_PACKET:
- case IB_QPT_RC:
- case IB_QPT_UC:
- case IB_QPT_UD:
- case IB_QPT_SMI:
- case MLX5_IB_QPT_HW_GSI:
- case MLX5_IB_QPT_REG_UMR:
- case MLX5_IB_QPT_DCI:
- qp = kzalloc(sizeof(*qp), GFP_KERNEL);
- if (!qp)
- return ERR_PTR(-ENOMEM);
+ if (qp->type == IB_QPT_XRC_TGT) {
+ err = create_xrc_tgt_qp(dev, qp, params);
+ goto out;
+ }
- err = create_qp_common(dev, pd, init_attr, udata, qp);
- if (err) {
- mlx5_ib_dbg(dev, "create_qp_common failed\n");
- kfree(qp);
- return ERR_PTR(err);
- }
+ if (params->udata)
+ err = create_user_qp(dev, pd, qp, params);
+ else
+ err = create_kernel_qp(dev, pd, qp, params);
- if (is_qp0(init_attr->qp_type))
- qp->ibqp.qp_num = 0;
- else if (is_qp1(init_attr->qp_type))
- qp->ibqp.qp_num = 1;
- else
- qp->ibqp.qp_num = qp->trans_qp.base.mqp.qpn;
+out:
+ if (err) {
+ mlx5_ib_err(dev, "Create QP type %d failed\n", qp->type);
+ return err;
+ }
- mlx5_ib_dbg(dev, "ib qpnum 0x%x, mlx qpn 0x%x, rcqn 0x%x, scqn 0x%x\n",
- qp->ibqp.qp_num, qp->trans_qp.base.mqp.qpn,
- init_attr->recv_cq ? to_mcq(init_attr->recv_cq)->mcq.cqn : -1,
- init_attr->send_cq ? to_mcq(init_attr->send_cq)->mcq.cqn : -1);
+ if (is_qp0(qp->type))
+ qp->ibqp.qp_num = 0;
+ else if (is_qp1(qp->type))
+ qp->ibqp.qp_num = 1;
+ else
+ qp->ibqp.qp_num = qp->trans_qp.base.mqp.qpn;
- qp->trans_qp.xrcdn = xrcdn;
+ mlx5_ib_dbg(dev,
+ "QP type %d, ib qpn 0x%X, mlx qpn 0x%x, rcqn 0x%x, scqn 0x%x, ece 0x%x\n",
+ qp->type, qp->ibqp.qp_num, qp->trans_qp.base.mqp.qpn,
+ params->attr->recv_cq ? to_mcq(params->attr->recv_cq)->mcq.cqn :
+ -1,
+ params->attr->send_cq ? to_mcq(params->attr->send_cq)->mcq.cqn :
+ -1,
+ params->resp.ece_options);
- break;
+ return 0;
+}
- case IB_QPT_GSI:
- return mlx5_ib_gsi_create_qp(pd, init_attr);
+static int check_qp_attr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ struct ib_qp_init_attr *attr)
+{
+ int ret = 0;
- case IB_QPT_RAW_IPV6:
- case IB_QPT_RAW_ETHERTYPE:
- case IB_QPT_MAX:
+ switch (qp->type) {
+ case MLX5_IB_QPT_DCT:
+ ret = (!attr->srq || !attr->recv_cq) ? -EINVAL : 0;
+ break;
+ case MLX5_IB_QPT_DCI:
+ ret = (attr->cap.max_recv_wr || attr->cap.max_recv_sge) ?
+ -EINVAL :
+ 0;
+ break;
+ case IB_QPT_RAW_PACKET:
+ ret = (attr->rwq_ind_tbl && attr->send_cq) ? -EINVAL : 0;
+ break;
default:
- mlx5_ib_dbg(dev, "unsupported qp type %d\n",
- init_attr->qp_type);
- /* Don't support raw QPs */
- return ERR_PTR(-EOPNOTSUPP);
+ break;
}
- if (verbs_init_attr->qp_type == IB_QPT_DRIVER)
- qp->qp_sub_type = init_attr->qp_type;
+ if (ret)
+ mlx5_ib_dbg(dev, "QP type %d has wrong attributes\n", qp->type);
- return &qp->ibqp;
+ return ret;
+}
+
+static int get_qp_uidx(struct mlx5_ib_qp *qp,
+ struct mlx5_create_qp_params *params)
+{
+ struct mlx5_ib_create_qp *ucmd = params->ucmd;
+ struct ib_udata *udata = params->udata;
+ struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
+ udata, struct mlx5_ib_ucontext, ibucontext);
+
+ if (params->is_rss_raw)
+ return 0;
+
+ return get_qp_user_index(ucontext, ucmd, sizeof(*ucmd), &params->uidx);
}
static int mlx5_ib_destroy_dct(struct mlx5_ib_qp *mqp)
@@ -2828,6 +2870,150 @@ static int mlx5_ib_destroy_dct(struct mlx5_ib_qp *mqp)
return 0;
}
+static int check_ucmd_data(struct mlx5_ib_dev *dev,
+ struct mlx5_create_qp_params *params)
+{
+ struct ib_qp_init_attr *attr = params->attr;
+ struct ib_udata *udata = params->udata;
+ size_t size, last;
+ int ret;
+
+ if (params->is_rss_raw)
+ /*
+ * These QPs don't have "reserved" field in their
+ * create_qp input struct, so their data is always valid.
+ */
+ last = sizeof(struct mlx5_ib_create_qp_rss);
+ else
+ /* IB_QPT_RAW_PACKET doesn't have ECE data */
+ switch (attr->qp_type) {
+ case IB_QPT_RAW_PACKET:
+ last = offsetof(struct mlx5_ib_create_qp, ece_options);
+ break;
+ default:
+ last = offsetof(struct mlx5_ib_create_qp, reserved);
+ }
+
+ if (udata->inlen <= last)
+ return 0;
+
+ /*
+ * User provides different create_qp structures based on the
+ * flow and we need to know if he cleared memory after our
+ * struct create_qp ends.
+ */
+ size = udata->inlen - last;
+ ret = ib_is_udata_cleared(params->udata, last, size);
+ if (!ret)
+ mlx5_ib_dbg(
+ dev,
+ "udata is not cleared, inlen = %lu, ucmd = %lu, last = %lu, size = %lu\n",
+ udata->inlen, params->ucmd_size, last, size);
+ return ret ? 0 : -EINVAL;
+}
+
+struct ib_qp *mlx5_ib_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attr,
+ struct ib_udata *udata)
+{
+ struct mlx5_create_qp_params params = {};
+ struct mlx5_ib_dev *dev;
+ struct mlx5_ib_qp *qp;
+ enum ib_qp_type type;
+ int err;
+
+ dev = pd ? to_mdev(pd->device) :
+ to_mdev(to_mxrcd(attr->xrcd)->ibxrcd.device);
+
+ err = check_qp_type(dev, attr, &type);
+ if (err)
+ return ERR_PTR(err);
+
+ err = check_valid_flow(dev, pd, attr, udata);
+ if (err)
+ return ERR_PTR(err);
+
+ if (attr->qp_type == IB_QPT_GSI)
+ return mlx5_ib_gsi_create_qp(pd, attr);
+
+ params.udata = udata;
+ params.uidx = MLX5_IB_DEFAULT_UIDX;
+ params.attr = attr;
+ params.is_rss_raw = !!attr->rwq_ind_tbl;
+
+ if (udata) {
+ err = process_udata_size(dev, &params);
+ if (err)
+ return ERR_PTR(err);
+
+ err = check_ucmd_data(dev, &params);
+ if (err)
+ return ERR_PTR(err);
+
+ params.ucmd = kzalloc(params.ucmd_size, GFP_KERNEL);
+ if (!params.ucmd)
+ return ERR_PTR(-ENOMEM);
+
+ err = ib_copy_from_udata(params.ucmd, udata, params.inlen);
+ if (err)
+ goto free_ucmd;
+ }
+
+ qp = kzalloc(sizeof(*qp), GFP_KERNEL);
+ if (!qp) {
+ err = -ENOMEM;
+ goto free_ucmd;
+ }
+
+ qp->type = type;
+ if (udata) {
+ err = process_vendor_flags(dev, qp, params.ucmd, attr);
+ if (err)
+ goto free_qp;
+
+ err = get_qp_uidx(qp, &params);
+ if (err)
+ goto free_qp;
+ }
+ err = process_create_flags(dev, qp, attr);
+ if (err)
+ goto free_qp;
+
+ err = check_qp_attr(dev, qp, attr);
+ if (err)
+ goto free_qp;
+
+ err = create_qp(dev, pd, qp, &params);
+ if (err)
+ goto free_qp;
+
+ kfree(params.ucmd);
+ params.ucmd = NULL;
+
+ if (udata)
+ /*
+ * It is safe to copy response for all user create QP flows,
+ * including MLX5_IB_QPT_DCT, which doesn't need it.
+ * In that case, resp will be filled with zeros.
+ */
+ err = ib_copy_to_udata(udata, &params.resp, params.outlen);
+ if (err)
+ goto destroy_qp;
+
+ return &qp->ibqp;
+
+destroy_qp:
+ if (qp->type == MLX5_IB_QPT_DCT)
+ mlx5_ib_destroy_dct(qp);
+ else
+ destroy_qp_common(dev, qp, udata);
+ qp = NULL;
+free_qp:
+ kfree(qp);
+free_ucmd:
+ kfree(params.ucmd);
+ return ERR_PTR(err);
+}
+
int mlx5_ib_destroy_qp(struct ib_qp *qp, struct ib_udata *udata)
{
struct mlx5_ib_dev *dev = to_mdev(qp->device);
@@ -2836,7 +3022,7 @@ int mlx5_ib_destroy_qp(struct ib_qp *qp, struct ib_udata *udata)
if (unlikely(qp->qp_type == IB_QPT_GSI))
return mlx5_ib_gsi_destroy_qp(qp);
- if (mqp->qp_sub_type == MLX5_IB_QPT_DCT)
+ if (mqp->type == MLX5_IB_QPT_DCT)
return mlx5_ib_destroy_dct(mqp);
destroy_qp_common(dev, mqp, udata);
@@ -2846,14 +3032,13 @@ int mlx5_ib_destroy_qp(struct ib_qp *qp, struct ib_udata *udata)
return 0;
}
-static int to_mlx5_access_flags(struct mlx5_ib_qp *qp,
- const struct ib_qp_attr *attr,
- int attr_mask, __be32 *hw_access_flags_be)
+static int set_qpc_atomic_flags(struct mlx5_ib_qp *qp,
+ const struct ib_qp_attr *attr, int attr_mask,
+ void *qpc)
{
- u8 dest_rd_atomic;
- u32 access_flags, hw_access_flags = 0;
-
struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.device);
+ u8 dest_rd_atomic;
+ u32 access_flags;
if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC)
dest_rd_atomic = attr->max_dest_rd_atomic;
@@ -2868,8 +3053,8 @@ static int to_mlx5_access_flags(struct mlx5_ib_qp *qp,
if (!dest_rd_atomic)
access_flags &= IB_ACCESS_REMOTE_WRITE;
- if (access_flags & IB_ACCESS_REMOTE_READ)
- hw_access_flags |= MLX5_QP_BIT_RRE;
+ MLX5_SET(qpc, qpc, rre, !!(access_flags & IB_ACCESS_REMOTE_READ));
+
if (access_flags & IB_ACCESS_REMOTE_ATOMIC) {
int atomic_mode;
@@ -2877,15 +3062,11 @@ static int to_mlx5_access_flags(struct mlx5_ib_qp *qp,
if (atomic_mode < 0)
return -EOPNOTSUPP;
- hw_access_flags |= MLX5_QP_BIT_RAE;
- hw_access_flags |= atomic_mode << MLX5_ATOMIC_MODE_OFFSET;
+ MLX5_SET(qpc, qpc, rae, 1);
+ MLX5_SET(qpc, qpc, atomic_mode, atomic_mode);
}
- if (access_flags & IB_ACCESS_REMOTE_WRITE)
- hw_access_flags |= MLX5_QP_BIT_RWE;
-
- *hw_access_flags_be = cpu_to_be32(hw_access_flags);
-
+ MLX5_SET(qpc, qpc, rwe, !!(access_flags & IB_ACCESS_REMOTE_WRITE));
return 0;
}
@@ -2965,11 +3146,22 @@ static int modify_raw_packet_tx_affinity(struct mlx5_core_dev *dev,
return err;
}
+static void mlx5_set_path_udp_sport(void *path, const struct rdma_ah_attr *ah,
+ u32 lqpn, u32 rqpn)
+
+{
+ u32 fl = ah->grh.flow_label;
+
+ if (!fl)
+ fl = rdma_calc_flow_label(lqpn, rqpn);
+
+ MLX5_SET(ads, path, udp_sport, rdma_flow_label_to_udp_sport(fl));
+}
+
static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
- const struct rdma_ah_attr *ah,
- struct mlx5_qp_path *path, u8 port, int attr_mask,
- u32 path_flags, const struct ib_qp_attr *attr,
- bool alt)
+ const struct rdma_ah_attr *ah, void *path, u8 port,
+ int attr_mask, u32 path_flags,
+ const struct ib_qp_attr *attr, bool alt)
{
const struct ib_global_route *grh = rdma_ah_read_grh(ah);
int err;
@@ -2978,8 +3170,8 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
u8 sl = rdma_ah_get_sl(ah);
if (attr_mask & IB_QP_PKEY_INDEX)
- path->pkey_index = cpu_to_be16(alt ? attr->alt_pkey_index :
- attr->pkey_index);
+ MLX5_SET(ads, path, pkey_index,
+ alt ? attr->alt_pkey_index : attr->pkey_index);
if (ah_flags & IB_AH_GRH) {
if (grh->sgid_index >=
@@ -2995,45 +3187,49 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
if (!(ah_flags & IB_AH_GRH))
return -EINVAL;
- memcpy(path->rmac, ah->roce.dmac, sizeof(ah->roce.dmac));
- if (qp->ibqp.qp_type == IB_QPT_RC ||
- qp->ibqp.qp_type == IB_QPT_UC ||
- qp->ibqp.qp_type == IB_QPT_XRC_INI ||
- qp->ibqp.qp_type == IB_QPT_XRC_TGT)
- path->udp_sport =
- mlx5_get_roce_udp_sport(dev, ah->grh.sgid_attr);
- path->dci_cfi_prio_sl = (sl & 0x7) << 4;
+ ether_addr_copy(MLX5_ADDR_OF(ads, path, rmac_47_32),
+ ah->roce.dmac);
+ if ((qp->ibqp.qp_type == IB_QPT_RC ||
+ qp->ibqp.qp_type == IB_QPT_UC ||
+ qp->ibqp.qp_type == IB_QPT_XRC_INI ||
+ qp->ibqp.qp_type == IB_QPT_XRC_TGT) &&
+ (grh->sgid_attr->gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) &&
+ (attr_mask & IB_QP_DEST_QPN))
+ mlx5_set_path_udp_sport(path, ah,
+ qp->ibqp.qp_num,
+ attr->dest_qp_num);
+ MLX5_SET(ads, path, eth_prio, sl & 0x7);
gid_type = ah->grh.sgid_attr->gid_type;
if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
- path->ecn_dscp = (grh->traffic_class >> 2) & 0x3f;
+ MLX5_SET(ads, path, dscp, grh->traffic_class >> 2);
} else {
- path->fl_free_ar = (path_flags & MLX5_PATH_FLAG_FL) ? 0x80 : 0;
- path->fl_free_ar |=
- (path_flags & MLX5_PATH_FLAG_FREE_AR) ? 0x40 : 0;
- path->rlid = cpu_to_be16(rdma_ah_get_dlid(ah));
- path->grh_mlid = rdma_ah_get_path_bits(ah) & 0x7f;
- if (ah_flags & IB_AH_GRH)
- path->grh_mlid |= 1 << 7;
- path->dci_cfi_prio_sl = sl & 0xf;
+ MLX5_SET(ads, path, fl, !!(path_flags & MLX5_PATH_FLAG_FL));
+ MLX5_SET(ads, path, free_ar,
+ !!(path_flags & MLX5_PATH_FLAG_FREE_AR));
+ MLX5_SET(ads, path, rlid, rdma_ah_get_dlid(ah));
+ MLX5_SET(ads, path, mlid, rdma_ah_get_path_bits(ah));
+ MLX5_SET(ads, path, grh, !!(ah_flags & IB_AH_GRH));
+ MLX5_SET(ads, path, sl, sl);
}
if (ah_flags & IB_AH_GRH) {
- path->mgid_index = grh->sgid_index;
- path->hop_limit = grh->hop_limit;
- path->tclass_flowlabel =
- cpu_to_be32((grh->traffic_class << 20) |
- (grh->flow_label));
- memcpy(path->rgid, grh->dgid.raw, 16);
+ MLX5_SET(ads, path, src_addr_index, grh->sgid_index);
+ MLX5_SET(ads, path, hop_limit, grh->hop_limit);
+ MLX5_SET(ads, path, tclass, grh->traffic_class);
+ MLX5_SET(ads, path, flow_label, grh->flow_label);
+ memcpy(MLX5_ADDR_OF(ads, path, rgid_rip), grh->dgid.raw,
+ sizeof(grh->dgid.raw));
}
err = ib_rate_to_mlx5(dev, rdma_ah_get_static_rate(ah));
if (err < 0)
return err;
- path->static_rate = err;
- path->port = port;
+ MLX5_SET(ads, path, stat_rate, err);
+ MLX5_SET(ads, path, vhca_port_num, port);
if (attr_mask & IB_QP_TIMEOUT)
- path->ackto_lt = (alt ? attr->alt_timeout : attr->timeout) << 3;
+ MLX5_SET(ads, path, ack_timeout,
+ alt ? attr->alt_timeout : attr->timeout);
if ((qp->ibqp.qp_type == IB_QPT_RAW_PACKET) && qp->sq.wqe_cnt)
return modify_raw_packet_eth_prio(dev->mdev,
@@ -3050,10 +3246,12 @@ static enum mlx5_qp_optpar opt_mask[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE][MLX5_Q
MLX5_QP_OPTPAR_RAE |
MLX5_QP_OPTPAR_RWE |
MLX5_QP_OPTPAR_PKEY_INDEX |
- MLX5_QP_OPTPAR_PRI_PORT,
+ MLX5_QP_OPTPAR_PRI_PORT |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
[MLX5_QP_ST_UC] = MLX5_QP_OPTPAR_RWE |
MLX5_QP_OPTPAR_PKEY_INDEX |
- MLX5_QP_OPTPAR_PRI_PORT,
+ MLX5_QP_OPTPAR_PRI_PORT |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
[MLX5_QP_ST_UD] = MLX5_QP_OPTPAR_PKEY_INDEX |
MLX5_QP_OPTPAR_Q_KEY |
MLX5_QP_OPTPAR_PRI_PORT,
@@ -3061,17 +3259,20 @@ static enum mlx5_qp_optpar opt_mask[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE][MLX5_Q
MLX5_QP_OPTPAR_RAE |
MLX5_QP_OPTPAR_RWE |
MLX5_QP_OPTPAR_PKEY_INDEX |
- MLX5_QP_OPTPAR_PRI_PORT,
+ MLX5_QP_OPTPAR_PRI_PORT |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
},
[MLX5_QP_STATE_RTR] = {
[MLX5_QP_ST_RC] = MLX5_QP_OPTPAR_ALT_ADDR_PATH |
MLX5_QP_OPTPAR_RRE |
MLX5_QP_OPTPAR_RAE |
MLX5_QP_OPTPAR_RWE |
- MLX5_QP_OPTPAR_PKEY_INDEX,
+ MLX5_QP_OPTPAR_PKEY_INDEX |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
[MLX5_QP_ST_UC] = MLX5_QP_OPTPAR_ALT_ADDR_PATH |
MLX5_QP_OPTPAR_RWE |
- MLX5_QP_OPTPAR_PKEY_INDEX,
+ MLX5_QP_OPTPAR_PKEY_INDEX |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
[MLX5_QP_ST_UD] = MLX5_QP_OPTPAR_PKEY_INDEX |
MLX5_QP_OPTPAR_Q_KEY,
[MLX5_QP_ST_MLX] = MLX5_QP_OPTPAR_PKEY_INDEX |
@@ -3080,7 +3281,8 @@ static enum mlx5_qp_optpar opt_mask[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE][MLX5_Q
MLX5_QP_OPTPAR_RRE |
MLX5_QP_OPTPAR_RAE |
MLX5_QP_OPTPAR_RWE |
- MLX5_QP_OPTPAR_PKEY_INDEX,
+ MLX5_QP_OPTPAR_PKEY_INDEX |
+ MLX5_QP_OPTPAR_LAG_TX_AFF,
},
},
[MLX5_QP_STATE_RTR] = {
@@ -3414,43 +3616,80 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
return 0;
}
-static unsigned int get_tx_affinity(struct mlx5_ib_dev *dev,
- struct mlx5_ib_pd *pd,
- struct mlx5_ib_qp_base *qp_base,
- u8 port_num, struct ib_udata *udata)
+static unsigned int get_tx_affinity_rr(struct mlx5_ib_dev *dev,
+ struct ib_udata *udata)
{
struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
udata, struct mlx5_ib_ucontext, ibucontext);
- unsigned int tx_port_affinity;
+ u8 port_num = mlx5_core_native_port_num(dev->mdev) - 1;
+ atomic_t *tx_port_affinity;
+
+ if (ucontext)
+ tx_port_affinity = &ucontext->tx_port_affinity;
+ else
+ tx_port_affinity = &dev->port[port_num].roce.tx_port_affinity;
- if (ucontext) {
- tx_port_affinity = (unsigned int)atomic_add_return(
- 1, &ucontext->tx_port_affinity) %
- MLX5_MAX_PORTS +
- 1;
+ return (unsigned int)atomic_add_return(1, tx_port_affinity) %
+ MLX5_MAX_PORTS + 1;
+}
+
+static bool qp_supports_affinity(struct ib_qp *qp)
+{
+ if ((qp->qp_type == IB_QPT_RC) ||
+ (qp->qp_type == IB_QPT_UD) ||
+ (qp->qp_type == IB_QPT_UC) ||
+ (qp->qp_type == IB_QPT_RAW_PACKET) ||
+ (qp->qp_type == IB_QPT_XRC_INI) ||
+ (qp->qp_type == IB_QPT_XRC_TGT))
+ return true;
+ return false;
+}
+
+static unsigned int get_tx_affinity(struct ib_qp *qp,
+ const struct ib_qp_attr *attr,
+ int attr_mask, u8 init,
+ struct ib_udata *udata)
+{
+ struct mlx5_ib_ucontext *ucontext = rdma_udata_to_drv_context(
+ udata, struct mlx5_ib_ucontext, ibucontext);
+ struct mlx5_ib_dev *dev = to_mdev(qp->device);
+ struct mlx5_ib_qp *mqp = to_mqp(qp);
+ struct mlx5_ib_qp_base *qp_base;
+ unsigned int tx_affinity;
+
+ if (!(mlx5_ib_lag_should_assign_affinity(dev) &&
+ qp_supports_affinity(qp)))
+ return 0;
+
+ if (mqp->flags & MLX5_IB_QP_CREATE_SQPN_QP1)
+ tx_affinity = mqp->gsi_lag_port;
+ else if (init)
+ tx_affinity = get_tx_affinity_rr(dev, udata);
+ else if ((attr_mask & IB_QP_AV) && attr->xmit_slave)
+ tx_affinity =
+ mlx5_lag_get_slave_port(dev->mdev, attr->xmit_slave);
+ else
+ return 0;
+
+ qp_base = &mqp->trans_qp.base;
+ if (ucontext)
mlx5_ib_dbg(dev, "Set tx affinity 0x%x to qpn 0x%x ucontext %p\n",
- tx_port_affinity, qp_base->mqp.qpn, ucontext);
- } else {
- tx_port_affinity =
- (unsigned int)atomic_add_return(
- 1, &dev->port[port_num].roce.tx_port_affinity) %
- MLX5_MAX_PORTS +
- 1;
+ tx_affinity, qp_base->mqp.qpn, ucontext);
+ else
mlx5_ib_dbg(dev, "Set tx affinity 0x%x to qpn 0x%x\n",
- tx_port_affinity, qp_base->mqp.qpn);
- }
-
- return tx_port_affinity;
+ tx_affinity, qp_base->mqp.qpn);
+ return tx_affinity;
}
static int __mlx5_ib_qp_set_counter(struct ib_qp *qp,
struct rdma_counter *counter)
{
struct mlx5_ib_dev *dev = to_mdev(qp->device);
+ u32 in[MLX5_ST_SZ_DW(rts2rts_qp_in)] = {};
struct mlx5_ib_qp *mqp = to_mqp(qp);
- struct mlx5_qp_context context = {};
struct mlx5_ib_qp_base *base;
u32 set_id;
+ u32 *qpc;
if (counter)
set_id = counter->id;
@@ -3458,11 +3697,15 @@ static int __mlx5_ib_qp_set_counter(struct ib_qp *qp,
set_id = mlx5_ib_get_counters_id(dev, mqp->port - 1);
base = &mqp->trans_qp.base;
- context.qp_counter_set_usr_page &= cpu_to_be32(0xffffff);
- context.qp_counter_set_usr_page |= cpu_to_be32(set_id << 24);
- return mlx5_core_qp_modify(dev, MLX5_CMD_OP_RTS2RTS_QP,
- MLX5_QP_OPTPAR_COUNTER_SET_ID, &context,
- &base->mqp);
+ MLX5_SET(rts2rts_qp_in, in, opcode, MLX5_CMD_OP_RTS2RTS_QP);
+ MLX5_SET(rts2rts_qp_in, in, qpn, base->mqp.qpn);
+ MLX5_SET(rts2rts_qp_in, in, uid, base->mqp.uid);
+ MLX5_SET(rts2rts_qp_in, in, opt_param_mask,
+ MLX5_QP_OPTPAR_COUNTER_SET_ID);
+
+ qpc = MLX5_ADDR_OF(rts2rts_qp_in, in, qpc);
+ MLX5_SET(qpc, qpc, counter_set_id, set_id);
+ return mlx5_cmd_exec_in(dev->mdev, rts2rts_qp, in);
}
static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
@@ -3470,6 +3713,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
enum ib_qp_state cur_state,
enum ib_qp_state new_state,
const struct mlx5_ib_modify_qp *ucmd,
+ struct mlx5_ib_modify_qp_resp *resp,
struct ib_udata *udata)
{
static const u16 optab[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE] = {
@@ -3513,67 +3757,60 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
struct mlx5_ib_qp *qp = to_mqp(ibqp);
struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
struct mlx5_ib_cq *send_cq, *recv_cq;
- struct mlx5_qp_context *context;
struct mlx5_ib_pd *pd;
enum mlx5_qp_state mlx5_cur, mlx5_new;
- enum mlx5_qp_optpar optpar;
+ void *qpc, *pri_path, *alt_path;
+ enum mlx5_qp_optpar optpar = 0;
u32 set_id = 0;
int mlx5_st;
int err;
u16 op;
u8 tx_affinity = 0;
- mlx5_st = to_mlx5_st(ibqp->qp_type == IB_QPT_DRIVER ?
- qp->qp_sub_type : ibqp->qp_type);
+ mlx5_st = to_mlx5_st(qp->type);
if (mlx5_st < 0)
return -EINVAL;
- context = kzalloc(sizeof(*context), GFP_KERNEL);
- if (!context)
+ qpc = kzalloc(MLX5_ST_SZ_BYTES(qpc), GFP_KERNEL);
+ if (!qpc)
return -ENOMEM;
- pd = get_pd(qp);
- context->flags = cpu_to_be32(mlx5_st << 16);
+ pd = to_mpd(qp->ibqp.pd);
+ MLX5_SET(qpc, qpc, st, mlx5_st);
if (!(attr_mask & IB_QP_PATH_MIG_STATE)) {
- context->flags |= cpu_to_be32(MLX5_QP_PM_MIGRATED << 11);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
} else {
switch (attr->path_mig_state) {
case IB_MIG_MIGRATED:
- context->flags |= cpu_to_be32(MLX5_QP_PM_MIGRATED << 11);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
break;
case IB_MIG_REARM:
- context->flags |= cpu_to_be32(MLX5_QP_PM_REARM << 11);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_REARM);
break;
case IB_MIG_ARMED:
- context->flags |= cpu_to_be32(MLX5_QP_PM_ARMED << 11);
+ MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_ARMED);
break;
}
}
- if ((cur_state == IB_QPS_RESET) && (new_state == IB_QPS_INIT)) {
- if ((ibqp->qp_type == IB_QPT_RC) ||
- (ibqp->qp_type == IB_QPT_UD &&
- !(qp->flags & MLX5_IB_QP_SQPN_QP1)) ||
- (ibqp->qp_type == IB_QPT_UC) ||
- (ibqp->qp_type == IB_QPT_RAW_PACKET) ||
- (ibqp->qp_type == IB_QPT_XRC_INI) ||
- (ibqp->qp_type == IB_QPT_XRC_TGT)) {
- if (dev->lag_active) {
- u8 p = mlx5_core_native_port_num(dev->mdev) - 1;
- tx_affinity = get_tx_affinity(dev, pd, base, p,
- udata);
- context->flags |= cpu_to_be32(tx_affinity << 24);
- }
- }
- }
+ tx_affinity = get_tx_affinity(ibqp, attr, attr_mask,
+ cur_state == IB_QPS_RESET &&
+ new_state == IB_QPS_INIT, udata);
+
+ MLX5_SET(qpc, qpc, lag_tx_port_affinity, tx_affinity);
+ if (tx_affinity && new_state == IB_QPS_RTR &&
+ MLX5_CAP_GEN(dev->mdev, init2_lag_tx_port_affinity))
+ optpar |= MLX5_QP_OPTPAR_LAG_TX_AFF;
if (is_sqp(ibqp->qp_type)) {
- context->mtu_msgmax = (IB_MTU_256 << 5) | 8;
+ MLX5_SET(qpc, qpc, mtu, IB_MTU_256);
+ MLX5_SET(qpc, qpc, log_msg_max, 8);
} else if ((ibqp->qp_type == IB_QPT_UD &&
- !(qp->flags & MLX5_IB_QP_UNDERLAY)) ||
+ !(qp->flags & IB_QP_CREATE_SOURCE_QPN)) ||
ibqp->qp_type == MLX5_IB_QPT_REG_UMR) {
- context->mtu_msgmax = (IB_MTU_4096 << 5) | 12;
+ MLX5_SET(qpc, qpc, mtu, IB_MTU_4096);
+ MLX5_SET(qpc, qpc, log_msg_max, 12);
} else if (attr_mask & IB_QP_PATH_MTU) {
if (attr->path_mtu < IB_MTU_256 ||
attr->path_mtu > IB_MTU_4096) {
@@ -3581,40 +3818,45 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
err = -EINVAL;
goto out;
}
- context->mtu_msgmax = (attr->path_mtu << 5) |
- (u8)MLX5_CAP_GEN(dev->mdev, log_max_msg);
+ MLX5_SET(qpc, qpc, mtu, attr->path_mtu);
+ MLX5_SET(qpc, qpc, log_msg_max,
+ MLX5_CAP_GEN(dev->mdev, log_max_msg));
}
if (attr_mask & IB_QP_DEST_QPN)
- context->log_pg_sz_remote_qpn = cpu_to_be32(attr->dest_qp_num);
+ MLX5_SET(qpc, qpc, remote_qpn, attr->dest_qp_num);
+
+ pri_path = MLX5_ADDR_OF(qpc, qpc, primary_address_path);
+ alt_path = MLX5_ADDR_OF(qpc, qpc, secondary_address_path);
if (attr_mask & IB_QP_PKEY_INDEX)
- context->pri_path.pkey_index = cpu_to_be16(attr->pkey_index);
+ MLX5_SET(ads, pri_path, pkey_index, attr->pkey_index);
/* todo implement counter_index functionality */
if (is_sqp(ibqp->qp_type))
- context->pri_path.port = qp->port;
+ MLX5_SET(ads, pri_path, vhca_port_num, qp->port);
if (attr_mask & IB_QP_PORT)
- context->pri_path.port = attr->port_num;
+ MLX5_SET(ads, pri_path, vhca_port_num, attr->port_num);
if (attr_mask & IB_QP_AV) {
- err = mlx5_set_path(dev, qp, &attr->ah_attr, &context->pri_path,
- attr_mask & IB_QP_PORT ? attr->port_num : qp->port,
+ err = mlx5_set_path(dev, qp, &attr->ah_attr, pri_path,
+ attr_mask & IB_QP_PORT ? attr->port_num :
+ qp->port,
attr_mask, 0, attr, false);
if (err)
goto out;
}
if (attr_mask & IB_QP_TIMEOUT)
- context->pri_path.ackto_lt |= attr->timeout << 3;
+ MLX5_SET(ads, pri_path, ack_timeout, attr->timeout);
if (attr_mask & IB_QP_ALT_PATH) {
- err = mlx5_set_path(dev, qp, &attr->alt_ah_attr,
- &context->alt_path,
+ err = mlx5_set_path(dev, qp, &attr->alt_ah_attr, alt_path,
attr->alt_port_num,
- attr_mask | IB_QP_PKEY_INDEX | IB_QP_TIMEOUT,
+ attr_mask | IB_QP_PKEY_INDEX |
+ IB_QP_TIMEOUT,
0, attr, true);
if (err)
goto out;
@@ -3623,75 +3865,68 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
get_cqs(qp->ibqp.qp_type, qp->ibqp.send_cq, qp->ibqp.recv_cq,
&send_cq, &recv_cq);
- context->flags_pd = cpu_to_be32(pd ? pd->pdn : to_mpd(dev->devr.p0)->pdn);
- context->cqn_send = send_cq ? cpu_to_be32(send_cq->mcq.cqn) : 0;
- context->cqn_recv = recv_cq ? cpu_to_be32(recv_cq->mcq.cqn) : 0;
- context->params1 = cpu_to_be32(MLX5_IB_ACK_REQ_FREQ << 28);
+ MLX5_SET(qpc, qpc, pd, pd ? pd->pdn : to_mpd(dev->devr.p0)->pdn);
+ if (send_cq)
+ MLX5_SET(qpc, qpc, cqn_snd, send_cq->mcq.cqn);
+ if (recv_cq)
+ MLX5_SET(qpc, qpc, cqn_rcv, recv_cq->mcq.cqn);
+
+ MLX5_SET(qpc, qpc, log_ack_req_freq, MLX5_IB_ACK_REQ_FREQ);
if (attr_mask & IB_QP_RNR_RETRY)
- context->params1 |= cpu_to_be32(attr->rnr_retry << 13);
+ MLX5_SET(qpc, qpc, rnr_retry, attr->rnr_retry);
if (attr_mask & IB_QP_RETRY_CNT)
- context->params1 |= cpu_to_be32(attr->retry_cnt << 16);
+ MLX5_SET(qpc, qpc, retry_count, attr->retry_cnt);
- if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC) {
- if (attr->max_rd_atomic)
- context->params1 |=
- cpu_to_be32(fls(attr->max_rd_atomic - 1) << 21);
- }
+ if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC && attr->max_rd_atomic)
+ MLX5_SET(qpc, qpc, log_sra_max, ilog2(attr->max_rd_atomic));
if (attr_mask & IB_QP_SQ_PSN)
- context->next_send_psn = cpu_to_be32(attr->sq_psn);
+ MLX5_SET(qpc, qpc, next_send_psn, attr->sq_psn);
- if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) {
- if (attr->max_dest_rd_atomic)
- context->params2 |=
- cpu_to_be32(fls(attr->max_dest_rd_atomic - 1) << 21);
- }
+ if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC && attr->max_dest_rd_atomic)
+ MLX5_SET(qpc, qpc, log_rra_max,
+ ilog2(attr->max_dest_rd_atomic));
if (attr_mask & (IB_QP_ACCESS_FLAGS | IB_QP_MAX_DEST_RD_ATOMIC)) {
- __be32 access_flags;
-
- err = to_mlx5_access_flags(qp, attr, attr_mask, &access_flags);
+ err = set_qpc_atomic_flags(qp, attr, attr_mask, qpc);
if (err)
goto out;
-
- context->params2 |= access_flags;
}
if (attr_mask & IB_QP_MIN_RNR_TIMER)
- context->rnr_nextrecvpsn |= cpu_to_be32(attr->min_rnr_timer << 24);
+ MLX5_SET(qpc, qpc, min_rnr_nak, attr->min_rnr_timer);
if (attr_mask & IB_QP_RQ_PSN)
- context->rnr_nextrecvpsn |= cpu_to_be32(attr->rq_psn);
+ MLX5_SET(qpc, qpc, next_rcv_psn, attr->rq_psn);
if (attr_mask & IB_QP_QKEY)
- context->qkey = cpu_to_be32(attr->qkey);
+ MLX5_SET(qpc, qpc, q_key, attr->qkey);
if (qp->rq.wqe_cnt && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT)
- context->db_rec_addr = cpu_to_be64(qp->db.dma);
+ MLX5_SET64(qpc, qpc, dbr_addr, qp->db.dma);
if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {
u8 port_num = (attr_mask & IB_QP_PORT ? attr->port_num :
qp->port) - 1;
/* Underlay port should be used - index 0 function per port */
- if (qp->flags & MLX5_IB_QP_UNDERLAY)
+ if (qp->flags & IB_QP_CREATE_SOURCE_QPN)
port_num = 0;
if (ibqp->counter)
set_id = ibqp->counter->id;
else
set_id = mlx5_ib_get_counters_id(dev, port_num);
- context->qp_counter_set_usr_page |=
- cpu_to_be32(set_id << 24);
+ MLX5_SET(qpc, qpc, counter_set_id, set_id);
}
if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT)
- context->sq_crq_size |= cpu_to_be16(1 << 4);
+ MLX5_SET(qpc, qpc, rlky, 1);
- if (qp->flags & MLX5_IB_QP_SQPN_QP1)
- context->deth_sqpn = cpu_to_be32(1);
+ if (qp->flags & MLX5_IB_QP_CREATE_SQPN_QP1)
+ MLX5_SET(qpc, qpc, deth_sqpn, 1);
mlx5_cur = to_mlx5_state(cur_state);
mlx5_new = to_mlx5_state(new_state);
@@ -3703,11 +3938,11 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
}
op = optab[mlx5_cur][mlx5_new];
- optpar = ib_mask_to_mlx5_opt(attr_mask);
+ optpar |= ib_mask_to_mlx5_opt(attr_mask);
optpar &= opt_mask[mlx5_cur][mlx5_new][mlx5_st];
if (qp->ibqp.qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) {
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) {
struct mlx5_modify_raw_qp_param raw_qp_param = {};
raw_qp_param.operation = op;
@@ -3749,7 +3984,15 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
err = modify_raw_packet_qp(dev, qp, &raw_qp_param, tx_affinity);
} else {
- err = mlx5_core_qp_modify(dev, op, optpar, context, &base->mqp);
+ if (udata) {
+ /* For the kernel flows, the resp will stay zero */
+ resp->ece_options =
+ MLX5_CAP_GEN(dev->mdev, ece_support) ?
+ ucmd->ece_options : 0;
+ resp->response_length = sizeof(*resp);
+ }
+ err = mlx5_core_qp_modify(dev, op, optpar, qpc, &base->mqp,
+ &resp->ece_options);
}
if (err)
@@ -3796,7 +4039,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
}
out:
- kfree(context);
+ kfree(qpc);
return err;
}
@@ -3854,7 +4097,8 @@ static bool modify_dci_qp_is_ok(enum ib_qp_state cur_state, enum ib_qp_state new
* Other transitions and attributes are illegal
*/
static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
- int attr_mask, struct ib_udata *udata)
+ int attr_mask, struct mlx5_ib_modify_qp *ucmd,
+ struct ib_udata *udata)
{
struct mlx5_ib_qp *qp = to_mqp(ibqp);
struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
@@ -3870,6 +4114,15 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
new_state = attr->qp_state;
dctc = MLX5_ADDR_OF(create_dct_in, qp->dct.in, dct_context_entry);
+ if (MLX5_CAP_GEN(dev->mdev, ece_support) && ucmd->ece_options)
+ /*
+ * DCT doesn't initialize QP till modify command is executed,
+ * so we need to overwrite previously set ECE field if user
+ * provided any value except zero, which means not set/not
+ * valid.
+ */
+ MLX5_SET(dctc, dctc, ece, ucmd->ece_options);
+
if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {
u16 set_id;
@@ -3902,17 +4155,23 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
set_id = mlx5_ib_get_counters_id(dev, attr->port_num - 1);
MLX5_SET(dctc, dctc, counter_set_id, set_id);
-
} else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_RTR) {
struct mlx5_ib_modify_qp_resp resp = {};
- u32 out[MLX5_ST_SZ_DW(create_dct_out)] = {0};
- u32 min_resp_len = offsetof(typeof(resp), dctn) +
- sizeof(resp.dctn);
+ u32 out[MLX5_ST_SZ_DW(create_dct_out)] = {};
+ u32 min_resp_len = offsetofend(typeof(resp), dctn);
if (udata->outlen < min_resp_len)
return -EINVAL;
resp.response_length = min_resp_len;
+ /*
+ * If we don't have enough space for the ECE options,
+ * simply indicate it with resp.response_length.
+ */
+ resp.response_length = (udata->outlen < sizeof(resp)) ?
+ min_resp_len :
+ sizeof(resp);
+
required |= IB_QP_MIN_RNR_TIMER | IB_QP_AV | IB_QP_PATH_MTU;
if (!is_valid_mask(attr_mask, required, 0))
return -EINVAL;
@@ -3929,6 +4188,8 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
if (err)
return err;
resp.dctn = qp->dct.mdct.mqp.qpn;
+ if (MLX5_CAP_GEN(dev->mdev, ece_support))
+ resp.ece_options = MLX5_GET(create_dct_out, out, ece);
err = ib_copy_to_udata(udata, &resp, resp.response_length);
if (err) {
mlx5_core_destroy_dct(dev, &qp->dct.mdct);
@@ -3949,11 +4210,11 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
int attr_mask, struct ib_udata *udata)
{
struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
+ struct mlx5_ib_modify_qp_resp resp = {};
struct mlx5_ib_qp *qp = to_mqp(ibqp);
struct mlx5_ib_modify_qp ucmd = {};
enum ib_qp_type qp_type;
enum ib_qp_state cur_state, new_state;
- size_t required_cmd_sz;
int err = -EINVAL;
int port;
@@ -3961,9 +4222,7 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
return -ENOSYS;
if (udata && udata->inlen) {
- required_cmd_sz = offsetof(typeof(ucmd), reserved) +
- sizeof(ucmd.reserved);
- if (udata->inlen < required_cmd_sz)
+ if (udata->inlen < offsetofend(typeof(ucmd), ece_options))
return -EINVAL;
if (udata->inlen > sizeof(ucmd) &&
@@ -3976,23 +4235,20 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
return -EFAULT;
if (ucmd.comp_mask ||
- memchr_inv(&ucmd.reserved, 0, sizeof(ucmd.reserved)) ||
memchr_inv(&ucmd.burst_info.reserved, 0,
sizeof(ucmd.burst_info.reserved)))
return -EOPNOTSUPP;
+
}
if (unlikely(ibqp->qp_type == IB_QPT_GSI))
return mlx5_ib_gsi_modify_qp(ibqp, attr, attr_mask);
- if (ibqp->qp_type == IB_QPT_DRIVER)
- qp_type = qp->qp_sub_type;
- else
- qp_type = (unlikely(ibqp->qp_type == MLX5_IB_QPT_HW_GSI)) ?
- IB_QPT_GSI : ibqp->qp_type;
+ qp_type = (unlikely(ibqp->qp_type == MLX5_IB_QPT_HW_GSI)) ? IB_QPT_GSI :
+ qp->type;
if (qp_type == MLX5_IB_QPT_DCT)
- return mlx5_ib_modify_dct(ibqp, attr, attr_mask, udata);
+ return mlx5_ib_modify_dct(ibqp, attr, attr_mask, &ucmd, udata);
mutex_lock(&qp->mutex);
@@ -4003,7 +4259,7 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
port = attr_mask & IB_QP_PORT ? attr->port_num : qp->port;
}
- if (qp->flags & MLX5_IB_QP_UNDERLAY) {
+ if (qp->flags & IB_QP_CREATE_SOURCE_QPN) {
if (attr_mask & ~(IB_QP_STATE | IB_QP_CUR_STATE)) {
mlx5_ib_dbg(dev, "invalid attr_mask 0x%x when underlay QP is used\n",
attr_mask);
@@ -4063,1439 +4319,19 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
}
err = __mlx5_ib_modify_qp(ibqp, attr, attr_mask, cur_state,
- new_state, &ucmd, udata);
-
-out:
- mutex_unlock(&qp->mutex);
- return err;
-}
-
-static void _handle_post_send_edge(struct mlx5_ib_wq *sq, void **seg,
- u32 wqe_sz, void **cur_edge)
-{
- u32 idx;
-
- idx = (sq->cur_post + (wqe_sz >> 2)) & (sq->wqe_cnt - 1);
- *cur_edge = get_sq_edge(sq, idx);
-
- *seg = mlx5_frag_buf_get_wqe(&sq->fbc, idx);
-}
-
-/* handle_post_send_edge - Check if we get to SQ edge. If yes, update to the
- * next nearby edge and get new address translation for current WQE position.
- * @sq - SQ buffer.
- * @seg: Current WQE position (16B aligned).
- * @wqe_sz: Total current WQE size [16B].
- * @cur_edge: Updated current edge.
- */
-static inline void handle_post_send_edge(struct mlx5_ib_wq *sq, void **seg,
- u32 wqe_sz, void **cur_edge)
-{
- if (likely(*seg != *cur_edge))
- return;
-
- _handle_post_send_edge(sq, seg, wqe_sz, cur_edge);
-}
-
-/* memcpy_send_wqe - copy data from src to WQE and update the relevant WQ's
- * pointers. At the end @seg is aligned to 16B regardless the copied size.
- * @sq - SQ buffer.
- * @cur_edge: Updated current edge.
- * @seg: Current WQE position (16B aligned).
- * @wqe_sz: Total current WQE size [16B].
- * @src: Pointer to copy from.
- * @n: Number of bytes to copy.
- */
-static inline void memcpy_send_wqe(struct mlx5_ib_wq *sq, void **cur_edge,
- void **seg, u32 *wqe_sz, const void *src,
- size_t n)
-{
- while (likely(n)) {
- size_t leftlen = *cur_edge - *seg;
- size_t copysz = min_t(size_t, leftlen, n);
- size_t stride;
-
- memcpy(*seg, src, copysz);
-
- n -= copysz;
- src += copysz;
- stride = !n ? ALIGN(copysz, 16) : copysz;
- *seg += stride;
- *wqe_sz += stride >> 4;
- handle_post_send_edge(sq, seg, *wqe_sz, cur_edge);
- }
-}
-
-static int mlx5_wq_overflow(struct mlx5_ib_wq *wq, int nreq, struct ib_cq *ib_cq)
-{
- struct mlx5_ib_cq *cq;
- unsigned cur;
-
- cur = wq->head - wq->tail;
- if (likely(cur + nreq < wq->max_post))
- return 0;
-
- cq = to_mcq(ib_cq);
- spin_lock(&cq->lock);
- cur = wq->head - wq->tail;
- spin_unlock(&cq->lock);
-
- return cur + nreq >= wq->max_post;
-}
-
-static __always_inline void set_raddr_seg(struct mlx5_wqe_raddr_seg *rseg,
- u64 remote_addr, u32 rkey)
-{
- rseg->raddr = cpu_to_be64(remote_addr);
- rseg->rkey = cpu_to_be32(rkey);
- rseg->reserved = 0;
-}
-
-static void set_eth_seg(const struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
- void **seg, int *size, void **cur_edge)
-{
- struct mlx5_wqe_eth_seg *eseg = *seg;
-
- memset(eseg, 0, sizeof(struct mlx5_wqe_eth_seg));
-
- if (wr->send_flags & IB_SEND_IP_CSUM)
- eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM |
- MLX5_ETH_WQE_L4_CSUM;
-
- if (wr->opcode == IB_WR_LSO) {
- struct ib_ud_wr *ud_wr = container_of(wr, struct ib_ud_wr, wr);
- size_t left, copysz;
- void *pdata = ud_wr->header;
- size_t stride;
-
- left = ud_wr->hlen;
- eseg->mss = cpu_to_be16(ud_wr->mss);
- eseg->inline_hdr.sz = cpu_to_be16(left);
-
- /* memcpy_send_wqe should get a 16B align address. Hence, we
- * first copy up to the current edge and then, if needed,
- * fall-through to memcpy_send_wqe.
- */
- copysz = min_t(u64, *cur_edge - (void *)eseg->inline_hdr.start,
- left);
- memcpy(eseg->inline_hdr.start, pdata, copysz);
- stride = ALIGN(sizeof(struct mlx5_wqe_eth_seg) -
- sizeof(eseg->inline_hdr.start) + copysz, 16);
- *size += stride / 16;
- *seg += stride;
-
- if (copysz < left) {
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
- left -= copysz;
- pdata += copysz;
- memcpy_send_wqe(&qp->sq, cur_edge, seg, size, pdata,
- left);
- }
-
- return;
- }
-
- *seg += sizeof(struct mlx5_wqe_eth_seg);
- *size += sizeof(struct mlx5_wqe_eth_seg) / 16;
-}
-
-static void set_datagram_seg(struct mlx5_wqe_datagram_seg *dseg,
- const struct ib_send_wr *wr)
-{
- memcpy(&dseg->av, &to_mah(ud_wr(wr)->ah)->av, sizeof(struct mlx5_av));
- dseg->av.dqp_dct = cpu_to_be32(ud_wr(wr)->remote_qpn | MLX5_EXTENDED_UD_AV);
- dseg->av.key.qkey.qkey = cpu_to_be32(ud_wr(wr)->remote_qkey);
-}
-
-static void set_data_ptr_seg(struct mlx5_wqe_data_seg *dseg, struct ib_sge *sg)
-{
- dseg->byte_count = cpu_to_be32(sg->length);
- dseg->lkey = cpu_to_be32(sg->lkey);
- dseg->addr = cpu_to_be64(sg->addr);
-}
-
-static u64 get_xlt_octo(u64 bytes)
-{
- return ALIGN(bytes, MLX5_IB_UMR_XLT_ALIGNMENT) /
- MLX5_IB_UMR_OCTOWORD;
-}
-
-static __be64 frwr_mkey_mask(bool atomic)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_LEN |
- MLX5_MKEY_MASK_PAGE_SIZE |
- MLX5_MKEY_MASK_START_ADDR |
- MLX5_MKEY_MASK_EN_RINVAL |
- MLX5_MKEY_MASK_KEY |
- MLX5_MKEY_MASK_LR |
- MLX5_MKEY_MASK_LW |
- MLX5_MKEY_MASK_RR |
- MLX5_MKEY_MASK_RW |
- MLX5_MKEY_MASK_SMALL_FENCE |
- MLX5_MKEY_MASK_FREE;
-
- if (atomic)
- result |= MLX5_MKEY_MASK_A;
-
- return cpu_to_be64(result);
-}
-
-static __be64 sig_mkey_mask(void)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_LEN |
- MLX5_MKEY_MASK_PAGE_SIZE |
- MLX5_MKEY_MASK_START_ADDR |
- MLX5_MKEY_MASK_EN_SIGERR |
- MLX5_MKEY_MASK_EN_RINVAL |
- MLX5_MKEY_MASK_KEY |
- MLX5_MKEY_MASK_LR |
- MLX5_MKEY_MASK_LW |
- MLX5_MKEY_MASK_RR |
- MLX5_MKEY_MASK_RW |
- MLX5_MKEY_MASK_SMALL_FENCE |
- MLX5_MKEY_MASK_FREE |
- MLX5_MKEY_MASK_BSF_EN;
-
- return cpu_to_be64(result);
-}
-
-static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
- struct mlx5_ib_mr *mr, u8 flags, bool atomic)
-{
- int size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
-
- memset(umr, 0, sizeof(*umr));
-
- umr->flags = flags;
- umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
- umr->mkey_mask = frwr_mkey_mask(atomic);
-}
-
-static void set_linv_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr)
-{
- memset(umr, 0, sizeof(*umr));
- umr->mkey_mask = cpu_to_be64(MLX5_MKEY_MASK_FREE);
- umr->flags = MLX5_UMR_INLINE;
-}
-
-static __be64 get_umr_enable_mr_mask(void)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_KEY |
- MLX5_MKEY_MASK_FREE;
-
- return cpu_to_be64(result);
-}
-
-static __be64 get_umr_disable_mr_mask(void)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_FREE;
-
- return cpu_to_be64(result);
-}
-
-static __be64 get_umr_update_translation_mask(void)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_LEN |
- MLX5_MKEY_MASK_PAGE_SIZE |
- MLX5_MKEY_MASK_START_ADDR;
-
- return cpu_to_be64(result);
-}
-
-static __be64 get_umr_update_access_mask(int atomic)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_LR |
- MLX5_MKEY_MASK_LW |
- MLX5_MKEY_MASK_RR |
- MLX5_MKEY_MASK_RW;
-
- if (atomic)
- result |= MLX5_MKEY_MASK_A;
-
- return cpu_to_be64(result);
-}
-
-static __be64 get_umr_update_pd_mask(void)
-{
- u64 result;
-
- result = MLX5_MKEY_MASK_PD;
-
- return cpu_to_be64(result);
-}
-
-static int umr_check_mkey_mask(struct mlx5_ib_dev *dev, u64 mask)
-{
- if ((mask & MLX5_MKEY_MASK_PAGE_SIZE &&
- MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled)) ||
- (mask & MLX5_MKEY_MASK_A &&
- MLX5_CAP_GEN(dev->mdev, umr_modify_atomic_disabled)))
- return -EPERM;
- return 0;
-}
-
-static int set_reg_umr_segment(struct mlx5_ib_dev *dev,
- struct mlx5_wqe_umr_ctrl_seg *umr,
- const struct ib_send_wr *wr, int atomic)
-{
- const struct mlx5_umr_wr *umrwr = umr_wr(wr);
-
- memset(umr, 0, sizeof(*umr));
-
- if (!umrwr->ignore_free_state) {
- if (wr->send_flags & MLX5_IB_SEND_UMR_FAIL_IF_FREE)
- /* fail if free */
- umr->flags = MLX5_UMR_CHECK_FREE;
- else
- /* fail if not free */
- umr->flags = MLX5_UMR_CHECK_NOT_FREE;
- }
-
- umr->xlt_octowords = cpu_to_be16(get_xlt_octo(umrwr->xlt_size));
- if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_XLT) {
- u64 offset = get_xlt_octo(umrwr->offset);
-
- umr->xlt_offset = cpu_to_be16(offset & 0xffff);
- umr->xlt_offset_47_16 = cpu_to_be32(offset >> 16);
- umr->flags |= MLX5_UMR_TRANSLATION_OFFSET_EN;
- }
- if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION)
- umr->mkey_mask |= get_umr_update_translation_mask();
- if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS) {
- umr->mkey_mask |= get_umr_update_access_mask(atomic);
- umr->mkey_mask |= get_umr_update_pd_mask();
- }
- if (wr->send_flags & MLX5_IB_SEND_UMR_ENABLE_MR)
- umr->mkey_mask |= get_umr_enable_mr_mask();
- if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR)
- umr->mkey_mask |= get_umr_disable_mr_mask();
-
- if (!wr->num_sge)
- umr->flags |= MLX5_UMR_INLINE;
-
- return umr_check_mkey_mask(dev, be64_to_cpu(umr->mkey_mask));
-}
-
-static u8 get_umr_flags(int acc)
-{
- return (acc & IB_ACCESS_REMOTE_ATOMIC ? MLX5_PERM_ATOMIC : 0) |
- (acc & IB_ACCESS_REMOTE_WRITE ? MLX5_PERM_REMOTE_WRITE : 0) |
- (acc & IB_ACCESS_REMOTE_READ ? MLX5_PERM_REMOTE_READ : 0) |
- (acc & IB_ACCESS_LOCAL_WRITE ? MLX5_PERM_LOCAL_WRITE : 0) |
- MLX5_PERM_LOCAL_READ | MLX5_PERM_UMR_EN;
-}
-
-static void set_reg_mkey_seg(struct mlx5_mkey_seg *seg,
- struct mlx5_ib_mr *mr,
- u32 key, int access)
-{
- int ndescs = ALIGN(mr->ndescs + mr->meta_ndescs, 8) >> 1;
-
- memset(seg, 0, sizeof(*seg));
-
- if (mr->access_mode == MLX5_MKC_ACCESS_MODE_MTT)
- seg->log2_page_size = ilog2(mr->ibmr.page_size);
- else if (mr->access_mode == MLX5_MKC_ACCESS_MODE_KLMS)
- /* KLMs take twice the size of MTTs */
- ndescs *= 2;
-
- seg->flags = get_umr_flags(access) | mr->access_mode;
- seg->qpn_mkey7_0 = cpu_to_be32((key & 0xff) | 0xffffff00);
- seg->flags_pd = cpu_to_be32(MLX5_MKEY_REMOTE_INVAL);
- seg->start_addr = cpu_to_be64(mr->ibmr.iova);
- seg->len = cpu_to_be64(mr->ibmr.length);
- seg->xlt_oct_size = cpu_to_be32(ndescs);
-}
-
-static void set_linv_mkey_seg(struct mlx5_mkey_seg *seg)
-{
- memset(seg, 0, sizeof(*seg));
- seg->status = MLX5_MKEY_STATUS_FREE;
-}
-
-static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg,
- const struct ib_send_wr *wr)
-{
- const struct mlx5_umr_wr *umrwr = umr_wr(wr);
-
- memset(seg, 0, sizeof(*seg));
- if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR)
- seg->status = MLX5_MKEY_STATUS_FREE;
-
- seg->flags = convert_access(umrwr->access_flags);
- if (umrwr->pd)
- seg->flags_pd = cpu_to_be32(to_mpd(umrwr->pd)->pdn);
- if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION &&
- !umrwr->length)
- seg->flags_pd |= cpu_to_be32(MLX5_MKEY_LEN64);
-
- seg->start_addr = cpu_to_be64(umrwr->virt_addr);
- seg->len = cpu_to_be64(umrwr->length);
- seg->log2_page_size = umrwr->page_shift;
- seg->qpn_mkey7_0 = cpu_to_be32(0xffffff00 |
- mlx5_mkey_variant(umrwr->mkey));
-}
-
-static void set_reg_data_seg(struct mlx5_wqe_data_seg *dseg,
- struct mlx5_ib_mr *mr,
- struct mlx5_ib_pd *pd)
-{
- int bcount = mr->desc_size * (mr->ndescs + mr->meta_ndescs);
-
- dseg->addr = cpu_to_be64(mr->desc_map);
- dseg->byte_count = cpu_to_be32(ALIGN(bcount, 64));
- dseg->lkey = cpu_to_be32(pd->ibpd.local_dma_lkey);
-}
-
-static __be32 send_ieth(const struct ib_send_wr *wr)
-{
- switch (wr->opcode) {
- case IB_WR_SEND_WITH_IMM:
- case IB_WR_RDMA_WRITE_WITH_IMM:
- return wr->ex.imm_data;
-
- case IB_WR_SEND_WITH_INV:
- return cpu_to_be32(wr->ex.invalidate_rkey);
-
- default:
- return 0;
- }
-}
-
-static u8 calc_sig(void *wqe, int size)
-{
- u8 *p = wqe;
- u8 res = 0;
- int i;
-
- for (i = 0; i < size; i++)
- res ^= p[i];
-
- return ~res;
-}
-
-static u8 wq_sig(void *wqe)
-{
- return calc_sig(wqe, (*((u8 *)wqe + 8) & 0x3f) << 4);
-}
-
-static int set_data_inl_seg(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
- void **wqe, int *wqe_sz, void **cur_edge)
-{
- struct mlx5_wqe_inline_seg *seg;
- size_t offset;
- int inl = 0;
- int i;
-
- seg = *wqe;
- *wqe += sizeof(*seg);
- offset = sizeof(*seg);
-
- for (i = 0; i < wr->num_sge; i++) {
- size_t len = wr->sg_list[i].length;
- void *addr = (void *)(unsigned long)(wr->sg_list[i].addr);
-
- inl += len;
-
- if (unlikely(inl > qp->max_inline_data))
- return -ENOMEM;
-
- while (likely(len)) {
- size_t leftlen;
- size_t copysz;
+ new_state, &ucmd, &resp, udata);
- handle_post_send_edge(&qp->sq, wqe,
- *wqe_sz + (offset >> 4),
- cur_edge);
-
- leftlen = *cur_edge - *wqe;
- copysz = min_t(size_t, leftlen, len);
-
- memcpy(*wqe, addr, copysz);
- len -= copysz;
- addr += copysz;
- *wqe += copysz;
- offset += copysz;
- }
- }
-
- seg->byte_count = cpu_to_be32(inl | MLX5_INLINE_SEG);
-
- *wqe_sz += ALIGN(inl + sizeof(seg->byte_count), 16) / 16;
-
- return 0;
-}
-
-static u16 prot_field_size(enum ib_signature_type type)
-{
- switch (type) {
- case IB_SIG_TYPE_T10_DIF:
- return MLX5_DIF_SIZE;
- default:
- return 0;
- }
-}
-
-static u8 bs_selector(int block_size)
-{
- switch (block_size) {
- case 512: return 0x1;
- case 520: return 0x2;
- case 4096: return 0x3;
- case 4160: return 0x4;
- case 1073741824: return 0x5;
- default: return 0;
- }
-}
-
-static void mlx5_fill_inl_bsf(struct ib_sig_domain *domain,
- struct mlx5_bsf_inl *inl)
-{
- /* Valid inline section and allow BSF refresh */
- inl->vld_refresh = cpu_to_be16(MLX5_BSF_INL_VALID |
- MLX5_BSF_REFRESH_DIF);
- inl->dif_apptag = cpu_to_be16(domain->sig.dif.app_tag);
- inl->dif_reftag = cpu_to_be32(domain->sig.dif.ref_tag);
- /* repeating block */
- inl->rp_inv_seed = MLX5_BSF_REPEAT_BLOCK;
- inl->sig_type = domain->sig.dif.bg_type == IB_T10DIF_CRC ?
- MLX5_DIF_CRC : MLX5_DIF_IPCS;
-
- if (domain->sig.dif.ref_remap)
- inl->dif_inc_ref_guard_check |= MLX5_BSF_INC_REFTAG;
-
- if (domain->sig.dif.app_escape) {
- if (domain->sig.dif.ref_escape)
- inl->dif_inc_ref_guard_check |= MLX5_BSF_APPREF_ESCAPE;
- else
- inl->dif_inc_ref_guard_check |= MLX5_BSF_APPTAG_ESCAPE;
- }
-
- inl->dif_app_bitmask_check =
- cpu_to_be16(domain->sig.dif.apptag_check_mask);
-}
-
-static int mlx5_set_bsf(struct ib_mr *sig_mr,
- struct ib_sig_attrs *sig_attrs,
- struct mlx5_bsf *bsf, u32 data_size)
-{
- struct mlx5_core_sig_ctx *msig = to_mmr(sig_mr)->sig;
- struct mlx5_bsf_basic *basic = &bsf->basic;
- struct ib_sig_domain *mem = &sig_attrs->mem;
- struct ib_sig_domain *wire = &sig_attrs->wire;
-
- memset(bsf, 0, sizeof(*bsf));
-
- /* Basic + Extended + Inline */
- basic->bsf_size_sbs = 1 << 7;
- /* Input domain check byte mask */
- basic->check_byte_mask = sig_attrs->check_mask;
- basic->raw_data_size = cpu_to_be32(data_size);
-
- /* Memory domain */
- switch (sig_attrs->mem.sig_type) {
- case IB_SIG_TYPE_NONE:
- break;
- case IB_SIG_TYPE_T10_DIF:
- basic->mem.bs_selector = bs_selector(mem->sig.dif.pi_interval);
- basic->m_bfs_psv = cpu_to_be32(msig->psv_memory.psv_idx);
- mlx5_fill_inl_bsf(mem, &bsf->m_inl);
- break;
- default:
- return -EINVAL;
- }
-
- /* Wire domain */
- switch (sig_attrs->wire.sig_type) {
- case IB_SIG_TYPE_NONE:
- break;
- case IB_SIG_TYPE_T10_DIF:
- if (mem->sig.dif.pi_interval == wire->sig.dif.pi_interval &&
- mem->sig_type == wire->sig_type) {
- /* Same block structure */
- basic->bsf_size_sbs |= 1 << 4;
- if (mem->sig.dif.bg_type == wire->sig.dif.bg_type)
- basic->wire.copy_byte_mask |= MLX5_CPY_GRD_MASK;
- if (mem->sig.dif.app_tag == wire->sig.dif.app_tag)
- basic->wire.copy_byte_mask |= MLX5_CPY_APP_MASK;
- if (mem->sig.dif.ref_tag == wire->sig.dif.ref_tag)
- basic->wire.copy_byte_mask |= MLX5_CPY_REF_MASK;
- } else
- basic->wire.bs_selector = bs_selector(wire->sig.dif.pi_interval);
-
- basic->w_bfs_psv = cpu_to_be32(msig->psv_wire.psv_idx);
- mlx5_fill_inl_bsf(wire, &bsf->w_inl);
- break;
- default:
- return -EINVAL;
- }
-
- return 0;
-}
-
-static int set_sig_data_segment(const struct ib_send_wr *send_wr,
- struct ib_mr *sig_mr,
- struct ib_sig_attrs *sig_attrs,
- struct mlx5_ib_qp *qp, void **seg, int *size,
- void **cur_edge)
-{
- struct mlx5_bsf *bsf;
- u32 data_len;
- u32 data_key;
- u64 data_va;
- u32 prot_len = 0;
- u32 prot_key = 0;
- u64 prot_va = 0;
- bool prot = false;
- int ret;
- int wqe_size;
- struct mlx5_ib_mr *mr = to_mmr(sig_mr);
- struct mlx5_ib_mr *pi_mr = mr->pi_mr;
-
- data_len = pi_mr->data_length;
- data_key = pi_mr->ibmr.lkey;
- data_va = pi_mr->data_iova;
- if (pi_mr->meta_ndescs) {
- prot_len = pi_mr->meta_length;
- prot_key = pi_mr->ibmr.lkey;
- prot_va = pi_mr->pi_iova;
- prot = true;
- }
-
- if (!prot || (data_key == prot_key && data_va == prot_va &&
- data_len == prot_len)) {
- /**
- * Source domain doesn't contain signature information
- * or data and protection are interleaved in memory.
- * So need construct:
- * ------------------
- * | data_klm |
- * ------------------
- * | BSF |
- * ------------------
- **/
- struct mlx5_klm *data_klm = *seg;
-
- data_klm->bcount = cpu_to_be32(data_len);
- data_klm->key = cpu_to_be32(data_key);
- data_klm->va = cpu_to_be64(data_va);
- wqe_size = ALIGN(sizeof(*data_klm), 64);
- } else {
- /**
- * Source domain contains signature information
- * So need construct a strided block format:
- * ---------------------------
- * | stride_block_ctrl |
- * ---------------------------
- * | data_klm |
- * ---------------------------
- * | prot_klm |
- * ---------------------------
- * | BSF |
- * ---------------------------
- **/
- struct mlx5_stride_block_ctrl_seg *sblock_ctrl;
- struct mlx5_stride_block_entry *data_sentry;
- struct mlx5_stride_block_entry *prot_sentry;
- u16 block_size = sig_attrs->mem.sig.dif.pi_interval;
- int prot_size;
-
- sblock_ctrl = *seg;
- data_sentry = (void *)sblock_ctrl + sizeof(*sblock_ctrl);
- prot_sentry = (void *)data_sentry + sizeof(*data_sentry);
-
- prot_size = prot_field_size(sig_attrs->mem.sig_type);
- if (!prot_size) {
- pr_err("Bad block size given: %u\n", block_size);
- return -EINVAL;
- }
- sblock_ctrl->bcount_per_cycle = cpu_to_be32(block_size +
- prot_size);
- sblock_ctrl->op = cpu_to_be32(MLX5_STRIDE_BLOCK_OP);
- sblock_ctrl->repeat_count = cpu_to_be32(data_len / block_size);
- sblock_ctrl->num_entries = cpu_to_be16(2);
-
- data_sentry->bcount = cpu_to_be16(block_size);
- data_sentry->key = cpu_to_be32(data_key);
- data_sentry->va = cpu_to_be64(data_va);
- data_sentry->stride = cpu_to_be16(block_size);
-
- prot_sentry->bcount = cpu_to_be16(prot_size);
- prot_sentry->key = cpu_to_be32(prot_key);
- prot_sentry->va = cpu_to_be64(prot_va);
- prot_sentry->stride = cpu_to_be16(prot_size);
-
- wqe_size = ALIGN(sizeof(*sblock_ctrl) + sizeof(*data_sentry) +
- sizeof(*prot_sentry), 64);
- }
-
- *seg += wqe_size;
- *size += wqe_size / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- bsf = *seg;
- ret = mlx5_set_bsf(sig_mr, sig_attrs, bsf, data_len);
- if (ret)
- return -EINVAL;
-
- *seg += sizeof(*bsf);
- *size += sizeof(*bsf) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- return 0;
-}
-
-static void set_sig_mkey_segment(struct mlx5_mkey_seg *seg,
- struct ib_mr *sig_mr, int access_flags,
- u32 size, u32 length, u32 pdn)
-{
- u32 sig_key = sig_mr->rkey;
- u8 sigerr = to_mmr(sig_mr)->sig->sigerr_count & 1;
-
- memset(seg, 0, sizeof(*seg));
-
- seg->flags = get_umr_flags(access_flags) | MLX5_MKC_ACCESS_MODE_KLMS;
- seg->qpn_mkey7_0 = cpu_to_be32((sig_key & 0xff) | 0xffffff00);
- seg->flags_pd = cpu_to_be32(MLX5_MKEY_REMOTE_INVAL | sigerr << 26 |
- MLX5_MKEY_BSF_EN | pdn);
- seg->len = cpu_to_be64(length);
- seg->xlt_oct_size = cpu_to_be32(get_xlt_octo(size));
- seg->bsfs_octo_size = cpu_to_be32(MLX5_MKEY_BSF_OCTO_SIZE);
-}
-
-static void set_sig_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr,
- u32 size)
-{
- memset(umr, 0, sizeof(*umr));
-
- umr->flags = MLX5_FLAGS_INLINE | MLX5_FLAGS_CHECK_FREE;
- umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
- umr->bsf_octowords = cpu_to_be16(MLX5_MKEY_BSF_OCTO_SIZE);
- umr->mkey_mask = sig_mkey_mask();
-}
-
-static int set_pi_umr_wr(const struct ib_send_wr *send_wr,
- struct mlx5_ib_qp *qp, void **seg, int *size,
- void **cur_edge)
-{
- const struct ib_reg_wr *wr = reg_wr(send_wr);
- struct mlx5_ib_mr *sig_mr = to_mmr(wr->mr);
- struct mlx5_ib_mr *pi_mr = sig_mr->pi_mr;
- struct ib_sig_attrs *sig_attrs = sig_mr->ibmr.sig_attrs;
- u32 pdn = get_pd(qp)->pdn;
- u32 xlt_size;
- int region_len, ret;
-
- if (unlikely(send_wr->num_sge != 0) ||
- unlikely(wr->access & IB_ACCESS_REMOTE_ATOMIC) ||
- unlikely(!sig_mr->sig) || unlikely(!qp->ibqp.integrity_en) ||
- unlikely(!sig_mr->sig->sig_status_checked))
- return -EINVAL;
-
- /* length of the protected region, data + protection */
- region_len = pi_mr->ibmr.length;
-
- /**
- * KLM octoword size - if protection was provided
- * then we use strided block format (3 octowords),
- * else we use single KLM (1 octoword)
- **/
- if (sig_attrs->mem.sig_type != IB_SIG_TYPE_NONE)
- xlt_size = 0x30;
- else
- xlt_size = sizeof(struct mlx5_klm);
-
- set_sig_umr_segment(*seg, xlt_size);
- *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
- *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- set_sig_mkey_segment(*seg, wr->mr, wr->access, xlt_size, region_len,
- pdn);
- *seg += sizeof(struct mlx5_mkey_seg);
- *size += sizeof(struct mlx5_mkey_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- ret = set_sig_data_segment(send_wr, wr->mr, sig_attrs, qp, seg, size,
- cur_edge);
- if (ret)
- return ret;
-
- sig_mr->sig->sig_status_checked = false;
- return 0;
-}
-
-static int set_psv_wr(struct ib_sig_domain *domain,
- u32 psv_idx, void **seg, int *size)
-{
- struct mlx5_seg_set_psv *psv_seg = *seg;
-
- memset(psv_seg, 0, sizeof(*psv_seg));
- psv_seg->psv_num = cpu_to_be32(psv_idx);
- switch (domain->sig_type) {
- case IB_SIG_TYPE_NONE:
- break;
- case IB_SIG_TYPE_T10_DIF:
- psv_seg->transient_sig = cpu_to_be32(domain->sig.dif.bg << 16 |
- domain->sig.dif.app_tag);
- psv_seg->ref_tag = cpu_to_be32(domain->sig.dif.ref_tag);
- break;
- default:
- pr_err("Bad signature type (%d) is given.\n",
- domain->sig_type);
- return -EINVAL;
- }
-
- *seg += sizeof(*psv_seg);
- *size += sizeof(*psv_seg) / 16;
-
- return 0;
-}
-
-static int set_reg_wr(struct mlx5_ib_qp *qp,
- const struct ib_reg_wr *wr,
- void **seg, int *size, void **cur_edge,
- bool check_not_free)
-{
- struct mlx5_ib_mr *mr = to_mmr(wr->mr);
- struct mlx5_ib_pd *pd = to_mpd(qp->ibqp.pd);
- struct mlx5_ib_dev *dev = to_mdev(pd->ibpd.device);
- int mr_list_size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
- bool umr_inline = mr_list_size <= MLX5_IB_SQ_UMR_INLINE_THRESHOLD;
- bool atomic = wr->access & IB_ACCESS_REMOTE_ATOMIC;
- u8 flags = 0;
-
- if (!mlx5_ib_can_use_umr(dev, atomic, wr->access)) {
- mlx5_ib_warn(to_mdev(qp->ibqp.device),
- "Fast update of %s for MR is disabled\n",
- (MLX5_CAP_GEN(dev->mdev,
- umr_modify_entity_size_disabled)) ?
- "entity size" :
- "atomic access");
- return -EINVAL;
- }
-
- if (unlikely(wr->wr.send_flags & IB_SEND_INLINE)) {
- mlx5_ib_warn(to_mdev(qp->ibqp.device),
- "Invalid IB_SEND_INLINE send flag\n");
- return -EINVAL;
- }
-
- if (check_not_free)
- flags |= MLX5_UMR_CHECK_NOT_FREE;
- if (umr_inline)
- flags |= MLX5_UMR_INLINE;
-
- set_reg_umr_seg(*seg, mr, flags, atomic);
- *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
- *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- set_reg_mkey_seg(*seg, mr, wr->key, wr->access);
- *seg += sizeof(struct mlx5_mkey_seg);
- *size += sizeof(struct mlx5_mkey_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-
- if (umr_inline) {
- memcpy_send_wqe(&qp->sq, cur_edge, seg, size, mr->descs,
- mr_list_size);
- *size = ALIGN(*size, MLX5_SEND_WQE_BB >> 4);
- } else {
- set_reg_data_seg(*seg, mr, pd);
- *seg += sizeof(struct mlx5_wqe_data_seg);
- *size += (sizeof(struct mlx5_wqe_data_seg) / 16);
- }
- return 0;
-}
-
-static void set_linv_wr(struct mlx5_ib_qp *qp, void **seg, int *size,
- void **cur_edge)
-{
- set_linv_umr_seg(*seg);
- *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
- *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
- set_linv_mkey_seg(*seg);
- *seg += sizeof(struct mlx5_mkey_seg);
- *size += sizeof(struct mlx5_mkey_seg) / 16;
- handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
-}
-
-static void dump_wqe(struct mlx5_ib_qp *qp, u32 idx, int size_16)
-{
- __be32 *p = NULL;
- int i, j;
-
- pr_debug("dump WQE index %u:\n", idx);
- for (i = 0, j = 0; i < size_16 * 4; i += 4, j += 4) {
- if ((i & 0xf) == 0) {
- p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, idx);
- pr_debug("WQBB at %p:\n", (void *)p);
- j = 0;
- idx = (idx + 1) & (qp->sq.wqe_cnt - 1);
- }
- pr_debug("%08x %08x %08x %08x\n", be32_to_cpu(p[j]),
- be32_to_cpu(p[j + 1]), be32_to_cpu(p[j + 2]),
- be32_to_cpu(p[j + 3]));
- }
-}
-
-static int __begin_wqe(struct mlx5_ib_qp *qp, void **seg,
- struct mlx5_wqe_ctrl_seg **ctrl,
- const struct ib_send_wr *wr, unsigned int *idx,
- int *size, void **cur_edge, int nreq,
- bool send_signaled, bool solicited)
-{
- if (unlikely(mlx5_wq_overflow(&qp->sq, nreq, qp->ibqp.send_cq)))
- return -ENOMEM;
-
- *idx = qp->sq.cur_post & (qp->sq.wqe_cnt - 1);
- *seg = mlx5_frag_buf_get_wqe(&qp->sq.fbc, *idx);
- *ctrl = *seg;
- *(uint32_t *)(*seg + 8) = 0;
- (*ctrl)->imm = send_ieth(wr);
- (*ctrl)->fm_ce_se = qp->sq_signal_bits |
- (send_signaled ? MLX5_WQE_CTRL_CQ_UPDATE : 0) |
- (solicited ? MLX5_WQE_CTRL_SOLICITED : 0);
-
- *seg += sizeof(**ctrl);
- *size = sizeof(**ctrl) / 16;
- *cur_edge = qp->sq.cur_edge;
-
- return 0;
-}
-
-static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
- struct mlx5_wqe_ctrl_seg **ctrl,
- const struct ib_send_wr *wr, unsigned *idx,
- int *size, void **cur_edge, int nreq)
-{
- return __begin_wqe(qp, seg, ctrl, wr, idx, size, cur_edge, nreq,
- wr->send_flags & IB_SEND_SIGNALED,
- wr->send_flags & IB_SEND_SOLICITED);
-}
-
-static void finish_wqe(struct mlx5_ib_qp *qp,
- struct mlx5_wqe_ctrl_seg *ctrl,
- void *seg, u8 size, void *cur_edge,
- unsigned int idx, u64 wr_id, int nreq, u8 fence,
- u32 mlx5_opcode)
-{
- u8 opmod = 0;
-
- ctrl->opmod_idx_opcode = cpu_to_be32(((u32)(qp->sq.cur_post) << 8) |
- mlx5_opcode | ((u32)opmod << 24));
- ctrl->qpn_ds = cpu_to_be32(size | (qp->trans_qp.base.mqp.qpn << 8));
- ctrl->fm_ce_se |= fence;
- if (unlikely(qp->wq_sig))
- ctrl->signature = wq_sig(ctrl);
-
- qp->sq.wrid[idx] = wr_id;
- qp->sq.w_list[idx].opcode = mlx5_opcode;
- qp->sq.wqe_head[idx] = qp->sq.head + nreq;
- qp->sq.cur_post += DIV_ROUND_UP(size * 16, MLX5_SEND_WQE_BB);
- qp->sq.w_list[idx].next = qp->sq.cur_post;
-
- /* We save the edge which was possibly updated during the WQE
- * construction, into SQ's cache.
- */
- seg = PTR_ALIGN(seg, MLX5_SEND_WQE_BB);
- qp->sq.cur_edge = (unlikely(seg == cur_edge)) ?
- get_sq_edge(&qp->sq, qp->sq.cur_post &
- (qp->sq.wqe_cnt - 1)) :
- cur_edge;
-}
-
-static int _mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
- const struct ib_send_wr **bad_wr, bool drain)
-{
- struct mlx5_wqe_ctrl_seg *ctrl = NULL; /* compiler warning */
- struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
- struct mlx5_core_dev *mdev = dev->mdev;
- struct ib_reg_wr reg_pi_wr;
- struct mlx5_ib_qp *qp;
- struct mlx5_ib_mr *mr;
- struct mlx5_ib_mr *pi_mr;
- struct mlx5_ib_mr pa_pi_mr;
- struct ib_sig_attrs *sig_attrs;
- struct mlx5_wqe_xrc_seg *xrc;
- struct mlx5_bf *bf;
- void *cur_edge;
- int uninitialized_var(size);
- unsigned long flags;
- unsigned idx;
- int err = 0;
- int num_sge;
- void *seg;
- int nreq;
- int i;
- u8 next_fence = 0;
- u8 fence;
-
- if (unlikely(mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR &&
- !drain)) {
- *bad_wr = wr;
- return -EIO;
- }
-
- if (unlikely(ibqp->qp_type == IB_QPT_GSI))
- return mlx5_ib_gsi_post_send(ibqp, wr, bad_wr);
-
- qp = to_mqp(ibqp);
- bf = &qp->bf;
-
- spin_lock_irqsave(&qp->sq.lock, flags);
-
- for (nreq = 0; wr; nreq++, wr = wr->next) {
- if (unlikely(wr->opcode >= ARRAY_SIZE(mlx5_ib_opcode))) {
- mlx5_ib_warn(dev, "\n");
- err = -EINVAL;
- *bad_wr = wr;
- goto out;
- }
-
- num_sge = wr->num_sge;
- if (unlikely(num_sge > qp->sq.max_gs)) {
- mlx5_ib_warn(dev, "\n");
- err = -EINVAL;
- *bad_wr = wr;
- goto out;
- }
-
- err = begin_wqe(qp, &seg, &ctrl, wr, &idx, &size, &cur_edge,
- nreq);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- err = -ENOMEM;
- *bad_wr = wr;
- goto out;
- }
-
- if (wr->opcode == IB_WR_REG_MR ||
- wr->opcode == IB_WR_REG_MR_INTEGRITY) {
- fence = dev->umr_fence;
- next_fence = MLX5_FENCE_MODE_INITIATOR_SMALL;
- } else {
- if (wr->send_flags & IB_SEND_FENCE) {
- if (qp->next_fence)
- fence = MLX5_FENCE_MODE_SMALL_AND_FENCE;
- else
- fence = MLX5_FENCE_MODE_FENCE;
- } else {
- fence = qp->next_fence;
- }
- }
-
- switch (ibqp->qp_type) {
- case IB_QPT_XRC_INI:
- xrc = seg;
- seg += sizeof(*xrc);
- size += sizeof(*xrc) / 16;
- /* fall through */
- case IB_QPT_RC:
- switch (wr->opcode) {
- case IB_WR_RDMA_READ:
- case IB_WR_RDMA_WRITE:
- case IB_WR_RDMA_WRITE_WITH_IMM:
- set_raddr_seg(seg, rdma_wr(wr)->remote_addr,
- rdma_wr(wr)->rkey);
- seg += sizeof(struct mlx5_wqe_raddr_seg);
- size += sizeof(struct mlx5_wqe_raddr_seg) / 16;
- break;
-
- case IB_WR_ATOMIC_CMP_AND_SWP:
- case IB_WR_ATOMIC_FETCH_AND_ADD:
- case IB_WR_MASKED_ATOMIC_CMP_AND_SWP:
- mlx5_ib_warn(dev, "Atomic operations are not supported yet\n");
- err = -ENOSYS;
- *bad_wr = wr;
- goto out;
-
- case IB_WR_LOCAL_INV:
- qp->sq.wr_data[idx] = IB_WR_LOCAL_INV;
- ctrl->imm = cpu_to_be32(wr->ex.invalidate_rkey);
- set_linv_wr(qp, &seg, &size, &cur_edge);
- num_sge = 0;
- break;
-
- case IB_WR_REG_MR:
- qp->sq.wr_data[idx] = IB_WR_REG_MR;
- ctrl->imm = cpu_to_be32(reg_wr(wr)->key);
- err = set_reg_wr(qp, reg_wr(wr), &seg, &size,
- &cur_edge, true);
- if (err) {
- *bad_wr = wr;
- goto out;
- }
- num_sge = 0;
- break;
-
- case IB_WR_REG_MR_INTEGRITY:
- qp->sq.wr_data[idx] = IB_WR_REG_MR_INTEGRITY;
-
- mr = to_mmr(reg_wr(wr)->mr);
- pi_mr = mr->pi_mr;
-
- if (pi_mr) {
- memset(&reg_pi_wr, 0,
- sizeof(struct ib_reg_wr));
-
- reg_pi_wr.mr = &pi_mr->ibmr;
- reg_pi_wr.access = reg_wr(wr)->access;
- reg_pi_wr.key = pi_mr->ibmr.rkey;
-
- ctrl->imm = cpu_to_be32(reg_pi_wr.key);
- /* UMR for data + prot registration */
- err = set_reg_wr(qp, &reg_pi_wr, &seg,
- &size, &cur_edge,
- false);
- if (err) {
- *bad_wr = wr;
- goto out;
- }
- finish_wqe(qp, ctrl, seg, size,
- cur_edge, idx, wr->wr_id,
- nreq, fence,
- MLX5_OPCODE_UMR);
-
- err = begin_wqe(qp, &seg, &ctrl, wr,
- &idx, &size, &cur_edge,
- nreq);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- err = -ENOMEM;
- *bad_wr = wr;
- goto out;
- }
- } else {
- memset(&pa_pi_mr, 0,
- sizeof(struct mlx5_ib_mr));
- /* No UMR, use local_dma_lkey */
- pa_pi_mr.ibmr.lkey =
- mr->ibmr.pd->local_dma_lkey;
-
- pa_pi_mr.ndescs = mr->ndescs;
- pa_pi_mr.data_length = mr->data_length;
- pa_pi_mr.data_iova = mr->data_iova;
- if (mr->meta_ndescs) {
- pa_pi_mr.meta_ndescs =
- mr->meta_ndescs;
- pa_pi_mr.meta_length =
- mr->meta_length;
- pa_pi_mr.pi_iova = mr->pi_iova;
- }
-
- pa_pi_mr.ibmr.length = mr->ibmr.length;
- mr->pi_mr = &pa_pi_mr;
- }
- ctrl->imm = cpu_to_be32(mr->ibmr.rkey);
- /* UMR for sig MR */
- err = set_pi_umr_wr(wr, qp, &seg, &size,
- &cur_edge);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- *bad_wr = wr;
- goto out;
- }
- finish_wqe(qp, ctrl, seg, size, cur_edge, idx,
- wr->wr_id, nreq, fence,
- MLX5_OPCODE_UMR);
-
- /*
- * SET_PSV WQEs are not signaled and solicited
- * on error
- */
- sig_attrs = mr->ibmr.sig_attrs;
- err = __begin_wqe(qp, &seg, &ctrl, wr, &idx,
- &size, &cur_edge, nreq, false,
- true);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- err = -ENOMEM;
- *bad_wr = wr;
- goto out;
- }
- err = set_psv_wr(&sig_attrs->mem,
- mr->sig->psv_memory.psv_idx,
- &seg, &size);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- *bad_wr = wr;
- goto out;
- }
- finish_wqe(qp, ctrl, seg, size, cur_edge, idx,
- wr->wr_id, nreq, next_fence,
- MLX5_OPCODE_SET_PSV);
-
- err = __begin_wqe(qp, &seg, &ctrl, wr, &idx,
- &size, &cur_edge, nreq, false,
- true);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- err = -ENOMEM;
- *bad_wr = wr;
- goto out;
- }
- err = set_psv_wr(&sig_attrs->wire,
- mr->sig->psv_wire.psv_idx,
- &seg, &size);
- if (err) {
- mlx5_ib_warn(dev, "\n");
- *bad_wr = wr;
- goto out;
- }
- finish_wqe(qp, ctrl, seg, size, cur_edge, idx,
- wr->wr_id, nreq, next_fence,
- MLX5_OPCODE_SET_PSV);
-
- qp->next_fence =
- MLX5_FENCE_MODE_INITIATOR_SMALL;
- num_sge = 0;
- goto skip_psv;
-
- default:
- break;
- }
- break;
-
- case IB_QPT_UC:
- switch (wr->opcode) {
- case IB_WR_RDMA_WRITE:
- case IB_WR_RDMA_WRITE_WITH_IMM:
- set_raddr_seg(seg, rdma_wr(wr)->remote_addr,
- rdma_wr(wr)->rkey);
- seg += sizeof(struct mlx5_wqe_raddr_seg);
- size += sizeof(struct mlx5_wqe_raddr_seg) / 16;
- break;
-
- default:
- break;
- }
- break;
-
- case IB_QPT_SMI:
- if (unlikely(!mdev->port_caps[qp->port - 1].has_smi)) {
- mlx5_ib_warn(dev, "Send SMP MADs is not allowed\n");
- err = -EPERM;
- *bad_wr = wr;
- goto out;
- }
- /* fall through */
- case MLX5_IB_QPT_HW_GSI:
- set_datagram_seg(seg, wr);
- seg += sizeof(struct mlx5_wqe_datagram_seg);
- size += sizeof(struct mlx5_wqe_datagram_seg) / 16;
- handle_post_send_edge(&qp->sq, &seg, size, &cur_edge);
-
- break;
- case IB_QPT_UD:
- set_datagram_seg(seg, wr);
- seg += sizeof(struct mlx5_wqe_datagram_seg);
- size += sizeof(struct mlx5_wqe_datagram_seg) / 16;
- handle_post_send_edge(&qp->sq, &seg, size, &cur_edge);
-
- /* handle qp that supports ud offload */
- if (qp->flags & IB_QP_CREATE_IPOIB_UD_LSO) {
- struct mlx5_wqe_eth_pad *pad;
-
- pad = seg;
- memset(pad, 0, sizeof(struct mlx5_wqe_eth_pad));
- seg += sizeof(struct mlx5_wqe_eth_pad);
- size += sizeof(struct mlx5_wqe_eth_pad) / 16;
- set_eth_seg(wr, qp, &seg, &size, &cur_edge);
- handle_post_send_edge(&qp->sq, &seg, size,
- &cur_edge);
- }
- break;
- case MLX5_IB_QPT_REG_UMR:
- if (wr->opcode != MLX5_IB_WR_UMR) {
- err = -EINVAL;
- mlx5_ib_warn(dev, "bad opcode\n");
- goto out;
- }
- qp->sq.wr_data[idx] = MLX5_IB_WR_UMR;
- ctrl->imm = cpu_to_be32(umr_wr(wr)->mkey);
- err = set_reg_umr_segment(dev, seg, wr, !!(MLX5_CAP_GEN(mdev, atomic)));
- if (unlikely(err))
- goto out;
- seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
- size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
- handle_post_send_edge(&qp->sq, &seg, size, &cur_edge);
- set_reg_mkey_segment(seg, wr);
- seg += sizeof(struct mlx5_mkey_seg);
- size += sizeof(struct mlx5_mkey_seg) / 16;
- handle_post_send_edge(&qp->sq, &seg, size, &cur_edge);
- break;
-
- default:
- break;
- }
-
- if (wr->send_flags & IB_SEND_INLINE && num_sge) {
- err = set_data_inl_seg(qp, wr, &seg, &size, &cur_edge);
- if (unlikely(err)) {
- mlx5_ib_warn(dev, "\n");
- *bad_wr = wr;
- goto out;
- }
- } else {
- for (i = 0; i < num_sge; i++) {
- handle_post_send_edge(&qp->sq, &seg, size,
- &cur_edge);
- if (likely(wr->sg_list[i].length)) {
- set_data_ptr_seg
- ((struct mlx5_wqe_data_seg *)seg,
- wr->sg_list + i);
- size += sizeof(struct mlx5_wqe_data_seg) / 16;
- seg += sizeof(struct mlx5_wqe_data_seg);
- }
- }
- }
-
- qp->next_fence = next_fence;
- finish_wqe(qp, ctrl, seg, size, cur_edge, idx, wr->wr_id, nreq,
- fence, mlx5_ib_opcode[wr->opcode]);
-skip_psv:
- if (0)
- dump_wqe(qp, idx, size);
- }
-
-out:
- if (likely(nreq)) {
- qp->sq.head += nreq;
-
- /* Make sure that descriptors are written before
- * updating doorbell record and ringing the doorbell
- */
- wmb();
-
- qp->db.db[MLX5_SND_DBR] = cpu_to_be32(qp->sq.cur_post);
-
- /* Make sure doorbell record is visible to the HCA before
- * we hit doorbell */
- wmb();
-
- mlx5_write64((__be32 *)ctrl, bf->bfreg->map + bf->offset);
- /* Make sure doorbells don't leak out of SQ spinlock
- * and reach the HCA out of order.
- */
- bf->offset ^= bf->buf_size;
- }
-
- spin_unlock_irqrestore(&qp->sq.lock, flags);
-
- return err;
-}
-
-int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
- const struct ib_send_wr **bad_wr)
-{
- return _mlx5_ib_post_send(ibqp, wr, bad_wr, false);
-}
-
-static void set_sig_seg(struct mlx5_rwqe_sig *sig, int size)
-{
- sig->signature = calc_sig(sig, size);
-}
-
-static int _mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
- const struct ib_recv_wr **bad_wr, bool drain)
-{
- struct mlx5_ib_qp *qp = to_mqp(ibqp);
- struct mlx5_wqe_data_seg *scat;
- struct mlx5_rwqe_sig *sig;
- struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
- struct mlx5_core_dev *mdev = dev->mdev;
- unsigned long flags;
- int err = 0;
- int nreq;
- int ind;
- int i;
-
- if (unlikely(mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR &&
- !drain)) {
- *bad_wr = wr;
- return -EIO;
- }
-
- if (unlikely(ibqp->qp_type == IB_QPT_GSI))
- return mlx5_ib_gsi_post_recv(ibqp, wr, bad_wr);
-
- spin_lock_irqsave(&qp->rq.lock, flags);
-
- ind = qp->rq.head & (qp->rq.wqe_cnt - 1);
-
- for (nreq = 0; wr; nreq++, wr = wr->next) {
- if (mlx5_wq_overflow(&qp->rq, nreq, qp->ibqp.recv_cq)) {
- err = -ENOMEM;
- *bad_wr = wr;
- goto out;
- }
-
- if (unlikely(wr->num_sge > qp->rq.max_gs)) {
- err = -EINVAL;
- *bad_wr = wr;
- goto out;
- }
-
- scat = mlx5_frag_buf_get_wqe(&qp->rq.fbc, ind);
- if (qp->wq_sig)
- scat++;
-
- for (i = 0; i < wr->num_sge; i++)
- set_data_ptr_seg(scat + i, wr->sg_list + i);
-
- if (i < qp->rq.max_gs) {
- scat[i].byte_count = 0;
- scat[i].lkey = cpu_to_be32(MLX5_INVALID_LKEY);
- scat[i].addr = 0;
- }
-
- if (qp->wq_sig) {
- sig = (struct mlx5_rwqe_sig *)scat;
- set_sig_seg(sig, (qp->rq.max_gs + 1) << 2);
- }
-
- qp->rq.wrid[ind] = wr->wr_id;
-
- ind = (ind + 1) & (qp->rq.wqe_cnt - 1);
- }
+ /* resp.response_length is set in ECE supported flows only */
+ if (!err && resp.response_length &&
+ udata->outlen >= resp.response_length)
+ /* Return -EFAULT to the user and expect him to destroy QP. */
+ err = ib_copy_to_udata(udata, &resp, resp.response_length);
out:
- if (likely(nreq)) {
- qp->rq.head += nreq;
-
- /* Make sure that descriptors are written before
- * doorbell record.
- */
- wmb();
-
- *qp->db.db = cpu_to_be32(qp->rq.head & 0xffff);
- }
-
- spin_unlock_irqrestore(&qp->rq.lock, flags);
-
+ mutex_unlock(&qp->mutex);
return err;
}
-int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
- const struct ib_recv_wr **bad_wr)
-{
- return _mlx5_ib_post_recv(ibqp, wr, bad_wr, false);
-}
-
static inline enum ib_qp_state to_ib_qp_state(enum mlx5_qp_state mlx5_state)
{
switch (mlx5_state) {
@@ -5521,50 +4357,35 @@ static inline enum ib_mig_state to_ib_mig_state(int mlx5_mig_state)
}
}
-static int to_ib_qp_access_flags(int mlx5_flags)
-{
- int ib_flags = 0;
-
- if (mlx5_flags & MLX5_QP_BIT_RRE)
- ib_flags |= IB_ACCESS_REMOTE_READ;
- if (mlx5_flags & MLX5_QP_BIT_RWE)
- ib_flags |= IB_ACCESS_REMOTE_WRITE;
- if (mlx5_flags & MLX5_QP_BIT_RAE)
- ib_flags |= IB_ACCESS_REMOTE_ATOMIC;
-
- return ib_flags;
-}
-
static void to_rdma_ah_attr(struct mlx5_ib_dev *ibdev,
- struct rdma_ah_attr *ah_attr,
- struct mlx5_qp_path *path)
+ struct rdma_ah_attr *ah_attr, void *path)
{
+ int port = MLX5_GET(ads, path, vhca_port_num);
+ int static_rate;
memset(ah_attr, 0, sizeof(*ah_attr));
- if (!path->port || path->port > ibdev->num_ports)
+ if (!port || port > ibdev->num_ports)
return;
- ah_attr->type = rdma_ah_find_type(&ibdev->ib_dev, path->port);
+ ah_attr->type = rdma_ah_find_type(&ibdev->ib_dev, port);
- rdma_ah_set_port_num(ah_attr, path->port);
- rdma_ah_set_sl(ah_attr, path->dci_cfi_prio_sl & 0xf);
+ rdma_ah_set_port_num(ah_attr, port);
+ rdma_ah_set_sl(ah_attr, MLX5_GET(ads, path, sl));
- rdma_ah_set_dlid(ah_attr, be16_to_cpu(path->rlid));
- rdma_ah_set_path_bits(ah_attr, path->grh_mlid & 0x7f);
- rdma_ah_set_static_rate(ah_attr,
- path->static_rate ? path->static_rate - 5 : 0);
+ rdma_ah_set_dlid(ah_attr, MLX5_GET(ads, path, rlid));
+ rdma_ah_set_path_bits(ah_attr, MLX5_GET(ads, path, mlid));
- if (path->grh_mlid & (1 << 7) ||
+ static_rate = MLX5_GET(ads, path, stat_rate);
+ rdma_ah_set_static_rate(ah_attr, static_rate ? static_rate - 5 : 0);
+ if (MLX5_GET(ads, path, grh) ||
ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
- u32 tc_fl = be32_to_cpu(path->tclass_flowlabel);
-
- rdma_ah_set_grh(ah_attr, NULL,
- tc_fl & 0xfffff,
- path->mgid_index,
- path->hop_limit,
- (tc_fl >> 20) & 0xff);
- rdma_ah_set_dgid_raw(ah_attr, path->rgid);
+ rdma_ah_set_grh(ah_attr, NULL, MLX5_GET(ads, path, flow_label),
+ MLX5_GET(ads, path, src_addr_index),
+ MLX5_GET(ads, path, hop_limit),
+ MLX5_GET(ads, path, tclass));
+ memcpy(ah_attr, MLX5_ADDR_OF(ads, path, rgid_rip),
+ MLX5_FLD_SZ_BYTES(ads, rgid_rip));
}
}
@@ -5686,10 +4507,9 @@ static int query_qp_attr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
struct ib_qp_attr *qp_attr)
{
int outlen = MLX5_ST_SZ_BYTES(query_qp_out);
- struct mlx5_qp_context *context;
- int mlx5_state;
+ void *qpc, *pri_path, *alt_path;
u32 *outb;
- int err = 0;
+ int err;
outb = kzalloc(outlen, GFP_KERNEL);
if (!outb)
@@ -5699,47 +4519,46 @@ static int query_qp_attr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
if (err)
goto out;
- /* FIXME: use MLX5_GET rather than mlx5_qp_context manual struct */
- context = (struct mlx5_qp_context *)MLX5_ADDR_OF(query_qp_out, outb, qpc);
+ qpc = MLX5_ADDR_OF(query_qp_out, outb, qpc);
- mlx5_state = be32_to_cpu(context->flags) >> 28;
+ qp->state = to_ib_qp_state(MLX5_GET(qpc, qpc, state));
+ if (MLX5_GET(qpc, qpc, state) == MLX5_QP_STATE_SQ_DRAINING)
+ qp_attr->sq_draining = 1;
- qp->state = to_ib_qp_state(mlx5_state);
- qp_attr->path_mtu = context->mtu_msgmax >> 5;
- qp_attr->path_mig_state =
- to_ib_mig_state((be32_to_cpu(context->flags) >> 11) & 0x3);
- qp_attr->qkey = be32_to_cpu(context->qkey);
- qp_attr->rq_psn = be32_to_cpu(context->rnr_nextrecvpsn) & 0xffffff;
- qp_attr->sq_psn = be32_to_cpu(context->next_send_psn) & 0xffffff;
- qp_attr->dest_qp_num = be32_to_cpu(context->log_pg_sz_remote_qpn) & 0xffffff;
- qp_attr->qp_access_flags =
- to_ib_qp_access_flags(be32_to_cpu(context->params2));
+ qp_attr->path_mtu = MLX5_GET(qpc, qpc, mtu);
+ qp_attr->path_mig_state = to_ib_mig_state(MLX5_GET(qpc, qpc, pm_state));
+ qp_attr->qkey = MLX5_GET(qpc, qpc, q_key);
+ qp_attr->rq_psn = MLX5_GET(qpc, qpc, next_rcv_psn);
+ qp_attr->sq_psn = MLX5_GET(qpc, qpc, next_send_psn);
+ qp_attr->dest_qp_num = MLX5_GET(qpc, qpc, remote_qpn);
- if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) {
- to_rdma_ah_attr(dev, &qp_attr->ah_attr, &context->pri_path);
- to_rdma_ah_attr(dev, &qp_attr->alt_ah_attr, &context->alt_path);
- qp_attr->alt_pkey_index =
- be16_to_cpu(context->alt_path.pkey_index);
- qp_attr->alt_port_num =
- rdma_ah_get_port_num(&qp_attr->alt_ah_attr);
- }
+ if (MLX5_GET(qpc, qpc, rre))
+ qp_attr->qp_access_flags |= IB_ACCESS_REMOTE_READ;
+ if (MLX5_GET(qpc, qpc, rwe))
+ qp_attr->qp_access_flags |= IB_ACCESS_REMOTE_WRITE;
+ if (MLX5_GET(qpc, qpc, rae))
+ qp_attr->qp_access_flags |= IB_ACCESS_REMOTE_ATOMIC;
- qp_attr->pkey_index = be16_to_cpu(context->pri_path.pkey_index);
- qp_attr->port_num = context->pri_path.port;
+ qp_attr->max_rd_atomic = 1 << MLX5_GET(qpc, qpc, log_sra_max);
+ qp_attr->max_dest_rd_atomic = 1 << MLX5_GET(qpc, qpc, log_rra_max);
+ qp_attr->min_rnr_timer = MLX5_GET(qpc, qpc, min_rnr_nak);
+ qp_attr->retry_cnt = MLX5_GET(qpc, qpc, retry_count);
+ qp_attr->rnr_retry = MLX5_GET(qpc, qpc, rnr_retry);
- /* qp_attr->en_sqd_async_notify is only applicable in modify qp */
- qp_attr->sq_draining = mlx5_state == MLX5_QP_STATE_SQ_DRAINING;
+ pri_path = MLX5_ADDR_OF(qpc, qpc, primary_address_path);
+ alt_path = MLX5_ADDR_OF(qpc, qpc, secondary_address_path);
- qp_attr->max_rd_atomic = 1 << ((be32_to_cpu(context->params1) >> 21) & 0x7);
+ if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) {
+ to_rdma_ah_attr(dev, &qp_attr->ah_attr, pri_path);
+ to_rdma_ah_attr(dev, &qp_attr->alt_ah_attr, alt_path);
+ qp_attr->alt_pkey_index = MLX5_GET(ads, alt_path, pkey_index);
+ qp_attr->alt_port_num = MLX5_GET(ads, alt_path, vhca_port_num);
+ }
- qp_attr->max_dest_rd_atomic =
- 1 << ((be32_to_cpu(context->params2) >> 21) & 0x7);
- qp_attr->min_rnr_timer =
- (be32_to_cpu(context->rnr_nextrecvpsn) >> 24) & 0x1f;
- qp_attr->timeout = context->pri_path.ackto_lt >> 3;
- qp_attr->retry_cnt = (be32_to_cpu(context->params1) >> 16) & 0x7;
- qp_attr->rnr_retry = (be32_to_cpu(context->params1) >> 13) & 0x7;
- qp_attr->alt_timeout = context->alt_path.ackto_lt >> 3;
+ qp_attr->pkey_index = MLX5_GET(ads, pri_path, pkey_index);
+ qp_attr->port_num = MLX5_GET(ads, pri_path, vhca_port_num);
+ qp_attr->timeout = MLX5_GET(ads, pri_path, ack_timeout);
+ qp_attr->alt_timeout = MLX5_GET(ads, alt_path, ack_timeout);
out:
kfree(outb);
@@ -5830,14 +4649,14 @@ int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
memset(qp_init_attr, 0, sizeof(*qp_init_attr));
memset(qp_attr, 0, sizeof(*qp_attr));
- if (unlikely(qp->qp_sub_type == MLX5_IB_QPT_DCT))
+ if (unlikely(qp->type == MLX5_IB_QPT_DCT))
return mlx5_ib_dct_query_qp(dev, qp, qp_attr,
qp_attr_mask, qp_init_attr);
mutex_lock(&qp->mutex);
if (qp->ibqp.qp_type == IB_QPT_RAW_PACKET ||
- qp->flags & MLX5_IB_QP_UNDERLAY) {
+ qp->flags & IB_QP_CREATE_SOURCE_QPN) {
err = query_raw_packet_qp_state(dev, qp, &raw_packet_qp_state);
if (err)
goto out;
@@ -5871,18 +4690,7 @@ int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
qp_init_attr->cap = qp_attr->cap;
- qp_init_attr->create_flags = 0;
- if (qp->flags & MLX5_IB_QP_BLOCK_MULTICAST_LOOPBACK)
- qp_init_attr->create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK;
-
- if (qp->flags & MLX5_IB_QP_CROSS_CHANNEL)
- qp_init_attr->create_flags |= IB_QP_CREATE_CROSS_CHANNEL;
- if (qp->flags & MLX5_IB_QP_MANAGED_SEND)
- qp_init_attr->create_flags |= IB_QP_CREATE_MANAGED_SEND;
- if (qp->flags & MLX5_IB_QP_MANAGED_RECV)
- qp_init_attr->create_flags |= IB_QP_CREATE_MANAGED_RECV;
- if (qp->flags & MLX5_IB_QP_SQPN_QP1)
- qp_init_attr->create_flags |= MLX5_IB_QP_CREATE_SQPN_QP1;
+ qp_init_attr->create_flags = qp->flags;
qp_init_attr->sq_sig_type = qp->sq_signal_bits & MLX5_WQE_CTRL_CQ_UPDATE ?
IB_SIGNAL_ALL_WR : IB_SIGNAL_REQ_WR;
@@ -6543,7 +5351,7 @@ void mlx5_ib_drain_sq(struct ib_qp *qp)
sdrain.cqe.done = mlx5_ib_drain_qp_done;
init_completion(&sdrain.done);
- ret = _mlx5_ib_post_send(qp, &swr.wr, &bad_swr, true);
+ ret = mlx5_ib_post_send_drain(qp, &swr.wr, &bad_swr);
if (ret) {
WARN_ONCE(ret, "failed to drain send queue: %d\n", ret);
return;
@@ -6573,7 +5381,7 @@ void mlx5_ib_drain_rq(struct ib_qp *qp)
rdrain.cqe.done = mlx5_ib_drain_qp_done;
init_completion(&rdrain.done);
- ret = _mlx5_ib_post_recv(qp, &rwr, &bad_rwr, true);
+ ret = mlx5_ib_post_recv_drain(qp, &rwr, &bad_rwr);
if (ret) {
WARN_ONCE(ret, "failed to drain recv queue: %d\n", ret);
return;
diff --git a/drivers/infiniband/hw/mlx5/qp.h b/drivers/infiniband/hw/mlx5/qp.h
index ad9d76e3e18a..82ea2b94dfa6 100644
--- a/drivers/infiniband/hw/mlx5/qp.h
+++ b/drivers/infiniband/hw/mlx5/qp.h
@@ -13,10 +13,10 @@ void mlx5_cleanup_qp_table(struct mlx5_ib_dev *dev);
int mlx5_core_create_dct(struct mlx5_ib_dev *dev, struct mlx5_core_dct *qp,
u32 *in, int inlen, u32 *out, int outlen);
-int mlx5_core_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
- u32 *in, int inlen);
+int mlx5_qpc_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+ u32 *in, int inlen, u32 *out);
int mlx5_core_qp_modify(struct mlx5_ib_dev *dev, u16 opcode, u32 opt_param_mask,
- void *qpc, struct mlx5_core_qp *qp);
+ void *qpc, struct mlx5_core_qp *qp, u32 *ece);
int mlx5_core_destroy_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp);
int mlx5_core_destroy_dct(struct mlx5_ib_dev *dev, struct mlx5_core_dct *dct);
int mlx5_core_qp_query(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
diff --git a/drivers/infiniband/hw/mlx5/qpc.c b/drivers/infiniband/hw/mlx5/qpc.c
index ea62735042f0..c19d91d6dce8 100644
--- a/drivers/infiniband/hw/mlx5/qpc.c
+++ b/drivers/infiniband/hw/mlx5/qpc.c
@@ -236,16 +236,16 @@ err_cmd:
return err;
}
-int mlx5_core_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
- u32 *in, int inlen)
+int mlx5_qpc_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+ u32 *in, int inlen, u32 *out)
{
- u32 out[MLX5_ST_SZ_DW(create_qp_out)] = {};
u32 din[MLX5_ST_SZ_DW(destroy_qp_in)] = {};
int err;
MLX5_SET(create_qp_in, in, opcode, MLX5_CMD_OP_CREATE_QP);
- err = mlx5_cmd_exec(dev->mdev, in, inlen, out, sizeof(out));
+ err = mlx5_cmd_exec(dev->mdev, in, inlen, out,
+ MLX5_ST_SZ_BYTES(create_qp_out));
if (err)
return err;
@@ -341,9 +341,30 @@ static void mbox_free(struct mbox_info *mbox)
kfree(mbox->out);
}
+static int get_ece_from_mbox(void *out, u16 opcode)
+{
+ int ece = 0;
+
+ switch (opcode) {
+ case MLX5_CMD_OP_INIT2RTR_QP:
+ ece = MLX5_GET(init2rtr_qp_out, out, ece);
+ break;
+ case MLX5_CMD_OP_RTR2RTS_QP:
+ ece = MLX5_GET(rtr2rts_qp_out, out, ece);
+ break;
+ case MLX5_CMD_OP_RTS2RTS_QP:
+ ece = MLX5_GET(rts2rts_qp_out, out, ece);
+ break;
+ default:
+ break;
+ }
+
+ return ece;
+}
+
static int modify_qp_mbox_alloc(struct mlx5_core_dev *dev, u16 opcode, int qpn,
u32 opt_param_mask, void *qpc,
- struct mbox_info *mbox, u16 uid)
+ struct mbox_info *mbox, u16 uid, u32 ece)
{
mbox->out = NULL;
mbox->in = NULL;
@@ -391,18 +412,21 @@ static int modify_qp_mbox_alloc(struct mlx5_core_dev *dev, u16 opcode, int qpn,
return -ENOMEM;
MOD_QP_IN_SET_QPC(init2rtr_qp, mbox->in, opcode, qpn,
opt_param_mask, qpc, uid);
+ MLX5_SET(init2rtr_qp_in, mbox->in, ece, ece);
break;
case MLX5_CMD_OP_RTR2RTS_QP:
if (MBOX_ALLOC(mbox, rtr2rts_qp))
return -ENOMEM;
MOD_QP_IN_SET_QPC(rtr2rts_qp, mbox->in, opcode, qpn,
opt_param_mask, qpc, uid);
+ MLX5_SET(rtr2rts_qp_in, mbox->in, ece, ece);
break;
case MLX5_CMD_OP_RTS2RTS_QP:
if (MBOX_ALLOC(mbox, rts2rts_qp))
return -ENOMEM;
MOD_QP_IN_SET_QPC(rts2rts_qp, mbox->in, opcode, qpn,
opt_param_mask, qpc, uid);
+ MLX5_SET(rts2rts_qp_in, mbox->in, ece, ece);
break;
case MLX5_CMD_OP_SQERR2RTS_QP:
if (MBOX_ALLOC(mbox, sqerr2rts_qp))
@@ -423,18 +447,22 @@ static int modify_qp_mbox_alloc(struct mlx5_core_dev *dev, u16 opcode, int qpn,
}
int mlx5_core_qp_modify(struct mlx5_ib_dev *dev, u16 opcode, u32 opt_param_mask,
- void *qpc, struct mlx5_core_qp *qp)
+ void *qpc, struct mlx5_core_qp *qp, u32 *ece)
{
struct mbox_info mbox;
int err;
- err = modify_qp_mbox_alloc(dev->mdev, opcode, qp->qpn,
- opt_param_mask, qpc, &mbox, qp->uid);
+ err = modify_qp_mbox_alloc(dev->mdev, opcode, qp->qpn, opt_param_mask,
+ qpc, &mbox, qp->uid, (ece) ? *ece : 0);
if (err)
return err;
err = mlx5_cmd_exec(dev->mdev, mbox.in, mbox.inlen, mbox.out,
mbox.outlen);
+
+ if (ece)
+ *ece = get_ece_from_mbox(mbox.out, opcode);
+
mbox_free(&mbox);
return err;
}
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index b1a8a9175040..6d1ff13d2283 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -310,12 +310,18 @@ int mlx5_ib_create_srq(struct ib_srq *ib_srq,
srq->msrq.event = mlx5_ib_srq_event;
srq->ibsrq.ext.xrc.srq_num = srq->msrq.srqn;
- if (udata)
- if (ib_copy_to_udata(udata, &srq->msrq.srqn, sizeof(__u32))) {
+ if (udata) {
+ struct mlx5_ib_create_srq_resp resp = {
+ .srqn = srq->msrq.srqn,
+ };
+
+ if (ib_copy_to_udata(udata, &resp, min(udata->outlen,
+ sizeof(resp)))) {
mlx5_ib_dbg(dev, "copy to user failed\n");
err = -EFAULT;
goto err_core;
}
+ }
init_attr->attr.max_wr = srq->msrq.max - 1;
diff --git a/drivers/infiniband/hw/mlx5/srq_cmd.c b/drivers/infiniband/hw/mlx5/srq_cmd.c
index c851570791af..6f5eadc4d183 100644
--- a/drivers/infiniband/hw/mlx5/srq_cmd.c
+++ b/drivers/infiniband/hw/mlx5/srq_cmd.c
@@ -132,38 +132,33 @@ static int create_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
static int destroy_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq)
{
- u32 srq_in[MLX5_ST_SZ_DW(destroy_srq_in)] = {0};
- u32 srq_out[MLX5_ST_SZ_DW(destroy_srq_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(destroy_srq_in)] = {};
- MLX5_SET(destroy_srq_in, srq_in, opcode,
- MLX5_CMD_OP_DESTROY_SRQ);
- MLX5_SET(destroy_srq_in, srq_in, srqn, srq->srqn);
- MLX5_SET(destroy_srq_in, srq_in, uid, srq->uid);
+ MLX5_SET(destroy_srq_in, in, opcode, MLX5_CMD_OP_DESTROY_SRQ);
+ MLX5_SET(destroy_srq_in, in, srqn, srq->srqn);
+ MLX5_SET(destroy_srq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, srq_in, sizeof(srq_in), srq_out,
- sizeof(srq_out));
+ return mlx5_cmd_exec_in(dev->mdev, destroy_srq, in);
}
static int arm_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
u16 lwm, int is_srq)
{
- u32 srq_in[MLX5_ST_SZ_DW(arm_rq_in)] = {0};
- u32 srq_out[MLX5_ST_SZ_DW(arm_rq_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(arm_rq_in)] = {};
- MLX5_SET(arm_rq_in, srq_in, opcode, MLX5_CMD_OP_ARM_RQ);
- MLX5_SET(arm_rq_in, srq_in, op_mod, MLX5_ARM_RQ_IN_OP_MOD_SRQ);
- MLX5_SET(arm_rq_in, srq_in, srq_number, srq->srqn);
- MLX5_SET(arm_rq_in, srq_in, lwm, lwm);
- MLX5_SET(arm_rq_in, srq_in, uid, srq->uid);
+ MLX5_SET(arm_rq_in, in, opcode, MLX5_CMD_OP_ARM_RQ);
+ MLX5_SET(arm_rq_in, in, op_mod, MLX5_ARM_RQ_IN_OP_MOD_SRQ);
+ MLX5_SET(arm_rq_in, in, srq_number, srq->srqn);
+ MLX5_SET(arm_rq_in, in, lwm, lwm);
+ MLX5_SET(arm_rq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, srq_in, sizeof(srq_in), srq_out,
- sizeof(srq_out));
+ return mlx5_cmd_exec_in(dev->mdev, arm_rq, in);
}
static int query_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
struct mlx5_srq_attr *out)
{
- u32 srq_in[MLX5_ST_SZ_DW(query_srq_in)] = {0};
+ u32 in[MLX5_ST_SZ_DW(query_srq_in)] = {};
u32 *srq_out;
void *srqc;
int err;
@@ -172,11 +167,9 @@ static int query_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
if (!srq_out)
return -ENOMEM;
- MLX5_SET(query_srq_in, srq_in, opcode,
- MLX5_CMD_OP_QUERY_SRQ);
- MLX5_SET(query_srq_in, srq_in, srqn, srq->srqn);
- err = mlx5_cmd_exec(dev->mdev, srq_in, sizeof(srq_in), srq_out,
- MLX5_ST_SZ_BYTES(query_srq_out));
+ MLX5_SET(query_srq_in, in, opcode, MLX5_CMD_OP_QUERY_SRQ);
+ MLX5_SET(query_srq_in, in, srqn, srq->srqn);
+ err = mlx5_cmd_exec_inout(dev->mdev, query_srq, in, srq_out);
if (err)
goto out;
@@ -234,39 +227,35 @@ out:
static int destroy_xrc_srq_cmd(struct mlx5_ib_dev *dev,
struct mlx5_core_srq *srq)
{
- u32 xrcsrq_in[MLX5_ST_SZ_DW(destroy_xrc_srq_in)] = {0};
- u32 xrcsrq_out[MLX5_ST_SZ_DW(destroy_xrc_srq_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(destroy_xrc_srq_in)] = {};
- MLX5_SET(destroy_xrc_srq_in, xrcsrq_in, opcode,
- MLX5_CMD_OP_DESTROY_XRC_SRQ);
- MLX5_SET(destroy_xrc_srq_in, xrcsrq_in, xrc_srqn, srq->srqn);
- MLX5_SET(destroy_xrc_srq_in, xrcsrq_in, uid, srq->uid);
+ MLX5_SET(destroy_xrc_srq_in, in, opcode, MLX5_CMD_OP_DESTROY_XRC_SRQ);
+ MLX5_SET(destroy_xrc_srq_in, in, xrc_srqn, srq->srqn);
+ MLX5_SET(destroy_xrc_srq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, xrcsrq_in, sizeof(xrcsrq_in),
- xrcsrq_out, sizeof(xrcsrq_out));
+ return mlx5_cmd_exec_in(dev->mdev, destroy_xrc_srq, in);
}
static int arm_xrc_srq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
u16 lwm)
{
- u32 xrcsrq_in[MLX5_ST_SZ_DW(arm_xrc_srq_in)] = {0};
- u32 xrcsrq_out[MLX5_ST_SZ_DW(arm_xrc_srq_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(arm_xrc_srq_in)] = {};
- MLX5_SET(arm_xrc_srq_in, xrcsrq_in, opcode, MLX5_CMD_OP_ARM_XRC_SRQ);
- MLX5_SET(arm_xrc_srq_in, xrcsrq_in, op_mod, MLX5_ARM_XRC_SRQ_IN_OP_MOD_XRC_SRQ);
- MLX5_SET(arm_xrc_srq_in, xrcsrq_in, xrc_srqn, srq->srqn);
- MLX5_SET(arm_xrc_srq_in, xrcsrq_in, lwm, lwm);
- MLX5_SET(arm_xrc_srq_in, xrcsrq_in, uid, srq->uid);
+ MLX5_SET(arm_xrc_srq_in, in, opcode, MLX5_CMD_OP_ARM_XRC_SRQ);
+ MLX5_SET(arm_xrc_srq_in, in, op_mod,
+ MLX5_ARM_XRC_SRQ_IN_OP_MOD_XRC_SRQ);
+ MLX5_SET(arm_xrc_srq_in, in, xrc_srqn, srq->srqn);
+ MLX5_SET(arm_xrc_srq_in, in, lwm, lwm);
+ MLX5_SET(arm_xrc_srq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, xrcsrq_in, sizeof(xrcsrq_in),
- xrcsrq_out, sizeof(xrcsrq_out));
+ return mlx5_cmd_exec_in(dev->mdev, arm_xrc_srq, in);
}
static int query_xrc_srq_cmd(struct mlx5_ib_dev *dev,
struct mlx5_core_srq *srq,
struct mlx5_srq_attr *out)
{
- u32 xrcsrq_in[MLX5_ST_SZ_DW(query_xrc_srq_in)];
+ u32 in[MLX5_ST_SZ_DW(query_xrc_srq_in)] = {};
u32 *xrcsrq_out;
void *xrc_srqc;
int err;
@@ -274,14 +263,11 @@ static int query_xrc_srq_cmd(struct mlx5_ib_dev *dev,
xrcsrq_out = kvzalloc(MLX5_ST_SZ_BYTES(query_xrc_srq_out), GFP_KERNEL);
if (!xrcsrq_out)
return -ENOMEM;
- memset(xrcsrq_in, 0, sizeof(xrcsrq_in));
- MLX5_SET(query_xrc_srq_in, xrcsrq_in, opcode,
- MLX5_CMD_OP_QUERY_XRC_SRQ);
- MLX5_SET(query_xrc_srq_in, xrcsrq_in, xrc_srqn, srq->srqn);
+ MLX5_SET(query_xrc_srq_in, in, opcode, MLX5_CMD_OP_QUERY_XRC_SRQ);
+ MLX5_SET(query_xrc_srq_in, in, xrc_srqn, srq->srqn);
- err = mlx5_cmd_exec(dev->mdev, xrcsrq_in, sizeof(xrcsrq_in),
- xrcsrq_out, MLX5_ST_SZ_BYTES(query_xrc_srq_out));
+ err = mlx5_cmd_exec_inout(dev->mdev, query_xrc_srq, in, xrcsrq_out);
if (err)
goto out;
@@ -341,13 +327,12 @@ out:
static int destroy_rmp_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq)
{
- u32 in[MLX5_ST_SZ_DW(destroy_rmp_in)] = {};
- u32 out[MLX5_ST_SZ_DW(destroy_rmp_out)] = {};
+ u32 in[MLX5_ST_SZ_DW(destroy_rmp_in)] = {};
MLX5_SET(destroy_rmp_in, in, opcode, MLX5_CMD_OP_DESTROY_RMP);
MLX5_SET(destroy_rmp_in, in, rmpn, srq->srqn);
MLX5_SET(destroy_rmp_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev->mdev, destroy_rmp, in);
}
static int arm_rmp_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
@@ -384,7 +369,7 @@ static int arm_rmp_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
MLX5_SET(rmpc, rmpc, state, MLX5_RMPC_STATE_RDY);
MLX5_SET(modify_rmp_in, in, opcode, MLX5_CMD_OP_MODIFY_RMP);
- err = mlx5_cmd_exec(dev->mdev, in, inlen, out, outlen);
+ err = mlx5_cmd_exec_inout(dev->mdev, modify_rmp, in, out);
out:
kvfree(in);
@@ -414,7 +399,7 @@ static int query_rmp_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
MLX5_SET(query_rmp_in, rmp_in, opcode, MLX5_CMD_OP_QUERY_RMP);
MLX5_SET(query_rmp_in, rmp_in, rmpn, srq->srqn);
- err = mlx5_cmd_exec(dev->mdev, rmp_in, inlen, rmp_out, outlen);
+ err = mlx5_cmd_exec_inout(dev->mdev, query_rmp, rmp_in, rmp_out);
if (err)
goto out;
@@ -477,36 +462,34 @@ static int create_xrq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
static int destroy_xrq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq)
{
- u32 in[MLX5_ST_SZ_DW(destroy_xrq_in)] = {0};
- u32 out[MLX5_ST_SZ_DW(destroy_xrq_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(destroy_xrq_in)] = {};
MLX5_SET(destroy_xrq_in, in, opcode, MLX5_CMD_OP_DESTROY_XRQ);
- MLX5_SET(destroy_xrq_in, in, xrqn, srq->srqn);
+ MLX5_SET(destroy_xrq_in, in, xrqn, srq->srqn);
MLX5_SET(destroy_xrq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev->mdev, destroy_xrq, in);
}
static int arm_xrq_cmd(struct mlx5_ib_dev *dev,
struct mlx5_core_srq *srq,
u16 lwm)
{
- u32 out[MLX5_ST_SZ_DW(arm_rq_out)] = {0};
- u32 in[MLX5_ST_SZ_DW(arm_rq_in)] = {0};
+ u32 in[MLX5_ST_SZ_DW(arm_rq_in)] = {};
- MLX5_SET(arm_rq_in, in, opcode, MLX5_CMD_OP_ARM_RQ);
- MLX5_SET(arm_rq_in, in, op_mod, MLX5_ARM_RQ_IN_OP_MOD_XRQ);
+ MLX5_SET(arm_rq_in, in, opcode, MLX5_CMD_OP_ARM_RQ);
+ MLX5_SET(arm_rq_in, in, op_mod, MLX5_ARM_RQ_IN_OP_MOD_XRQ);
MLX5_SET(arm_rq_in, in, srq_number, srq->srqn);
- MLX5_SET(arm_rq_in, in, lwm, lwm);
+ MLX5_SET(arm_rq_in, in, lwm, lwm);
MLX5_SET(arm_rq_in, in, uid, srq->uid);
- return mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+ return mlx5_cmd_exec_in(dev->mdev, arm_rq, in);
}
static int query_xrq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
struct mlx5_srq_attr *out)
{
- u32 in[MLX5_ST_SZ_DW(query_xrq_in)] = {0};
+ u32 in[MLX5_ST_SZ_DW(query_xrq_in)] = {};
u32 *xrq_out;
int outlen = MLX5_ST_SZ_BYTES(query_xrq_out);
void *xrqc;
@@ -519,7 +502,7 @@ static int query_xrq_cmd(struct mlx5_ib_dev *dev, struct mlx5_core_srq *srq,
MLX5_SET(query_xrq_in, in, opcode, MLX5_CMD_OP_QUERY_XRQ);
MLX5_SET(query_xrq_in, in, xrqn, srq->srqn);
- err = mlx5_cmd_exec(dev->mdev, in, sizeof(in), xrq_out, outlen);
+ err = mlx5_cmd_exec_inout(dev->mdev, query_xrq, in, xrq_out);
if (err)
goto out;
diff --git a/drivers/infiniband/hw/mlx5/wr.c b/drivers/infiniband/hw/mlx5/wr.c
new file mode 100644
index 000000000000..2c6df1c43b55
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/wr.c
@@ -0,0 +1,1504 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright (c) 2020, Mellanox Technologies inc. All rights reserved.
+ */
+
+#include <linux/gfp.h>
+#include <linux/mlx5/qp.h>
+#include <linux/mlx5/driver.h>
+#include "wr.h"
+
+static const u32 mlx5_ib_opcode[] = {
+ [IB_WR_SEND] = MLX5_OPCODE_SEND,
+ [IB_WR_LSO] = MLX5_OPCODE_LSO,
+ [IB_WR_SEND_WITH_IMM] = MLX5_OPCODE_SEND_IMM,
+ [IB_WR_RDMA_WRITE] = MLX5_OPCODE_RDMA_WRITE,
+ [IB_WR_RDMA_WRITE_WITH_IMM] = MLX5_OPCODE_RDMA_WRITE_IMM,
+ [IB_WR_RDMA_READ] = MLX5_OPCODE_RDMA_READ,
+ [IB_WR_ATOMIC_CMP_AND_SWP] = MLX5_OPCODE_ATOMIC_CS,
+ [IB_WR_ATOMIC_FETCH_AND_ADD] = MLX5_OPCODE_ATOMIC_FA,
+ [IB_WR_SEND_WITH_INV] = MLX5_OPCODE_SEND_INVAL,
+ [IB_WR_LOCAL_INV] = MLX5_OPCODE_UMR,
+ [IB_WR_REG_MR] = MLX5_OPCODE_UMR,
+ [IB_WR_MASKED_ATOMIC_CMP_AND_SWP] = MLX5_OPCODE_ATOMIC_MASKED_CS,
+ [IB_WR_MASKED_ATOMIC_FETCH_AND_ADD] = MLX5_OPCODE_ATOMIC_MASKED_FA,
+ [MLX5_IB_WR_UMR] = MLX5_OPCODE_UMR,
+};
+
+/* handle_post_send_edge - Check if we get to SQ edge. If yes, update to the
+ * next nearby edge and get new address translation for current WQE position.
+ * @sq - SQ buffer.
+ * @seg: Current WQE position (16B aligned).
+ * @wqe_sz: Total current WQE size [16B].
+ * @cur_edge: Updated current edge.
+ */
+static inline void handle_post_send_edge(struct mlx5_ib_wq *sq, void **seg,
+ u32 wqe_sz, void **cur_edge)
+{
+ u32 idx;
+
+ if (likely(*seg != *cur_edge))
+ return;
+
+ idx = (sq->cur_post + (wqe_sz >> 2)) & (sq->wqe_cnt - 1);
+ *cur_edge = get_sq_edge(sq, idx);
+
+ *seg = mlx5_frag_buf_get_wqe(&sq->fbc, idx);
+}
+
+/* memcpy_send_wqe - copy data from src to WQE and update the relevant WQ's
+ * pointers. At the end @seg is aligned to 16B regardless the copied size.
+ * @sq - SQ buffer.
+ * @cur_edge: Updated current edge.
+ * @seg: Current WQE position (16B aligned).
+ * @wqe_sz: Total current WQE size [16B].
+ * @src: Pointer to copy from.
+ * @n: Number of bytes to copy.
+ */
+static inline void memcpy_send_wqe(struct mlx5_ib_wq *sq, void **cur_edge,
+ void **seg, u32 *wqe_sz, const void *src,
+ size_t n)
+{
+ while (likely(n)) {
+ size_t leftlen = *cur_edge - *seg;
+ size_t copysz = min_t(size_t, leftlen, n);
+ size_t stride;
+
+ memcpy(*seg, src, copysz);
+
+ n -= copysz;
+ src += copysz;
+ stride = !n ? ALIGN(copysz, 16) : copysz;
+ *seg += stride;
+ *wqe_sz += stride >> 4;
+ handle_post_send_edge(sq, seg, *wqe_sz, cur_edge);
+ }
+}
+
+static int mlx5_wq_overflow(struct mlx5_ib_wq *wq, int nreq,
+ struct ib_cq *ib_cq)
+{
+ struct mlx5_ib_cq *cq;
+ unsigned int cur;
+
+ cur = wq->head - wq->tail;
+ if (likely(cur + nreq < wq->max_post))
+ return 0;
+
+ cq = to_mcq(ib_cq);
+ spin_lock(&cq->lock);
+ cur = wq->head - wq->tail;
+ spin_unlock(&cq->lock);
+
+ return cur + nreq >= wq->max_post;
+}
+
+static __always_inline void set_raddr_seg(struct mlx5_wqe_raddr_seg *rseg,
+ u64 remote_addr, u32 rkey)
+{
+ rseg->raddr = cpu_to_be64(remote_addr);
+ rseg->rkey = cpu_to_be32(rkey);
+ rseg->reserved = 0;
+}
+
+static void set_eth_seg(const struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
+ void **seg, int *size, void **cur_edge)
+{
+ struct mlx5_wqe_eth_seg *eseg = *seg;
+
+ memset(eseg, 0, sizeof(struct mlx5_wqe_eth_seg));
+
+ if (wr->send_flags & IB_SEND_IP_CSUM)
+ eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM |
+ MLX5_ETH_WQE_L4_CSUM;
+
+ if (wr->opcode == IB_WR_LSO) {
+ struct ib_ud_wr *ud_wr = container_of(wr, struct ib_ud_wr, wr);
+ size_t left, copysz;
+ void *pdata = ud_wr->header;
+ size_t stride;
+
+ left = ud_wr->hlen;
+ eseg->mss = cpu_to_be16(ud_wr->mss);
+ eseg->inline_hdr.sz = cpu_to_be16(left);
+
+ /* memcpy_send_wqe should get a 16B align address. Hence, we
+ * first copy up to the current edge and then, if needed,
+ * continue to memcpy_send_wqe.
+ */
+ copysz = min_t(u64, *cur_edge - (void *)eseg->inline_hdr.start,
+ left);
+ memcpy(eseg->inline_hdr.start, pdata, copysz);
+ stride = ALIGN(sizeof(struct mlx5_wqe_eth_seg) -
+ sizeof(eseg->inline_hdr.start) + copysz, 16);
+ *size += stride / 16;
+ *seg += stride;
+
+ if (copysz < left) {
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+ left -= copysz;
+ pdata += copysz;
+ memcpy_send_wqe(&qp->sq, cur_edge, seg, size, pdata,
+ left);
+ }
+
+ return;
+ }
+
+ *seg += sizeof(struct mlx5_wqe_eth_seg);
+ *size += sizeof(struct mlx5_wqe_eth_seg) / 16;
+}
+
+static void set_datagram_seg(struct mlx5_wqe_datagram_seg *dseg,
+ const struct ib_send_wr *wr)
+{
+ memcpy(&dseg->av, &to_mah(ud_wr(wr)->ah)->av, sizeof(struct mlx5_av));
+ dseg->av.dqp_dct =
+ cpu_to_be32(ud_wr(wr)->remote_qpn | MLX5_EXTENDED_UD_AV);
+ dseg->av.key.qkey.qkey = cpu_to_be32(ud_wr(wr)->remote_qkey);
+}
+
+static void set_data_ptr_seg(struct mlx5_wqe_data_seg *dseg, struct ib_sge *sg)
+{
+ dseg->byte_count = cpu_to_be32(sg->length);
+ dseg->lkey = cpu_to_be32(sg->lkey);
+ dseg->addr = cpu_to_be64(sg->addr);
+}
+
+static u64 get_xlt_octo(u64 bytes)
+{
+ return ALIGN(bytes, MLX5_IB_UMR_XLT_ALIGNMENT) /
+ MLX5_IB_UMR_OCTOWORD;
+}
+
+static __be64 frwr_mkey_mask(bool atomic)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_LEN |
+ MLX5_MKEY_MASK_PAGE_SIZE |
+ MLX5_MKEY_MASK_START_ADDR |
+ MLX5_MKEY_MASK_EN_RINVAL |
+ MLX5_MKEY_MASK_KEY |
+ MLX5_MKEY_MASK_LR |
+ MLX5_MKEY_MASK_LW |
+ MLX5_MKEY_MASK_RR |
+ MLX5_MKEY_MASK_RW |
+ MLX5_MKEY_MASK_SMALL_FENCE |
+ MLX5_MKEY_MASK_FREE;
+
+ if (atomic)
+ result |= MLX5_MKEY_MASK_A;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 sig_mkey_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_LEN |
+ MLX5_MKEY_MASK_PAGE_SIZE |
+ MLX5_MKEY_MASK_START_ADDR |
+ MLX5_MKEY_MASK_EN_SIGERR |
+ MLX5_MKEY_MASK_EN_RINVAL |
+ MLX5_MKEY_MASK_KEY |
+ MLX5_MKEY_MASK_LR |
+ MLX5_MKEY_MASK_LW |
+ MLX5_MKEY_MASK_RR |
+ MLX5_MKEY_MASK_RW |
+ MLX5_MKEY_MASK_SMALL_FENCE |
+ MLX5_MKEY_MASK_FREE |
+ MLX5_MKEY_MASK_BSF_EN;
+
+ return cpu_to_be64(result);
+}
+
+static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
+ struct mlx5_ib_mr *mr, u8 flags, bool atomic)
+{
+ int size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
+
+ memset(umr, 0, sizeof(*umr));
+
+ umr->flags = flags;
+ umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
+ umr->mkey_mask = frwr_mkey_mask(atomic);
+}
+
+static void set_linv_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr)
+{
+ memset(umr, 0, sizeof(*umr));
+ umr->mkey_mask = cpu_to_be64(MLX5_MKEY_MASK_FREE);
+ umr->flags = MLX5_UMR_INLINE;
+}
+
+static __be64 get_umr_enable_mr_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_KEY |
+ MLX5_MKEY_MASK_FREE;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_disable_mr_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_FREE;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_update_translation_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_LEN |
+ MLX5_MKEY_MASK_PAGE_SIZE |
+ MLX5_MKEY_MASK_START_ADDR;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_update_access_mask(int atomic)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_LR |
+ MLX5_MKEY_MASK_LW |
+ MLX5_MKEY_MASK_RR |
+ MLX5_MKEY_MASK_RW;
+
+ if (atomic)
+ result |= MLX5_MKEY_MASK_A;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_update_pd_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_PD;
+
+ return cpu_to_be64(result);
+}
+
+static int umr_check_mkey_mask(struct mlx5_ib_dev *dev, u64 mask)
+{
+ if ((mask & MLX5_MKEY_MASK_PAGE_SIZE &&
+ MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled)) ||
+ (mask & MLX5_MKEY_MASK_A &&
+ MLX5_CAP_GEN(dev->mdev, umr_modify_atomic_disabled)))
+ return -EPERM;
+ return 0;
+}
+
+static int set_reg_umr_segment(struct mlx5_ib_dev *dev,
+ struct mlx5_wqe_umr_ctrl_seg *umr,
+ const struct ib_send_wr *wr, int atomic)
+{
+ const struct mlx5_umr_wr *umrwr = umr_wr(wr);
+
+ memset(umr, 0, sizeof(*umr));
+
+ if (!umrwr->ignore_free_state) {
+ if (wr->send_flags & MLX5_IB_SEND_UMR_FAIL_IF_FREE)
+ /* fail if free */
+ umr->flags = MLX5_UMR_CHECK_FREE;
+ else
+ /* fail if not free */
+ umr->flags = MLX5_UMR_CHECK_NOT_FREE;
+ }
+
+ umr->xlt_octowords = cpu_to_be16(get_xlt_octo(umrwr->xlt_size));
+ if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_XLT) {
+ u64 offset = get_xlt_octo(umrwr->offset);
+
+ umr->xlt_offset = cpu_to_be16(offset & 0xffff);
+ umr->xlt_offset_47_16 = cpu_to_be32(offset >> 16);
+ umr->flags |= MLX5_UMR_TRANSLATION_OFFSET_EN;
+ }
+ if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION)
+ umr->mkey_mask |= get_umr_update_translation_mask();
+ if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS) {
+ umr->mkey_mask |= get_umr_update_access_mask(atomic);
+ umr->mkey_mask |= get_umr_update_pd_mask();
+ }
+ if (wr->send_flags & MLX5_IB_SEND_UMR_ENABLE_MR)
+ umr->mkey_mask |= get_umr_enable_mr_mask();
+ if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR)
+ umr->mkey_mask |= get_umr_disable_mr_mask();
+
+ if (!wr->num_sge)
+ umr->flags |= MLX5_UMR_INLINE;
+
+ return umr_check_mkey_mask(dev, be64_to_cpu(umr->mkey_mask));
+}
+
+static u8 get_umr_flags(int acc)
+{
+ return (acc & IB_ACCESS_REMOTE_ATOMIC ? MLX5_PERM_ATOMIC : 0) |
+ (acc & IB_ACCESS_REMOTE_WRITE ? MLX5_PERM_REMOTE_WRITE : 0) |
+ (acc & IB_ACCESS_REMOTE_READ ? MLX5_PERM_REMOTE_READ : 0) |
+ (acc & IB_ACCESS_LOCAL_WRITE ? MLX5_PERM_LOCAL_WRITE : 0) |
+ MLX5_PERM_LOCAL_READ | MLX5_PERM_UMR_EN;
+}
+
+static void set_reg_mkey_seg(struct mlx5_mkey_seg *seg,
+ struct mlx5_ib_mr *mr,
+ u32 key, int access)
+{
+ int ndescs = ALIGN(mr->ndescs + mr->meta_ndescs, 8) >> 1;
+
+ memset(seg, 0, sizeof(*seg));
+
+ if (mr->access_mode == MLX5_MKC_ACCESS_MODE_MTT)
+ seg->log2_page_size = ilog2(mr->ibmr.page_size);
+ else if (mr->access_mode == MLX5_MKC_ACCESS_MODE_KLMS)
+ /* KLMs take twice the size of MTTs */
+ ndescs *= 2;
+
+ seg->flags = get_umr_flags(access) | mr->access_mode;
+ seg->qpn_mkey7_0 = cpu_to_be32((key & 0xff) | 0xffffff00);
+ seg->flags_pd = cpu_to_be32(MLX5_MKEY_REMOTE_INVAL);
+ seg->start_addr = cpu_to_be64(mr->ibmr.iova);
+ seg->len = cpu_to_be64(mr->ibmr.length);
+ seg->xlt_oct_size = cpu_to_be32(ndescs);
+}
+
+static void set_linv_mkey_seg(struct mlx5_mkey_seg *seg)
+{
+ memset(seg, 0, sizeof(*seg));
+ seg->status = MLX5_MKEY_STATUS_FREE;
+}
+
+static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg,
+ const struct ib_send_wr *wr)
+{
+ const struct mlx5_umr_wr *umrwr = umr_wr(wr);
+
+ memset(seg, 0, sizeof(*seg));
+ if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR)
+ seg->status = MLX5_MKEY_STATUS_FREE;
+
+ seg->flags = convert_access(umrwr->access_flags);
+ if (umrwr->pd)
+ seg->flags_pd = cpu_to_be32(to_mpd(umrwr->pd)->pdn);
+ if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION &&
+ !umrwr->length)
+ seg->flags_pd |= cpu_to_be32(MLX5_MKEY_LEN64);
+
+ seg->start_addr = cpu_to_be64(umrwr->virt_addr);
+ seg->len = cpu_to_be64(umrwr->length);
+ seg->log2_page_size = umrwr->page_shift;
+ seg->qpn_mkey7_0 = cpu_to_be32(0xffffff00 |
+ mlx5_mkey_variant(umrwr->mkey));
+}
+
+static void set_reg_data_seg(struct mlx5_wqe_data_seg *dseg,
+ struct mlx5_ib_mr *mr,
+ struct mlx5_ib_pd *pd)
+{
+ int bcount = mr->desc_size * (mr->ndescs + mr->meta_ndescs);
+
+ dseg->addr = cpu_to_be64(mr->desc_map);
+ dseg->byte_count = cpu_to_be32(ALIGN(bcount, 64));
+ dseg->lkey = cpu_to_be32(pd->ibpd.local_dma_lkey);
+}
+
+static __be32 send_ieth(const struct ib_send_wr *wr)
+{
+ switch (wr->opcode) {
+ case IB_WR_SEND_WITH_IMM:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ return wr->ex.imm_data;
+
+ case IB_WR_SEND_WITH_INV:
+ return cpu_to_be32(wr->ex.invalidate_rkey);
+
+ default:
+ return 0;
+ }
+}
+
+static u8 calc_sig(void *wqe, int size)
+{
+ u8 *p = wqe;
+ u8 res = 0;
+ int i;
+
+ for (i = 0; i < size; i++)
+ res ^= p[i];
+
+ return ~res;
+}
+
+static u8 wq_sig(void *wqe)
+{
+ return calc_sig(wqe, (*((u8 *)wqe + 8) & 0x3f) << 4);
+}
+
+static int set_data_inl_seg(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
+ void **wqe, int *wqe_sz, void **cur_edge)
+{
+ struct mlx5_wqe_inline_seg *seg;
+ size_t offset;
+ int inl = 0;
+ int i;
+
+ seg = *wqe;
+ *wqe += sizeof(*seg);
+ offset = sizeof(*seg);
+
+ for (i = 0; i < wr->num_sge; i++) {
+ size_t len = wr->sg_list[i].length;
+ void *addr = (void *)(unsigned long)(wr->sg_list[i].addr);
+
+ inl += len;
+
+ if (unlikely(inl > qp->max_inline_data))
+ return -ENOMEM;
+
+ while (likely(len)) {
+ size_t leftlen;
+ size_t copysz;
+
+ handle_post_send_edge(&qp->sq, wqe,
+ *wqe_sz + (offset >> 4),
+ cur_edge);
+
+ leftlen = *cur_edge - *wqe;
+ copysz = min_t(size_t, leftlen, len);
+
+ memcpy(*wqe, addr, copysz);
+ len -= copysz;
+ addr += copysz;
+ *wqe += copysz;
+ offset += copysz;
+ }
+ }
+
+ seg->byte_count = cpu_to_be32(inl | MLX5_INLINE_SEG);
+
+ *wqe_sz += ALIGN(inl + sizeof(seg->byte_count), 16) / 16;
+
+ return 0;
+}
+
+static u16 prot_field_size(enum ib_signature_type type)
+{
+ switch (type) {
+ case IB_SIG_TYPE_T10_DIF:
+ return MLX5_DIF_SIZE;
+ default:
+ return 0;
+ }
+}
+
+static u8 bs_selector(int block_size)
+{
+ switch (block_size) {
+ case 512: return 0x1;
+ case 520: return 0x2;
+ case 4096: return 0x3;
+ case 4160: return 0x4;
+ case 1073741824: return 0x5;
+ default: return 0;
+ }
+}
+
+static void mlx5_fill_inl_bsf(struct ib_sig_domain *domain,
+ struct mlx5_bsf_inl *inl)
+{
+ /* Valid inline section and allow BSF refresh */
+ inl->vld_refresh = cpu_to_be16(MLX5_BSF_INL_VALID |
+ MLX5_BSF_REFRESH_DIF);
+ inl->dif_apptag = cpu_to_be16(domain->sig.dif.app_tag);
+ inl->dif_reftag = cpu_to_be32(domain->sig.dif.ref_tag);
+ /* repeating block */
+ inl->rp_inv_seed = MLX5_BSF_REPEAT_BLOCK;
+ inl->sig_type = domain->sig.dif.bg_type == IB_T10DIF_CRC ?
+ MLX5_DIF_CRC : MLX5_DIF_IPCS;
+
+ if (domain->sig.dif.ref_remap)
+ inl->dif_inc_ref_guard_check |= MLX5_BSF_INC_REFTAG;
+
+ if (domain->sig.dif.app_escape) {
+ if (domain->sig.dif.ref_escape)
+ inl->dif_inc_ref_guard_check |= MLX5_BSF_APPREF_ESCAPE;
+ else
+ inl->dif_inc_ref_guard_check |= MLX5_BSF_APPTAG_ESCAPE;
+ }
+
+ inl->dif_app_bitmask_check =
+ cpu_to_be16(domain->sig.dif.apptag_check_mask);
+}
+
+static int mlx5_set_bsf(struct ib_mr *sig_mr,
+ struct ib_sig_attrs *sig_attrs,
+ struct mlx5_bsf *bsf, u32 data_size)
+{
+ struct mlx5_core_sig_ctx *msig = to_mmr(sig_mr)->sig;
+ struct mlx5_bsf_basic *basic = &bsf->basic;
+ struct ib_sig_domain *mem = &sig_attrs->mem;
+ struct ib_sig_domain *wire = &sig_attrs->wire;
+
+ memset(bsf, 0, sizeof(*bsf));
+
+ /* Basic + Extended + Inline */
+ basic->bsf_size_sbs = 1 << 7;
+ /* Input domain check byte mask */
+ basic->check_byte_mask = sig_attrs->check_mask;
+ basic->raw_data_size = cpu_to_be32(data_size);
+
+ /* Memory domain */
+ switch (sig_attrs->mem.sig_type) {
+ case IB_SIG_TYPE_NONE:
+ break;
+ case IB_SIG_TYPE_T10_DIF:
+ basic->mem.bs_selector = bs_selector(mem->sig.dif.pi_interval);
+ basic->m_bfs_psv = cpu_to_be32(msig->psv_memory.psv_idx);
+ mlx5_fill_inl_bsf(mem, &bsf->m_inl);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ /* Wire domain */
+ switch (sig_attrs->wire.sig_type) {
+ case IB_SIG_TYPE_NONE:
+ break;
+ case IB_SIG_TYPE_T10_DIF:
+ if (mem->sig.dif.pi_interval == wire->sig.dif.pi_interval &&
+ mem->sig_type == wire->sig_type) {
+ /* Same block structure */
+ basic->bsf_size_sbs |= 1 << 4;
+ if (mem->sig.dif.bg_type == wire->sig.dif.bg_type)
+ basic->wire.copy_byte_mask |= MLX5_CPY_GRD_MASK;
+ if (mem->sig.dif.app_tag == wire->sig.dif.app_tag)
+ basic->wire.copy_byte_mask |= MLX5_CPY_APP_MASK;
+ if (mem->sig.dif.ref_tag == wire->sig.dif.ref_tag)
+ basic->wire.copy_byte_mask |= MLX5_CPY_REF_MASK;
+ } else
+ basic->wire.bs_selector =
+ bs_selector(wire->sig.dif.pi_interval);
+
+ basic->w_bfs_psv = cpu_to_be32(msig->psv_wire.psv_idx);
+ mlx5_fill_inl_bsf(wire, &bsf->w_inl);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+
+static int set_sig_data_segment(const struct ib_send_wr *send_wr,
+ struct ib_mr *sig_mr,
+ struct ib_sig_attrs *sig_attrs,
+ struct mlx5_ib_qp *qp, void **seg, int *size,
+ void **cur_edge)
+{
+ struct mlx5_bsf *bsf;
+ u32 data_len;
+ u32 data_key;
+ u64 data_va;
+ u32 prot_len = 0;
+ u32 prot_key = 0;
+ u64 prot_va = 0;
+ bool prot = false;
+ int ret;
+ int wqe_size;
+ struct mlx5_ib_mr *mr = to_mmr(sig_mr);
+ struct mlx5_ib_mr *pi_mr = mr->pi_mr;
+
+ data_len = pi_mr->data_length;
+ data_key = pi_mr->ibmr.lkey;
+ data_va = pi_mr->data_iova;
+ if (pi_mr->meta_ndescs) {
+ prot_len = pi_mr->meta_length;
+ prot_key = pi_mr->ibmr.lkey;
+ prot_va = pi_mr->pi_iova;
+ prot = true;
+ }
+
+ if (!prot || (data_key == prot_key && data_va == prot_va &&
+ data_len == prot_len)) {
+ /**
+ * Source domain doesn't contain signature information
+ * or data and protection are interleaved in memory.
+ * So need construct:
+ * ------------------
+ * | data_klm |
+ * ------------------
+ * | BSF |
+ * ------------------
+ **/
+ struct mlx5_klm *data_klm = *seg;
+
+ data_klm->bcount = cpu_to_be32(data_len);
+ data_klm->key = cpu_to_be32(data_key);
+ data_klm->va = cpu_to_be64(data_va);
+ wqe_size = ALIGN(sizeof(*data_klm), 64);
+ } else {
+ /**
+ * Source domain contains signature information
+ * So need construct a strided block format:
+ * ---------------------------
+ * | stride_block_ctrl |
+ * ---------------------------
+ * | data_klm |
+ * ---------------------------
+ * | prot_klm |
+ * ---------------------------
+ * | BSF |
+ * ---------------------------
+ **/
+ struct mlx5_stride_block_ctrl_seg *sblock_ctrl;
+ struct mlx5_stride_block_entry *data_sentry;
+ struct mlx5_stride_block_entry *prot_sentry;
+ u16 block_size = sig_attrs->mem.sig.dif.pi_interval;
+ int prot_size;
+
+ sblock_ctrl = *seg;
+ data_sentry = (void *)sblock_ctrl + sizeof(*sblock_ctrl);
+ prot_sentry = (void *)data_sentry + sizeof(*data_sentry);
+
+ prot_size = prot_field_size(sig_attrs->mem.sig_type);
+ if (!prot_size) {
+ pr_err("Bad block size given: %u\n", block_size);
+ return -EINVAL;
+ }
+ sblock_ctrl->bcount_per_cycle = cpu_to_be32(block_size +
+ prot_size);
+ sblock_ctrl->op = cpu_to_be32(MLX5_STRIDE_BLOCK_OP);
+ sblock_ctrl->repeat_count = cpu_to_be32(data_len / block_size);
+ sblock_ctrl->num_entries = cpu_to_be16(2);
+
+ data_sentry->bcount = cpu_to_be16(block_size);
+ data_sentry->key = cpu_to_be32(data_key);
+ data_sentry->va = cpu_to_be64(data_va);
+ data_sentry->stride = cpu_to_be16(block_size);
+
+ prot_sentry->bcount = cpu_to_be16(prot_size);
+ prot_sentry->key = cpu_to_be32(prot_key);
+ prot_sentry->va = cpu_to_be64(prot_va);
+ prot_sentry->stride = cpu_to_be16(prot_size);
+
+ wqe_size = ALIGN(sizeof(*sblock_ctrl) + sizeof(*data_sentry) +
+ sizeof(*prot_sentry), 64);
+ }
+
+ *seg += wqe_size;
+ *size += wqe_size / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ bsf = *seg;
+ ret = mlx5_set_bsf(sig_mr, sig_attrs, bsf, data_len);
+ if (ret)
+ return -EINVAL;
+
+ *seg += sizeof(*bsf);
+ *size += sizeof(*bsf) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ return 0;
+}
+
+static void set_sig_mkey_segment(struct mlx5_mkey_seg *seg,
+ struct ib_mr *sig_mr, int access_flags,
+ u32 size, u32 length, u32 pdn)
+{
+ u32 sig_key = sig_mr->rkey;
+ u8 sigerr = to_mmr(sig_mr)->sig->sigerr_count & 1;
+
+ memset(seg, 0, sizeof(*seg));
+
+ seg->flags = get_umr_flags(access_flags) | MLX5_MKC_ACCESS_MODE_KLMS;
+ seg->qpn_mkey7_0 = cpu_to_be32((sig_key & 0xff) | 0xffffff00);
+ seg->flags_pd = cpu_to_be32(MLX5_MKEY_REMOTE_INVAL | sigerr << 26 |
+ MLX5_MKEY_BSF_EN | pdn);
+ seg->len = cpu_to_be64(length);
+ seg->xlt_oct_size = cpu_to_be32(get_xlt_octo(size));
+ seg->bsfs_octo_size = cpu_to_be32(MLX5_MKEY_BSF_OCTO_SIZE);
+}
+
+static void set_sig_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr,
+ u32 size)
+{
+ memset(umr, 0, sizeof(*umr));
+
+ umr->flags = MLX5_FLAGS_INLINE | MLX5_FLAGS_CHECK_FREE;
+ umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
+ umr->bsf_octowords = cpu_to_be16(MLX5_MKEY_BSF_OCTO_SIZE);
+ umr->mkey_mask = sig_mkey_mask();
+}
+
+static int set_pi_umr_wr(const struct ib_send_wr *send_wr,
+ struct mlx5_ib_qp *qp, void **seg, int *size,
+ void **cur_edge)
+{
+ const struct ib_reg_wr *wr = reg_wr(send_wr);
+ struct mlx5_ib_mr *sig_mr = to_mmr(wr->mr);
+ struct mlx5_ib_mr *pi_mr = sig_mr->pi_mr;
+ struct ib_sig_attrs *sig_attrs = sig_mr->ibmr.sig_attrs;
+ u32 pdn = to_mpd(qp->ibqp.pd)->pdn;
+ u32 xlt_size;
+ int region_len, ret;
+
+ if (unlikely(send_wr->num_sge != 0) ||
+ unlikely(wr->access & IB_ACCESS_REMOTE_ATOMIC) ||
+ unlikely(!sig_mr->sig) || unlikely(!qp->ibqp.integrity_en) ||
+ unlikely(!sig_mr->sig->sig_status_checked))
+ return -EINVAL;
+
+ /* length of the protected region, data + protection */
+ region_len = pi_mr->ibmr.length;
+
+ /**
+ * KLM octoword size - if protection was provided
+ * then we use strided block format (3 octowords),
+ * else we use single KLM (1 octoword)
+ **/
+ if (sig_attrs->mem.sig_type != IB_SIG_TYPE_NONE)
+ xlt_size = 0x30;
+ else
+ xlt_size = sizeof(struct mlx5_klm);
+
+ set_sig_umr_segment(*seg, xlt_size);
+ *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
+ *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ set_sig_mkey_segment(*seg, wr->mr, wr->access, xlt_size, region_len,
+ pdn);
+ *seg += sizeof(struct mlx5_mkey_seg);
+ *size += sizeof(struct mlx5_mkey_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ ret = set_sig_data_segment(send_wr, wr->mr, sig_attrs, qp, seg, size,
+ cur_edge);
+ if (ret)
+ return ret;
+
+ sig_mr->sig->sig_status_checked = false;
+ return 0;
+}
+
+static int set_psv_wr(struct ib_sig_domain *domain,
+ u32 psv_idx, void **seg, int *size)
+{
+ struct mlx5_seg_set_psv *psv_seg = *seg;
+
+ memset(psv_seg, 0, sizeof(*psv_seg));
+ psv_seg->psv_num = cpu_to_be32(psv_idx);
+ switch (domain->sig_type) {
+ case IB_SIG_TYPE_NONE:
+ break;
+ case IB_SIG_TYPE_T10_DIF:
+ psv_seg->transient_sig = cpu_to_be32(domain->sig.dif.bg << 16 |
+ domain->sig.dif.app_tag);
+ psv_seg->ref_tag = cpu_to_be32(domain->sig.dif.ref_tag);
+ break;
+ default:
+ pr_err("Bad signature type (%d) is given.\n",
+ domain->sig_type);
+ return -EINVAL;
+ }
+
+ *seg += sizeof(*psv_seg);
+ *size += sizeof(*psv_seg) / 16;
+
+ return 0;
+}
+
+static int set_reg_wr(struct mlx5_ib_qp *qp,
+ const struct ib_reg_wr *wr,
+ void **seg, int *size, void **cur_edge,
+ bool check_not_free)
+{
+ struct mlx5_ib_mr *mr = to_mmr(wr->mr);
+ struct mlx5_ib_pd *pd = to_mpd(qp->ibqp.pd);
+ struct mlx5_ib_dev *dev = to_mdev(pd->ibpd.device);
+ int mr_list_size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
+ bool umr_inline = mr_list_size <= MLX5_IB_SQ_UMR_INLINE_THRESHOLD;
+ bool atomic = wr->access & IB_ACCESS_REMOTE_ATOMIC;
+ u8 flags = 0;
+
+ if (!mlx5_ib_can_use_umr(dev, atomic, wr->access)) {
+ mlx5_ib_warn(to_mdev(qp->ibqp.device),
+ "Fast update of %s for MR is disabled\n",
+ (MLX5_CAP_GEN(dev->mdev,
+ umr_modify_entity_size_disabled)) ?
+ "entity size" :
+ "atomic access");
+ return -EINVAL;
+ }
+
+ if (unlikely(wr->wr.send_flags & IB_SEND_INLINE)) {
+ mlx5_ib_warn(to_mdev(qp->ibqp.device),
+ "Invalid IB_SEND_INLINE send flag\n");
+ return -EINVAL;
+ }
+
+ if (check_not_free)
+ flags |= MLX5_UMR_CHECK_NOT_FREE;
+ if (umr_inline)
+ flags |= MLX5_UMR_INLINE;
+
+ set_reg_umr_seg(*seg, mr, flags, atomic);
+ *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
+ *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ set_reg_mkey_seg(*seg, mr, wr->key, wr->access);
+ *seg += sizeof(struct mlx5_mkey_seg);
+ *size += sizeof(struct mlx5_mkey_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ if (umr_inline) {
+ memcpy_send_wqe(&qp->sq, cur_edge, seg, size, mr->descs,
+ mr_list_size);
+ *size = ALIGN(*size, MLX5_SEND_WQE_BB >> 4);
+ } else {
+ set_reg_data_seg(*seg, mr, pd);
+ *seg += sizeof(struct mlx5_wqe_data_seg);
+ *size += (sizeof(struct mlx5_wqe_data_seg) / 16);
+ }
+ return 0;
+}
+
+static void set_linv_wr(struct mlx5_ib_qp *qp, void **seg, int *size,
+ void **cur_edge)
+{
+ set_linv_umr_seg(*seg);
+ *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
+ *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+ set_linv_mkey_seg(*seg);
+ *seg += sizeof(struct mlx5_mkey_seg);
+ *size += sizeof(struct mlx5_mkey_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+}
+
+static void dump_wqe(struct mlx5_ib_qp *qp, u32 idx, int size_16)
+{
+ __be32 *p = NULL;
+ int i, j;
+
+ pr_debug("dump WQE index %u:\n", idx);
+ for (i = 0, j = 0; i < size_16 * 4; i += 4, j += 4) {
+ if ((i & 0xf) == 0) {
+ p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, idx);
+ pr_debug("WQBB at %p:\n", (void *)p);
+ j = 0;
+ idx = (idx + 1) & (qp->sq.wqe_cnt - 1);
+ }
+ pr_debug("%08x %08x %08x %08x\n", be32_to_cpu(p[j]),
+ be32_to_cpu(p[j + 1]), be32_to_cpu(p[j + 2]),
+ be32_to_cpu(p[j + 3]));
+ }
+}
+
+static int __begin_wqe(struct mlx5_ib_qp *qp, void **seg,
+ struct mlx5_wqe_ctrl_seg **ctrl,
+ const struct ib_send_wr *wr, unsigned int *idx,
+ int *size, void **cur_edge, int nreq,
+ bool send_signaled, bool solicited)
+{
+ if (unlikely(mlx5_wq_overflow(&qp->sq, nreq, qp->ibqp.send_cq)))
+ return -ENOMEM;
+
+ *idx = qp->sq.cur_post & (qp->sq.wqe_cnt - 1);
+ *seg = mlx5_frag_buf_get_wqe(&qp->sq.fbc, *idx);
+ *ctrl = *seg;
+ *(uint32_t *)(*seg + 8) = 0;
+ (*ctrl)->imm = send_ieth(wr);
+ (*ctrl)->fm_ce_se = qp->sq_signal_bits |
+ (send_signaled ? MLX5_WQE_CTRL_CQ_UPDATE : 0) |
+ (solicited ? MLX5_WQE_CTRL_SOLICITED : 0);
+
+ *seg += sizeof(**ctrl);
+ *size = sizeof(**ctrl) / 16;
+ *cur_edge = qp->sq.cur_edge;
+
+ return 0;
+}
+
+static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
+ struct mlx5_wqe_ctrl_seg **ctrl,
+ const struct ib_send_wr *wr, unsigned int *idx, int *size,
+ void **cur_edge, int nreq)
+{
+ return __begin_wqe(qp, seg, ctrl, wr, idx, size, cur_edge, nreq,
+ wr->send_flags & IB_SEND_SIGNALED,
+ wr->send_flags & IB_SEND_SOLICITED);
+}
+
+static void finish_wqe(struct mlx5_ib_qp *qp,
+ struct mlx5_wqe_ctrl_seg *ctrl,
+ void *seg, u8 size, void *cur_edge,
+ unsigned int idx, u64 wr_id, int nreq, u8 fence,
+ u32 mlx5_opcode)
+{
+ u8 opmod = 0;
+
+ ctrl->opmod_idx_opcode = cpu_to_be32(((u32)(qp->sq.cur_post) << 8) |
+ mlx5_opcode | ((u32)opmod << 24));
+ ctrl->qpn_ds = cpu_to_be32(size | (qp->trans_qp.base.mqp.qpn << 8));
+ ctrl->fm_ce_se |= fence;
+ if (unlikely(qp->flags_en & MLX5_QP_FLAG_SIGNATURE))
+ ctrl->signature = wq_sig(ctrl);
+
+ qp->sq.wrid[idx] = wr_id;
+ qp->sq.w_list[idx].opcode = mlx5_opcode;
+ qp->sq.wqe_head[idx] = qp->sq.head + nreq;
+ qp->sq.cur_post += DIV_ROUND_UP(size * 16, MLX5_SEND_WQE_BB);
+ qp->sq.w_list[idx].next = qp->sq.cur_post;
+
+ /* We save the edge which was possibly updated during the WQE
+ * construction, into SQ's cache.
+ */
+ seg = PTR_ALIGN(seg, MLX5_SEND_WQE_BB);
+ qp->sq.cur_edge = (unlikely(seg == cur_edge)) ?
+ get_sq_edge(&qp->sq, qp->sq.cur_post &
+ (qp->sq.wqe_cnt - 1)) :
+ cur_edge;
+}
+
+static void handle_rdma_op(const struct ib_send_wr *wr, void **seg, int *size)
+{
+ set_raddr_seg(*seg, rdma_wr(wr)->remote_addr, rdma_wr(wr)->rkey);
+ *seg += sizeof(struct mlx5_wqe_raddr_seg);
+ *size += sizeof(struct mlx5_wqe_raddr_seg) / 16;
+}
+
+static void handle_local_inv(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg,
+ int *size, void **cur_edge, unsigned int idx)
+{
+ qp->sq.wr_data[idx] = IB_WR_LOCAL_INV;
+ (*ctrl)->imm = cpu_to_be32(wr->ex.invalidate_rkey);
+ set_linv_wr(qp, seg, size, cur_edge);
+}
+
+static int handle_reg_mr(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg, int *size,
+ void **cur_edge, unsigned int idx)
+{
+ qp->sq.wr_data[idx] = IB_WR_REG_MR;
+ (*ctrl)->imm = cpu_to_be32(reg_wr(wr)->key);
+ return set_reg_wr(qp, reg_wr(wr), seg, size, cur_edge, true);
+}
+
+static int handle_psv(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg, int *size,
+ void **cur_edge, unsigned int *idx, int nreq,
+ struct ib_sig_domain *domain, u32 psv_index,
+ u8 next_fence)
+{
+ int err;
+
+ /*
+ * SET_PSV WQEs are not signaled and solicited on error.
+ */
+ err = __begin_wqe(qp, seg, ctrl, wr, idx, size, cur_edge, nreq,
+ false, true);
+ if (unlikely(err)) {
+ mlx5_ib_warn(dev, "\n");
+ err = -ENOMEM;
+ goto out;
+ }
+ err = set_psv_wr(domain, psv_index, seg, size);
+ if (unlikely(err)) {
+ mlx5_ib_warn(dev, "\n");
+ goto out;
+ }
+ finish_wqe(qp, *ctrl, *seg, *size, *cur_edge, *idx, wr->wr_id, nreq,
+ next_fence, MLX5_OPCODE_SET_PSV);
+
+out:
+ return err;
+}
+
+static int handle_reg_mr_integrity(struct mlx5_ib_dev *dev,
+ struct mlx5_ib_qp *qp,
+ const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg,
+ int *size, void **cur_edge,
+ unsigned int *idx, int nreq, u8 fence,
+ u8 next_fence)
+{
+ struct mlx5_ib_mr *mr;
+ struct mlx5_ib_mr *pi_mr;
+ struct mlx5_ib_mr pa_pi_mr;
+ struct ib_sig_attrs *sig_attrs;
+ struct ib_reg_wr reg_pi_wr;
+ int err;
+
+ qp->sq.wr_data[*idx] = IB_WR_REG_MR_INTEGRITY;
+
+ mr = to_mmr(reg_wr(wr)->mr);
+ pi_mr = mr->pi_mr;
+
+ if (pi_mr) {
+ memset(&reg_pi_wr, 0,
+ sizeof(struct ib_reg_wr));
+
+ reg_pi_wr.mr = &pi_mr->ibmr;
+ reg_pi_wr.access = reg_wr(wr)->access;
+ reg_pi_wr.key = pi_mr->ibmr.rkey;
+
+ (*ctrl)->imm = cpu_to_be32(reg_pi_wr.key);
+ /* UMR for data + prot registration */
+ err = set_reg_wr(qp, &reg_pi_wr, seg, size, cur_edge, false);
+ if (unlikely(err))
+ goto out;
+
+ finish_wqe(qp, *ctrl, *seg, *size, *cur_edge, *idx, wr->wr_id,
+ nreq, fence, MLX5_OPCODE_UMR);
+
+ err = begin_wqe(qp, seg, ctrl, wr, idx, size, cur_edge, nreq);
+ if (unlikely(err)) {
+ mlx5_ib_warn(dev, "\n");
+ err = -ENOMEM;
+ goto out;
+ }
+ } else {
+ memset(&pa_pi_mr, 0, sizeof(struct mlx5_ib_mr));
+ /* No UMR, use local_dma_lkey */
+ pa_pi_mr.ibmr.lkey = mr->ibmr.pd->local_dma_lkey;
+ pa_pi_mr.ndescs = mr->ndescs;
+ pa_pi_mr.data_length = mr->data_length;
+ pa_pi_mr.data_iova = mr->data_iova;
+ if (mr->meta_ndescs) {
+ pa_pi_mr.meta_ndescs = mr->meta_ndescs;
+ pa_pi_mr.meta_length = mr->meta_length;
+ pa_pi_mr.pi_iova = mr->pi_iova;
+ }
+
+ pa_pi_mr.ibmr.length = mr->ibmr.length;
+ mr->pi_mr = &pa_pi_mr;
+ }
+ (*ctrl)->imm = cpu_to_be32(mr->ibmr.rkey);
+ /* UMR for sig MR */
+ err = set_pi_umr_wr(wr, qp, seg, size, cur_edge);
+ if (unlikely(err)) {
+ mlx5_ib_warn(dev, "\n");
+ goto out;
+ }
+ finish_wqe(qp, *ctrl, *seg, *size, *cur_edge, *idx, wr->wr_id, nreq,
+ fence, MLX5_OPCODE_UMR);
+
+ sig_attrs = mr->ibmr.sig_attrs;
+ err = handle_psv(dev, qp, wr, ctrl, seg, size, cur_edge, idx, nreq,
+ &sig_attrs->mem, mr->sig->psv_memory.psv_idx,
+ next_fence);
+ if (unlikely(err))
+ goto out;
+
+ err = handle_psv(dev, qp, wr, ctrl, seg, size, cur_edge, idx, nreq,
+ &sig_attrs->wire, mr->sig->psv_wire.psv_idx,
+ next_fence);
+ if (unlikely(err))
+ goto out;
+
+ qp->next_fence = MLX5_FENCE_MODE_INITIATOR_SMALL;
+
+out:
+ return err;
+}
+
+static int handle_qpt_rc(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg, int *size,
+ void **cur_edge, unsigned int *idx, int nreq, u8 fence,
+ u8 next_fence, int *num_sge)
+{
+ int err = 0;
+
+ switch (wr->opcode) {
+ case IB_WR_RDMA_READ:
+ case IB_WR_RDMA_WRITE:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ handle_rdma_op(wr, seg, size);
+ break;
+
+ case IB_WR_ATOMIC_CMP_AND_SWP:
+ case IB_WR_ATOMIC_FETCH_AND_ADD:
+ case IB_WR_MASKED_ATOMIC_CMP_AND_SWP:
+ mlx5_ib_warn(dev, "Atomic operations are not supported yet\n");
+ err = -EOPNOTSUPP;
+ goto out;
+
+ case IB_WR_LOCAL_INV:
+ handle_local_inv(qp, wr, ctrl, seg, size, cur_edge, *idx);
+ *num_sge = 0;
+ break;
+
+ case IB_WR_REG_MR:
+ err = handle_reg_mr(qp, wr, ctrl, seg, size, cur_edge, *idx);
+ if (unlikely(err))
+ goto out;
+ *num_sge = 0;
+ break;
+
+ case IB_WR_REG_MR_INTEGRITY:
+ err = handle_reg_mr_integrity(dev, qp, wr, ctrl, seg, size,
+ cur_edge, idx, nreq, fence,
+ next_fence);
+ if (unlikely(err))
+ goto out;
+ *num_sge = 0;
+ break;
+
+ default:
+ break;
+ }
+
+out:
+ return err;
+}
+
+static void handle_qpt_uc(const struct ib_send_wr *wr, void **seg, int *size)
+{
+ switch (wr->opcode) {
+ case IB_WR_RDMA_WRITE:
+ case IB_WR_RDMA_WRITE_WITH_IMM:
+ handle_rdma_op(wr, seg, size);
+ break;
+ default:
+ break;
+ }
+}
+
+static void handle_qpt_hw_gsi(struct mlx5_ib_qp *qp,
+ const struct ib_send_wr *wr, void **seg,
+ int *size, void **cur_edge)
+{
+ set_datagram_seg(*seg, wr);
+ *seg += sizeof(struct mlx5_wqe_datagram_seg);
+ *size += sizeof(struct mlx5_wqe_datagram_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+}
+
+static void handle_qpt_ud(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
+ void **seg, int *size, void **cur_edge)
+{
+ set_datagram_seg(*seg, wr);
+ *seg += sizeof(struct mlx5_wqe_datagram_seg);
+ *size += sizeof(struct mlx5_wqe_datagram_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+
+ /* handle qp that supports ud offload */
+ if (qp->flags & IB_QP_CREATE_IPOIB_UD_LSO) {
+ struct mlx5_wqe_eth_pad *pad;
+
+ pad = *seg;
+ memset(pad, 0, sizeof(struct mlx5_wqe_eth_pad));
+ *seg += sizeof(struct mlx5_wqe_eth_pad);
+ *size += sizeof(struct mlx5_wqe_eth_pad) / 16;
+ set_eth_seg(wr, qp, seg, size, cur_edge);
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+ }
+}
+
+static int handle_qpt_reg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
+ const struct ib_send_wr *wr,
+ struct mlx5_wqe_ctrl_seg **ctrl, void **seg,
+ int *size, void **cur_edge, unsigned int idx)
+{
+ int err = 0;
+
+ if (unlikely(wr->opcode != MLX5_IB_WR_UMR)) {
+ err = -EINVAL;
+ mlx5_ib_warn(dev, "bad opcode %d\n", wr->opcode);
+ goto out;
+ }
+
+ qp->sq.wr_data[idx] = MLX5_IB_WR_UMR;
+ (*ctrl)->imm = cpu_to_be32(umr_wr(wr)->mkey);
+ err = set_reg_umr_segment(dev, *seg, wr,
+ !!(MLX5_CAP_GEN(dev->mdev, atomic)));
+ if (unlikely(err))
+ goto out;
+ *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
+ *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+ set_reg_mkey_segment(*seg, wr);
+ *seg += sizeof(struct mlx5_mkey_seg);
+ *size += sizeof(struct mlx5_mkey_seg) / 16;
+ handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
+out:
+ return err;
+}
+
+int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
+ const struct ib_send_wr **bad_wr, bool drain)
+{
+ struct mlx5_wqe_ctrl_seg *ctrl = NULL; /* compiler warning */
+ struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
+ struct mlx5_core_dev *mdev = dev->mdev;
+ struct mlx5_ib_qp *qp;
+ struct mlx5_wqe_xrc_seg *xrc;
+ struct mlx5_bf *bf;
+ void *cur_edge;
+ int uninitialized_var(size);
+ unsigned long flags;
+ unsigned int idx;
+ int err = 0;
+ int num_sge;
+ void *seg;
+ int nreq;
+ int i;
+ u8 next_fence = 0;
+ u8 fence;
+
+ if (unlikely(mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR &&
+ !drain)) {
+ *bad_wr = wr;
+ return -EIO;
+ }
+
+ if (unlikely(ibqp->qp_type == IB_QPT_GSI))
+ return mlx5_ib_gsi_post_send(ibqp, wr, bad_wr);
+
+ qp = to_mqp(ibqp);
+ bf = &qp->bf;
+
+ spin_lock_irqsave(&qp->sq.lock, flags);
+
+ for (nreq = 0; wr; nreq++, wr = wr->next) {
+ if (unlikely(wr->opcode >= ARRAY_SIZE(mlx5_ib_opcode))) {
+ mlx5_ib_warn(dev, "\n");
+ err = -EINVAL;
+ *bad_wr = wr;
+ goto out;
+ }
+
+ num_sge = wr->num_sge;
+ if (unlikely(num_sge > qp->sq.max_gs)) {
+ mlx5_ib_warn(dev, "\n");
+ err = -EINVAL;
+ *bad_wr = wr;
+ goto out;
+ }
+
+ err = begin_wqe(qp, &seg, &ctrl, wr, &idx, &size, &cur_edge,
+ nreq);
+ if (err) {
+ mlx5_ib_warn(dev, "\n");
+ err = -ENOMEM;
+ *bad_wr = wr;
+ goto out;
+ }
+
+ if (wr->opcode == IB_WR_REG_MR ||
+ wr->opcode == IB_WR_REG_MR_INTEGRITY) {
+ fence = dev->umr_fence;
+ next_fence = MLX5_FENCE_MODE_INITIATOR_SMALL;
+ } else {
+ if (wr->send_flags & IB_SEND_FENCE) {
+ if (qp->next_fence)
+ fence = MLX5_FENCE_MODE_SMALL_AND_FENCE;
+ else
+ fence = MLX5_FENCE_MODE_FENCE;
+ } else {
+ fence = qp->next_fence;
+ }
+ }
+
+ switch (ibqp->qp_type) {
+ case IB_QPT_XRC_INI:
+ xrc = seg;
+ seg += sizeof(*xrc);
+ size += sizeof(*xrc) / 16;
+ fallthrough;
+ case IB_QPT_RC:
+ err = handle_qpt_rc(dev, qp, wr, &ctrl, &seg, &size,
+ &cur_edge, &idx, nreq, fence,
+ next_fence, &num_sge);
+ if (unlikely(err)) {
+ *bad_wr = wr;
+ goto out;
+ } else if (wr->opcode == IB_WR_REG_MR_INTEGRITY) {
+ goto skip_psv;
+ }
+ break;
+
+ case IB_QPT_UC:
+ handle_qpt_uc(wr, &seg, &size);
+ break;
+ case IB_QPT_SMI:
+ if (unlikely(!mdev->port_caps[qp->port - 1].has_smi)) {
+ mlx5_ib_warn(dev, "Send SMP MADs is not allowed\n");
+ err = -EPERM;
+ *bad_wr = wr;
+ goto out;
+ }
+ fallthrough;
+ case MLX5_IB_QPT_HW_GSI:
+ handle_qpt_hw_gsi(qp, wr, &seg, &size, &cur_edge);
+ break;
+ case IB_QPT_UD:
+ handle_qpt_ud(qp, wr, &seg, &size, &cur_edge);
+ break;
+ case MLX5_IB_QPT_REG_UMR:
+ err = handle_qpt_reg_umr(dev, qp, wr, &ctrl, &seg,
+ &size, &cur_edge, idx);
+ if (unlikely(err))
+ goto out;
+ break;
+
+ default:
+ break;
+ }
+
+ if (wr->send_flags & IB_SEND_INLINE && num_sge) {
+ err = set_data_inl_seg(qp, wr, &seg, &size, &cur_edge);
+ if (unlikely(err)) {
+ mlx5_ib_warn(dev, "\n");
+ *bad_wr = wr;
+ goto out;
+ }
+ } else {
+ for (i = 0; i < num_sge; i++) {
+ handle_post_send_edge(&qp->sq, &seg, size,
+ &cur_edge);
+ if (unlikely(!wr->sg_list[i].length))
+ continue;
+
+ set_data_ptr_seg(
+ (struct mlx5_wqe_data_seg *)seg,
+ wr->sg_list + i);
+ size += sizeof(struct mlx5_wqe_data_seg) / 16;
+ seg += sizeof(struct mlx5_wqe_data_seg);
+ }
+ }
+
+ qp->next_fence = next_fence;
+ finish_wqe(qp, ctrl, seg, size, cur_edge, idx, wr->wr_id, nreq,
+ fence, mlx5_ib_opcode[wr->opcode]);
+skip_psv:
+ if (0)
+ dump_wqe(qp, idx, size);
+ }
+
+out:
+ if (likely(nreq)) {
+ qp->sq.head += nreq;
+
+ /* Make sure that descriptors are written before
+ * updating doorbell record and ringing the doorbell
+ */
+ wmb();
+
+ qp->db.db[MLX5_SND_DBR] = cpu_to_be32(qp->sq.cur_post);
+
+ /* Make sure doorbell record is visible to the HCA before
+ * we hit doorbell.
+ */
+ wmb();
+
+ mlx5_write64((__be32 *)ctrl, bf->bfreg->map + bf->offset);
+ /* Make sure doorbells don't leak out of SQ spinlock
+ * and reach the HCA out of order.
+ */
+ bf->offset ^= bf->buf_size;
+ }
+
+ spin_unlock_irqrestore(&qp->sq.lock, flags);
+
+ return err;
+}
+
+static void set_sig_seg(struct mlx5_rwqe_sig *sig, int max_gs)
+{
+ sig->signature = calc_sig(sig, (max_gs + 1) << 2);
+}
+
+int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
+ const struct ib_recv_wr **bad_wr, bool drain)
+{
+ struct mlx5_ib_qp *qp = to_mqp(ibqp);
+ struct mlx5_wqe_data_seg *scat;
+ struct mlx5_rwqe_sig *sig;
+ struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
+ struct mlx5_core_dev *mdev = dev->mdev;
+ unsigned long flags;
+ int err = 0;
+ int nreq;
+ int ind;
+ int i;
+
+ if (unlikely(mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR &&
+ !drain)) {
+ *bad_wr = wr;
+ return -EIO;
+ }
+
+ if (unlikely(ibqp->qp_type == IB_QPT_GSI))
+ return mlx5_ib_gsi_post_recv(ibqp, wr, bad_wr);
+
+ spin_lock_irqsave(&qp->rq.lock, flags);
+
+ ind = qp->rq.head & (qp->rq.wqe_cnt - 1);
+
+ for (nreq = 0; wr; nreq++, wr = wr->next) {
+ if (mlx5_wq_overflow(&qp->rq, nreq, qp->ibqp.recv_cq)) {
+ err = -ENOMEM;
+ *bad_wr = wr;
+ goto out;
+ }
+
+ if (unlikely(wr->num_sge > qp->rq.max_gs)) {
+ err = -EINVAL;
+ *bad_wr = wr;
+ goto out;
+ }
+
+ scat = mlx5_frag_buf_get_wqe(&qp->rq.fbc, ind);
+ if (qp->flags_en & MLX5_QP_FLAG_SIGNATURE)
+ scat++;
+
+ for (i = 0; i < wr->num_sge; i++)
+ set_data_ptr_seg(scat + i, wr->sg_list + i);
+
+ if (i < qp->rq.max_gs) {
+ scat[i].byte_count = 0;
+ scat[i].lkey = cpu_to_be32(MLX5_INVALID_LKEY);
+ scat[i].addr = 0;
+ }
+
+ if (qp->flags_en & MLX5_QP_FLAG_SIGNATURE) {
+ sig = (struct mlx5_rwqe_sig *)scat;
+ set_sig_seg(sig, qp->rq.max_gs);
+ }
+
+ qp->rq.wrid[ind] = wr->wr_id;
+
+ ind = (ind + 1) & (qp->rq.wqe_cnt - 1);
+ }
+
+out:
+ if (likely(nreq)) {
+ qp->rq.head += nreq;
+
+ /* Make sure that descriptors are written before
+ * doorbell record.
+ */
+ wmb();
+
+ *qp->db.db = cpu_to_be32(qp->rq.head & 0xffff);
+ }
+
+ spin_unlock_irqrestore(&qp->rq.lock, flags);
+
+ return err;
+}
diff --git a/drivers/infiniband/hw/mlx5/wr.h b/drivers/infiniband/hw/mlx5/wr.h
new file mode 100644
index 000000000000..4f0057516402
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/wr.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/*
+ * Copyright (c) 2020, Mellanox Technologies inc. All rights reserved.
+ */
+
+#ifndef _MLX5_IB_WR_H
+#define _MLX5_IB_WR_H
+
+#include "mlx5_ib.h"
+
+enum {
+ MLX5_IB_SQ_UMR_INLINE_THRESHOLD = 64,
+};
+
+struct mlx5_wqe_eth_pad {
+ u8 rsvd0[16];
+};
+
+
+/* get_sq_edge - Get the next nearby edge.
+ *
+ * An 'edge' is defined as the first following address after the end
+ * of the fragment or the SQ. Accordingly, during the WQE construction
+ * which repetitively increases the pointer to write the next data, it
+ * simply should check if it gets to an edge.
+ *
+ * @sq - SQ buffer.
+ * @idx - Stride index in the SQ buffer.
+ *
+ * Return:
+ * The new edge.
+ */
+static inline void *get_sq_edge(struct mlx5_ib_wq *sq, u32 idx)
+{
+ void *fragment_end;
+
+ fragment_end = mlx5_frag_buf_get_wqe
+ (&sq->fbc,
+ mlx5_frag_buf_get_idx_last_contig_stride(&sq->fbc, idx));
+
+ return fragment_end + MLX5_SEND_WQE_BB;
+}
+
+int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
+ const struct ib_send_wr **bad_wr, bool drain);
+int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
+ const struct ib_recv_wr **bad_wr, bool drain);
+
+static inline int mlx5_ib_post_send_nodrain(struct ib_qp *ibqp,
+ const struct ib_send_wr *wr,
+ const struct ib_send_wr **bad_wr)
+{
+ return mlx5_ib_post_send(ibqp, wr, bad_wr, false);
+}
+
+static inline int mlx5_ib_post_send_drain(struct ib_qp *ibqp,
+ const struct ib_send_wr *wr,
+ const struct ib_send_wr **bad_wr)
+{
+ return mlx5_ib_post_send(ibqp, wr, bad_wr, true);
+}
+
+static inline int mlx5_ib_post_recv_nodrain(struct ib_qp *ibqp,
+ const struct ib_recv_wr *wr,
+ const struct ib_recv_wr **bad_wr)
+{
+ return mlx5_ib_post_recv(ibqp, wr, bad_wr, false);
+}
+
+static inline int mlx5_ib_post_recv_drain(struct ib_qp *ibqp,
+ const struct ib_recv_wr *wr,
+ const struct ib_recv_wr **bad_wr)
+{
+ return mlx5_ib_post_recv(ibqp, wr, bad_wr, true);
+}
+#endif /* _MLX5_IB_WR_H */
diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index 599794c5a78f..7550e9d03dec 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -478,16 +478,6 @@ int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
u32 access, struct mthca_mr *mr);
void mthca_free_mr(struct mthca_dev *dev, struct mthca_mr *mr);
-int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
- u32 access, struct mthca_fmr *fmr);
-int mthca_tavor_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
- int list_len, u64 iova);
-void mthca_tavor_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr);
-int mthca_arbel_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
- int list_len, u64 iova);
-void mthca_arbel_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr);
-int mthca_free_fmr(struct mthca_dev *dev, struct mthca_fmr *fmr);
-
int mthca_map_eq_icm(struct mthca_dev *dev, u64 icm_virt);
void mthca_unmap_eq_icm(struct mthca_dev *dev);
diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c
index 4250b2c18c64..ce0e0867e488 100644
--- a/drivers/infiniband/hw/mthca/mthca_mr.c
+++ b/drivers/infiniband/hw/mthca/mthca_mr.c
@@ -541,7 +541,7 @@ int mthca_mr_alloc_phys(struct mthca_dev *dev, u32 pd,
return err;
}
-/* Free mr or fmr */
+/* Free mr */
static void mthca_free_region(struct mthca_dev *dev, u32 lkey)
{
mthca_table_put(dev, dev->mr_table.mpt_table,
@@ -564,266 +564,6 @@ void mthca_free_mr(struct mthca_dev *dev, struct mthca_mr *mr)
mthca_free_mtt(dev, mr->mtt);
}
-int mthca_fmr_alloc(struct mthca_dev *dev, u32 pd,
- u32 access, struct mthca_fmr *mr)
-{
- struct mthca_mpt_entry *mpt_entry;
- struct mthca_mailbox *mailbox;
- u64 mtt_seg;
- u32 key, idx;
- int list_len = mr->attr.max_pages;
- int err = -ENOMEM;
- int i;
-
- if (mr->attr.page_shift < 12 || mr->attr.page_shift >= 32)
- return -EINVAL;
-
- /* For Arbel, all MTTs must fit in the same page. */
- if (mthca_is_memfree(dev) &&
- mr->attr.max_pages * sizeof *mr->mem.arbel.mtts > PAGE_SIZE)
- return -EINVAL;
-
- mr->maps = 0;
-
- key = mthca_alloc(&dev->mr_table.mpt_alloc);
- if (key == -1)
- return -ENOMEM;
- key = adjust_key(dev, key);
-
- idx = key & (dev->limits.num_mpts - 1);
- mr->ibmr.rkey = mr->ibmr.lkey = hw_index_to_key(dev, key);
-
- if (mthca_is_memfree(dev)) {
- err = mthca_table_get(dev, dev->mr_table.mpt_table, key);
- if (err)
- goto err_out_mpt_free;
-
- mr->mem.arbel.mpt = mthca_table_find(dev->mr_table.mpt_table, key, NULL);
- BUG_ON(!mr->mem.arbel.mpt);
- } else
- mr->mem.tavor.mpt = dev->mr_table.tavor_fmr.mpt_base +
- sizeof *(mr->mem.tavor.mpt) * idx;
-
- mr->mtt = __mthca_alloc_mtt(dev, list_len, dev->mr_table.fmr_mtt_buddy);
- if (IS_ERR(mr->mtt)) {
- err = PTR_ERR(mr->mtt);
- goto err_out_table;
- }
-
- mtt_seg = mr->mtt->first_seg * dev->limits.mtt_seg_size;
-
- if (mthca_is_memfree(dev)) {
- mr->mem.arbel.mtts = mthca_table_find(dev->mr_table.mtt_table,
- mr->mtt->first_seg,
- &mr->mem.arbel.dma_handle);
- BUG_ON(!mr->mem.arbel.mtts);
- } else
- mr->mem.tavor.mtts = dev->mr_table.tavor_fmr.mtt_base + mtt_seg;
-
- mailbox = mthca_alloc_mailbox(dev, GFP_KERNEL);
- if (IS_ERR(mailbox)) {
- err = PTR_ERR(mailbox);
- goto err_out_free_mtt;
- }
-
- mpt_entry = mailbox->buf;
-
- mpt_entry->flags = cpu_to_be32(MTHCA_MPT_FLAG_SW_OWNS |
- MTHCA_MPT_FLAG_MIO |
- MTHCA_MPT_FLAG_REGION |
- access);
-
- mpt_entry->page_size = cpu_to_be32(mr->attr.page_shift - 12);
- mpt_entry->key = cpu_to_be32(key);
- mpt_entry->pd = cpu_to_be32(pd);
- memset(&mpt_entry->start, 0,
- sizeof *mpt_entry - offsetof(struct mthca_mpt_entry, start));
- mpt_entry->mtt_seg = cpu_to_be64(dev->mr_table.mtt_base + mtt_seg);
-
- if (0) {
- mthca_dbg(dev, "Dumping MPT entry %08x:\n", mr->ibmr.lkey);
- for (i = 0; i < sizeof (struct mthca_mpt_entry) / 4; ++i) {
- if (i % 4 == 0)
- printk("[%02x] ", i * 4);
- printk(" %08x", be32_to_cpu(((__be32 *) mpt_entry)[i]));
- if ((i + 1) % 4 == 0)
- printk("\n");
- }
- }
-
- err = mthca_SW2HW_MPT(dev, mailbox,
- key & (dev->limits.num_mpts - 1));
- if (err) {
- mthca_warn(dev, "SW2HW_MPT failed (%d)\n", err);
- goto err_out_mailbox_free;
- }
-
- mthca_free_mailbox(dev, mailbox);
- return 0;
-
-err_out_mailbox_free:
- mthca_free_mailbox(dev, mailbox);
-
-err_out_free_mtt:
- mthca_free_mtt(dev, mr->mtt);
-
-err_out_table:
- mthca_table_put(dev, dev->mr_table.mpt_table, key);
-
-err_out_mpt_free:
- mthca_free(&dev->mr_table.mpt_alloc, key);
- return err;
-}
-
-int mthca_free_fmr(struct mthca_dev *dev, struct mthca_fmr *fmr)
-{
- if (fmr->maps)
- return -EBUSY;
-
- mthca_free_region(dev, fmr->ibmr.lkey);
- mthca_free_mtt(dev, fmr->mtt);
-
- return 0;
-}
-
-static inline int mthca_check_fmr(struct mthca_fmr *fmr, u64 *page_list,
- int list_len, u64 iova)
-{
- int i, page_mask;
-
- if (list_len > fmr->attr.max_pages)
- return -EINVAL;
-
- page_mask = (1 << fmr->attr.page_shift) - 1;
-
- /* We are getting page lists, so va must be page aligned. */
- if (iova & page_mask)
- return -EINVAL;
-
- /* Trust the user not to pass misaligned data in page_list */
- if (0)
- for (i = 0; i < list_len; ++i) {
- if (page_list[i] & ~page_mask)
- return -EINVAL;
- }
-
- if (fmr->maps >= fmr->attr.max_maps)
- return -EINVAL;
-
- return 0;
-}
-
-
-int mthca_tavor_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
- int list_len, u64 iova)
-{
- struct mthca_fmr *fmr = to_mfmr(ibfmr);
- struct mthca_dev *dev = to_mdev(ibfmr->device);
- struct mthca_mpt_entry mpt_entry;
- u32 key;
- int i, err;
-
- err = mthca_check_fmr(fmr, page_list, list_len, iova);
- if (err)
- return err;
-
- ++fmr->maps;
-
- key = tavor_key_to_hw_index(fmr->ibmr.lkey);
- key += dev->limits.num_mpts;
- fmr->ibmr.lkey = fmr->ibmr.rkey = tavor_hw_index_to_key(key);
-
- writeb(MTHCA_MPT_STATUS_SW, fmr->mem.tavor.mpt);
-
- for (i = 0; i < list_len; ++i) {
- __be64 mtt_entry = cpu_to_be64(page_list[i] |
- MTHCA_MTT_FLAG_PRESENT);
- mthca_write64_raw(mtt_entry, fmr->mem.tavor.mtts + i);
- }
-
- mpt_entry.lkey = cpu_to_be32(key);
- mpt_entry.length = cpu_to_be64(list_len * (1ull << fmr->attr.page_shift));
- mpt_entry.start = cpu_to_be64(iova);
-
- __raw_writel((__force u32) mpt_entry.lkey, &fmr->mem.tavor.mpt->key);
- memcpy_toio(&fmr->mem.tavor.mpt->start, &mpt_entry.start,
- offsetof(struct mthca_mpt_entry, window_count) -
- offsetof(struct mthca_mpt_entry, start));
-
- writeb(MTHCA_MPT_STATUS_HW, fmr->mem.tavor.mpt);
-
- return 0;
-}
-
-int mthca_arbel_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
- int list_len, u64 iova)
-{
- struct mthca_fmr *fmr = to_mfmr(ibfmr);
- struct mthca_dev *dev = to_mdev(ibfmr->device);
- u32 key;
- int i, err;
-
- err = mthca_check_fmr(fmr, page_list, list_len, iova);
- if (err)
- return err;
-
- ++fmr->maps;
-
- key = arbel_key_to_hw_index(fmr->ibmr.lkey);
- if (dev->mthca_flags & MTHCA_FLAG_SINAI_OPT)
- key += SINAI_FMR_KEY_INC;
- else
- key += dev->limits.num_mpts;
- fmr->ibmr.lkey = fmr->ibmr.rkey = arbel_hw_index_to_key(key);
-
- *(u8 *) fmr->mem.arbel.mpt = MTHCA_MPT_STATUS_SW;
-
- wmb();
-
- dma_sync_single_for_cpu(&dev->pdev->dev, fmr->mem.arbel.dma_handle,
- list_len * sizeof(u64), DMA_TO_DEVICE);
-
- for (i = 0; i < list_len; ++i)
- fmr->mem.arbel.mtts[i] = cpu_to_be64(page_list[i] |
- MTHCA_MTT_FLAG_PRESENT);
-
- dma_sync_single_for_device(&dev->pdev->dev, fmr->mem.arbel.dma_handle,
- list_len * sizeof(u64), DMA_TO_DEVICE);
-
- fmr->mem.arbel.mpt->key = cpu_to_be32(key);
- fmr->mem.arbel.mpt->lkey = cpu_to_be32(key);
- fmr->mem.arbel.mpt->length = cpu_to_be64(list_len * (1ull << fmr->attr.page_shift));
- fmr->mem.arbel.mpt->start = cpu_to_be64(iova);
-
- wmb();
-
- *(u8 *) fmr->mem.arbel.mpt = MTHCA_MPT_STATUS_HW;
-
- wmb();
-
- return 0;
-}
-
-void mthca_tavor_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr)
-{
- if (!fmr->maps)
- return;
-
- fmr->maps = 0;
-
- writeb(MTHCA_MPT_STATUS_SW, fmr->mem.tavor.mpt);
-}
-
-void mthca_arbel_fmr_unmap(struct mthca_dev *dev, struct mthca_fmr *fmr)
-{
- if (!fmr->maps)
- return;
-
- fmr->maps = 0;
-
- *(u8 *) fmr->mem.arbel.mpt = MTHCA_MPT_STATUS_SW;
-}
-
int mthca_init_mr_table(struct mthca_dev *dev)
{
phys_addr_t addr;
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c
index 69a3e4f62fb1..9fa2f9164a47 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.c
+++ b/drivers/infiniband/hw/mthca/mthca_provider.c
@@ -118,16 +118,6 @@ static int mthca_query_device(struct ib_device *ibdev, struct ib_device_attr *pr
props->max_mcast_qp_attach = MTHCA_QP_PER_MGM;
props->max_total_mcast_qp_attach = props->max_mcast_qp_attach *
props->max_mcast_grp;
- /*
- * If Sinai memory key optimization is being used, then only
- * the 8-bit key portion will change. For other HCAs, the
- * unused index bits will also be used for FMR remapping.
- */
- if (mdev->mthca_flags & MTHCA_FLAG_SINAI_OPT)
- props->max_map_per_fmr = 255;
- else
- props->max_map_per_fmr =
- (1 << (32 - ilog2(mdev->limits.num_mpts))) - 1;
err = 0;
out:
@@ -388,14 +378,15 @@ static void mthca_dealloc_pd(struct ib_pd *pd, struct ib_udata *udata)
mthca_pd_free(to_mdev(pd->device), to_mpd(pd));
}
-static int mthca_ah_create(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
+static int mthca_ah_create(struct ib_ah *ibah,
+ struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
struct mthca_ah *ah = to_mah(ibah);
- return mthca_create_ah(to_mdev(ibah->device), to_mpd(ibah->pd), ah_attr,
- ah);
+ return mthca_create_ah(to_mdev(ibah->device), to_mpd(ibah->pd),
+ init_attr->ah_attr, ah);
}
static void mthca_ah_destroy(struct ib_ah *ah, u32 flags)
@@ -957,69 +948,6 @@ static int mthca_dereg_mr(struct ib_mr *mr, struct ib_udata *udata)
return 0;
}
-static struct ib_fmr *mthca_alloc_fmr(struct ib_pd *pd, int mr_access_flags,
- struct ib_fmr_attr *fmr_attr)
-{
- struct mthca_fmr *fmr;
- int err;
-
- fmr = kmalloc(sizeof *fmr, GFP_KERNEL);
- if (!fmr)
- return ERR_PTR(-ENOMEM);
-
- memcpy(&fmr->attr, fmr_attr, sizeof *fmr_attr);
- err = mthca_fmr_alloc(to_mdev(pd->device), to_mpd(pd)->pd_num,
- convert_access(mr_access_flags), fmr);
-
- if (err) {
- kfree(fmr);
- return ERR_PTR(err);
- }
-
- return &fmr->ibmr;
-}
-
-static int mthca_dealloc_fmr(struct ib_fmr *fmr)
-{
- struct mthca_fmr *mfmr = to_mfmr(fmr);
- int err;
-
- err = mthca_free_fmr(to_mdev(fmr->device), mfmr);
- if (err)
- return err;
-
- kfree(mfmr);
- return 0;
-}
-
-static int mthca_unmap_fmr(struct list_head *fmr_list)
-{
- struct ib_fmr *fmr;
- int err;
- struct mthca_dev *mdev = NULL;
-
- list_for_each_entry(fmr, fmr_list, list) {
- if (mdev && to_mdev(fmr->device) != mdev)
- return -EINVAL;
- mdev = to_mdev(fmr->device);
- }
-
- if (!mdev)
- return 0;
-
- if (mthca_is_memfree(mdev)) {
- list_for_each_entry(fmr, fmr_list, list)
- mthca_arbel_fmr_unmap(mdev, to_mfmr(fmr));
-
- wmb();
- } else
- list_for_each_entry(fmr, fmr_list, list)
- mthca_tavor_fmr_unmap(mdev, to_mfmr(fmr));
-
- err = mthca_SYNC_TPT(mdev);
- return err;
-}
-
static ssize_t hw_rev_show(struct device *device,
struct device_attribute *attr, char *buf)
{
@@ -1203,20 +1131,6 @@ static const struct ib_device_ops mthca_dev_tavor_srq_ops = {
INIT_RDMA_OBJ_SIZE(ib_srq, mthca_srq, ibsrq),
};
-static const struct ib_device_ops mthca_dev_arbel_fmr_ops = {
- .alloc_fmr = mthca_alloc_fmr,
- .dealloc_fmr = mthca_dealloc_fmr,
- .map_phys_fmr = mthca_arbel_map_phys_fmr,
- .unmap_fmr = mthca_unmap_fmr,
-};
-
-static const struct ib_device_ops mthca_dev_tavor_fmr_ops = {
- .alloc_fmr = mthca_alloc_fmr,
- .dealloc_fmr = mthca_dealloc_fmr,
- .map_phys_fmr = mthca_tavor_map_phys_fmr,
- .unmap_fmr = mthca_unmap_fmr,
-};
-
static const struct ib_device_ops mthca_dev_arbel_ops = {
.post_recv = mthca_arbel_post_receive,
.post_send = mthca_arbel_post_send,
@@ -1275,15 +1189,6 @@ int mthca_register_device(struct mthca_dev *dev)
&mthca_dev_tavor_srq_ops);
}
- if (dev->mthca_flags & MTHCA_FLAG_FMR) {
- if (mthca_is_memfree(dev))
- ib_set_device_ops(&dev->ib_dev,
- &mthca_dev_arbel_fmr_ops);
- else
- ib_set_device_ops(&dev->ib_dev,
- &mthca_dev_tavor_fmr_ops);
- }
-
ib_set_device_ops(&dev->ib_dev, &mthca_dev_ops);
if (mthca_is_memfree(dev))
diff --git a/drivers/infiniband/hw/mthca/mthca_provider.h b/drivers/infiniband/hw/mthca/mthca_provider.h
index 596acc45569b..84c64bff0d92 100644
--- a/drivers/infiniband/hw/mthca/mthca_provider.h
+++ b/drivers/infiniband/hw/mthca/mthca_provider.h
@@ -76,24 +76,6 @@ struct mthca_mr {
struct mthca_mtt *mtt;
};
-struct mthca_fmr {
- struct ib_fmr ibmr;
- struct ib_fmr_attr attr;
- struct mthca_mtt *mtt;
- int maps;
- union {
- struct {
- struct mthca_mpt_entry __iomem *mpt;
- u64 __iomem *mtts;
- } tavor;
- struct {
- struct mthca_mpt_entry *mpt;
- __be64 *mtts;
- dma_addr_t dma_handle;
- } arbel;
- } mem;
-};
-
struct mthca_pd {
struct ib_pd ibpd;
u32 pd_num;
@@ -301,11 +283,6 @@ static inline struct mthca_ucontext *to_mucontext(struct ib_ucontext *ibucontext
return container_of(ibucontext, struct mthca_ucontext, ibucontext);
}
-static inline struct mthca_fmr *to_mfmr(struct ib_fmr *ibmr)
-{
- return container_of(ibmr, struct mthca_fmr, ibmr);
-}
-
static inline struct mthca_mr *to_mmr(struct ib_mr *ibmr)
{
return container_of(ibmr, struct mthca_mr, ibmr);
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma.h b/drivers/infiniband/hw/ocrdma/ocrdma.h
index 7baedc74e39d..fcfe0e82197a 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma.h
+++ b/drivers/infiniband/hw/ocrdma/ocrdma.h
@@ -98,7 +98,6 @@ struct ocrdma_dev_attr {
u64 max_mr_size;
u32 max_num_mr_pbl;
int max_mw;
- int max_fmr;
int max_map_per_fmr;
int max_pages_per_frmr;
u16 max_ord_per_qp;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
index 2b7f00ac41b0..6eea02b18968 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
@@ -155,7 +155,7 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah,
return status;
}
-int ocrdma_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr, u32 flags,
+int ocrdma_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata)
{
u32 *ahid_addr;
@@ -165,6 +165,7 @@ int ocrdma_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr, u32 flags,
u16 vlan_tag = 0xffff;
const struct ib_gid_attr *sgid_attr;
struct ocrdma_pd *pd = get_ocrdma_pd(ibah->pd);
+ struct rdma_ah_attr *attr = init_attr->ah_attr;
struct ocrdma_dev *dev = get_ocrdma_dev(ibah->device);
if ((attr->type != RDMA_AH_ATTR_TYPE_ROCE) ||
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h
index 9780afcde780..8b73b3489f3a 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h
@@ -51,7 +51,7 @@ enum {
OCRDMA_AH_L3_TYPE_SHIFT = 0x1D /* 29 bits */
};
-int ocrdma_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr, u32 flags,
+int ocrdma_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
void ocrdma_destroy_ah(struct ib_ah *ah, u32 flags);
int ocrdma_query_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr);
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
index d82d3ec3649e..e07bf0b2209a 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
@@ -1190,7 +1190,6 @@ static void ocrdma_get_attr(struct ocrdma_dev *dev,
attr->max_mr = rsp->max_mr;
attr->max_mr_size = ((u64)rsp->max_mr_size_hi << 32) |
rsp->max_mr_size_lo;
- attr->max_fmr = 0;
attr->max_pages_per_frmr = rsp->max_pages_per_frmr;
attr->max_num_mr_pbl = rsp->max_num_mr_pbl;
attr->max_cqe = rsp->max_cq_cqes_per_cq &
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index 10e343894595..d11c74390a12 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -99,8 +99,6 @@ int ocrdma_query_device(struct ib_device *ibdev, struct ib_device_attr *attr,
attr->max_mw = dev->attr.max_mw;
attr->max_pd = dev->attr.max_pd;
attr->atomic_cap = 0;
- attr->max_fmr = 0;
- attr->max_map_per_fmr = 0;
attr->max_qp_rd_atom =
min(dev->attr.max_ord_per_qp, dev->attr.max_ird_per_qp);
attr->max_qp_init_rd_atom = dev->attr.max_ord_per_qp;
diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c
index dcdc85a1ab25..ccaedfd53e49 100644
--- a/drivers/infiniband/hw/qedr/main.c
+++ b/drivers/infiniband/hw/qedr/main.c
@@ -632,7 +632,6 @@ static int qedr_set_device_attr(struct qedr_dev *dev)
attr->max_mr_size = qed_attr->max_mr_size;
attr->max_cqe = min_t(u64, qed_attr->max_cqe, QEDR_MAX_CQES);
attr->max_mw = qed_attr->max_mw;
- attr->max_fmr = qed_attr->max_fmr;
attr->max_mr_mw_fmr_pbl = qed_attr->max_mr_mw_fmr_pbl;
attr->max_mr_mw_fmr_size = qed_attr->max_mr_mw_fmr_size;
attr->max_pd = qed_attr->max_pd;
diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h
index 5488dbd59d3c..fdf90ecb2699 100644
--- a/drivers/infiniband/hw/qedr/qedr.h
+++ b/drivers/infiniband/hw/qedr/qedr.h
@@ -103,7 +103,6 @@ struct qedr_device_attr {
u64 max_mr_size;
u32 max_cqe;
u32 max_mw;
- u32 max_fmr;
u32 max_mr_mw_fmr_pbl;
u64 max_mr_mw_fmr_size;
u32 max_pd;
diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index a5bd3adaf90a..9b9e80266367 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -145,8 +145,6 @@ int qedr_query_device(struct ib_device *ibdev,
attr->max_mw = qattr->max_mw;
attr->max_pd = qattr->max_pd;
attr->atomic_cap = dev->atomic_cap;
- attr->max_fmr = qattr->max_fmr;
- attr->max_map_per_fmr = 16;
attr->max_qp_init_rd_atom =
1 << (fls(qattr->max_qp_req_rd_atomic_resc) - 1);
attr->max_qp_rd_atom =
@@ -2750,12 +2748,12 @@ int qedr_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata)
return 0;
}
-int qedr_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr, u32 flags,
+int qedr_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata)
{
struct qedr_ah *ah = get_qedr_ah(ibah);
- rdma_copy_ah_attr(&ah->attr, attr);
+ rdma_copy_ah_attr(&ah->attr, init_attr->ah_attr);
return 0;
}
diff --git a/drivers/infiniband/hw/qedr/verbs.h b/drivers/infiniband/hw/qedr/verbs.h
index 18027844eb87..5e02387e068d 100644
--- a/drivers/infiniband/hw/qedr/verbs.h
+++ b/drivers/infiniband/hw/qedr/verbs.h
@@ -70,7 +70,7 @@ int qedr_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr);
void qedr_destroy_srq(struct ib_srq *ibsrq, struct ib_udata *udata);
int qedr_post_srq_recv(struct ib_srq *ibsrq, const struct ib_recv_wr *wr,
const struct ib_recv_wr **bad_recv_wr);
-int qedr_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *attr, u32 flags,
+int qedr_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
void qedr_destroy_ah(struct ib_ah *ibah, u32 flags);
diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c
index 91d64dd71a8a..8bcbc884e5b6 100644
--- a/drivers/infiniband/hw/qib/qib_iba7322.c
+++ b/drivers/infiniband/hw/qib/qib_iba7322.c
@@ -2375,7 +2375,6 @@ static int qib_7322_bringup_serdes(struct qib_pportdata *ppd)
struct qib_devdata *dd = ppd->dd;
u64 val, guid, ibc;
unsigned long flags;
- int ret = 0;
/*
* SerDes model not in Pd, but still need to
@@ -2510,7 +2509,7 @@ static int qib_7322_bringup_serdes(struct qib_pportdata *ppd)
val | ERR_MASK_N(IBStatusChanged));
/* Always zero until we start messing with SerDes for real */
- return ret;
+ return 0;
}
/**
@@ -6875,7 +6874,7 @@ static int init_sdma_7322_regs(struct qib_pportdata *ppd)
struct qib_devdata *dd = ppd->dd;
unsigned lastbuf, erstbuf;
u64 senddmabufmask[3] = { 0 };
- int n, ret = 0;
+ int n;
qib_write_kreg_port(ppd, krp_senddmabase, ppd->sdma_descq_phys);
qib_sdma_7322_setlengen(ppd);
@@ -6904,7 +6903,7 @@ static int init_sdma_7322_regs(struct qib_pportdata *ppd)
qib_write_kreg_port(ppd, krp_senddmabufmask0, senddmabufmask[0]);
qib_write_kreg_port(ppd, krp_senddmabufmask1, senddmabufmask[1]);
qib_write_kreg_port(ppd, krp_senddmabufmask2, senddmabufmask[2]);
- return ret;
+ return 0;
}
/* sdma_lock must be held */
diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c
index 7508abb6a0fa..7acf9ba5358a 100644
--- a/drivers/infiniband/hw/qib/qib_verbs.c
+++ b/drivers/infiniband/hw/qib/qib_verbs.c
@@ -1460,7 +1460,6 @@ static void qib_fill_device_attr(struct qib_devdata *dd)
rdi->dparms.props.max_cq = ib_qib_max_cqs;
rdi->dparms.props.max_cqe = ib_qib_max_cqes;
rdi->dparms.props.max_ah = ib_qib_max_ahs;
- rdi->dparms.props.max_map_per_fmr = 32767;
rdi->dparms.props.max_qp_rd_atom = QIB_MAX_RDMA_ATOMIC;
rdi->dparms.props.max_qp_init_rd_atom = 255;
rdi->dparms.props.max_srq = ib_qib_max_srqs;
diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
index 71f82339446c..b8a77ce11590 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
@@ -322,7 +322,6 @@ int usnic_ib_query_device(struct ib_device *ibdev,
props->max_mcast_grp = 0;
props->max_mcast_qp_attach = 0;
props->max_total_mcast_qp_attach = 0;
- props->max_map_per_fmr = 0;
/* Owned by Userspace
* max_qp_wr, max_sge, max_sge_rd, max_cqe */
mutex_unlock(&us_ibdev->usdev_lock);
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c
index faf7ecd7b3fa..ccbded2d26ce 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c
@@ -509,9 +509,10 @@ void pvrdma_dealloc_pd(struct ib_pd *pd, struct ib_udata *udata)
*
* @return: 0 on success, otherwise errno.
*/
-int pvrdma_create_ah(struct ib_ah *ibah, struct rdma_ah_attr *ah_attr,
- u32 flags, struct ib_udata *udata)
+int pvrdma_create_ah(struct ib_ah *ibah, struct rdma_ah_init_attr *init_attr,
+ struct ib_udata *udata)
{
+ struct rdma_ah_attr *ah_attr = init_attr->ah_attr;
struct pvrdma_dev *dev = to_vdev(ibah->device);
struct pvrdma_ah *ah = to_vah(ibah);
const struct ib_global_route *grh;
diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h
index e4a48f5c0c85..267702226f10 100644
--- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h
+++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h
@@ -414,7 +414,7 @@ int pvrdma_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
void pvrdma_destroy_cq(struct ib_cq *cq, struct ib_udata *udata);
int pvrdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc);
int pvrdma_req_notify_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags);
-int pvrdma_create_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr, u32 flags,
+int pvrdma_create_ah(struct ib_ah *ah, struct rdma_ah_init_attr *init_attr,
struct ib_udata *udata);
void pvrdma_destroy_ah(struct ib_ah *ah, u32 flags);