Skip to contentSkip to conversation

RDMA/rxe: Send last wqe reached event on qp cleanup

PATCH In review Bob Pearson 1 participant 2 messages Jun 2, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 RDMA/rxe: Send last wqe reached event on qp cleanup

The IBA requires:
o11-5.2.5: If the HCA supports SRQ, for RC and UD service, the CI shall generate a Last WQE Reached Affiliated Asynchronous Event on a QP that is in the Error State and is associated with an SRQ when either:
• a CQE is generated for the last WQE, or
• the QP gets in the Error State and there are no more WQEs on the RQ.

This patch implements this behavior in flush_recv_queue() which is called as a result of rxe_qp_error() being called whenever the qp is put into the error state. The rxe responder executes SRQ WQEs directly from the SRQ so there are never more WQES on the RQ.

---
 drivers/infiniband/sw/rxe/rxe_resp.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/infiniband/sw/rxe/rxe_resp.c +10 -1
index 172c8f916470..0c24facd12cb 100644
@@ -1492,8 +1492,17 @@ static void flush_recv_queue(struct rxe_qp *qp, bool notify)
 	struct rxe_recv_wqe *wqe;
 	int err;
 
-	if (qp->srq)
+	if (qp->srq) {
+		if (notify && qp->ibqp.event_handler) {
+			struct ib_event ev;
+
+			ev.device = qp->ibqp.device;
+			ev.element.qp = &qp->ibqp;
+			ev.event = IB_EVENT_QP_LAST_WQE_REACHED;
+			qp->ibqp.event_handler(&ev, qp->ibqp.qp_context);
+		}
 		return;
+	}
 
 	while ((wqe = queue_head(q, q->type))) {
 		if (notify) {
Signature · 1 line
2.39.2
#1 lore raw

Sorry ignore this one. I forgot the for-next.

Bob

Quoted 42 lines · Bob Pearson

The IBA requires:
o11-5.2.5: If the HCA supports SRQ, for RC and UD service, the CI shall generate a Last WQE Reached Affiliated Asynchronous Event on a QP that is in the Error State and is associated with an SRQ when either:
• a CQE is generated for the last WQE, or
• the QP gets in the Error State and there are no more WQEs on the RQ.

This patch implements this behavior in flush_recv_queue() which is called as a result of rxe_qp_error() being called whenever the qp is put into the error state. The rxe responder executes SRQ WQEs directly from the SRQ so there are never more WQES on the RQ.

---
 drivers/infiniband/sw/rxe/rxe_resp.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/infiniband/sw/rxe/rxe_resp.c +10 -1
index 172c8f916470..0c24facd12cb 100644
@@ -1492,8 +1492,17 @@ static void flush_recv_queue(struct rxe_qp *qp, bool notify)
 	struct rxe_recv_wqe *wqe;
 	int err;
 
-	if (qp->srq)
+	if (qp->srq) {
+		if (notify && qp->ibqp.event_handler) {
+			struct ib_event ev;
+
+			ev.device = qp->ibqp.device;
+			ev.element.qp = &qp->ibqp;
+			ev.event = IB_EVENT_QP_LAST_WQE_REACHED;
+			qp->ibqp.event_handler(&ev, qp->ibqp.qp_context);
+		}
 		return;
+	}
 
 	while ((wqe = queue_head(q, q->type))) {
 		if (notify) {