Commit ca95f802 authored Nov 28, 2018 by Kaike Wan Committed by Jason Gunthorpe Dec 06, 2018

IB/hfi1: Unreserve a reserved request when it is completed



Currently, When a reserved operation is completed, its entry in the send
queue will not be unreserved, which leads to the miscalculation of
qp->s_avail and thus the triggering of a WARN_ON call trace. This patch
fixes the problem by unreserving the reserved operation when it is
completed.

Fixes: 856cc4c2 ("IB/hfi1: Add the capability for reserved operations")
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

parent ff8b67fc

drivers/infiniband/hw/hfi1/rc.c

+2 −0

Original line number	Diff line number	Diff line
		@@ -1157,6 +1157,7 @@ void hfi1_rc_send_complete(struct rvt_qp qp, struct hfi1_opa_header opah)
		if (cmp_psn(wqe->lpsn, qp->s_sending_psn) >= 0 &&
		cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0)
		break;
		rvt_qp_wqe_unreserve(qp, wqe);
		s_last = qp->s_last;
		trace_hfi1_qp_send_completion(qp, wqe, s_last);
		if (++s_last >= qp->s_size)
		@@ -1209,6 +1210,7 @@ static struct rvt_swqe do_rc_completion(struct rvt_qp qp,
		u32 s_last;

		rvt_put_swqe(wqe);
		rvt_qp_wqe_unreserve(qp, wqe);
		s_last = qp->s_last;
		trace_hfi1_qp_send_completion(qp, wqe, s_last);
		if (++s_last >= qp->s_size)

Admin message