IB/qib: Use rdmavt version of post_send

This patch removes the post_send and post_one_send from the qib driver.
The "posting" of sends will be done by rdmavt which will walk a WQE and
queue work. This patch will still provide the capability to schedule that
work as well as kick the progress. These are provided to the rdmavt layer.

Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This commit is contained in:
Harish Chegondi
2016-01-22 13:07:42 -08:00
committed by Doug Ledford
parent 4bb88e5f84
commit db3ef0eb84
10 changed files with 81 additions and 291 deletions

View File

@@ -377,8 +377,8 @@ static u32 qib_rcv_hdrerr(struct qib_ctxtdata *rcd, struct qib_pportdata *ppd,
spin_lock(&qp->r_lock);
/* Check for valid receive state. */
if (!(ib_qib_state_ops[qp->state] &
QIB_PROCESS_RECV_OK)) {
if (!(ib_rvt_state_ops[qp->state] &
RVT_PROCESS_RECV_OK)) {
ibp->rvp.n_pkt_drops++;
goto unlock;
}
@@ -592,8 +592,8 @@ move_along:
qp->r_flags &= ~RVT_R_RSP_SEND;
spin_lock_irqsave(&qp->s_lock, flags);
if (ib_qib_state_ops[qp->state] &
QIB_PROCESS_OR_FLUSH_SEND)
if (ib_rvt_state_ops[qp->state] &
RVT_PROCESS_OR_FLUSH_SEND)
qib_schedule_send(qp);
spin_unlock_irqrestore(&qp->s_lock, flags);
}