Re: [PATCH 3/9] IB: add a helper to safely drain a QP

From: Sagi Grimberg
Date: Mon Nov 23 2015 - 05:29:13 EST

That won't work for iWARP. Is this code new? I didn't see any errors that would result from this code when I tested iSER over
cxgb4 with the old iwarp support patches.


I think I figured out why this works with iWARP.

For iWARP, rdma_disconnect() calls iw_cm_disconnect() with abrupt=0
which would make iw_cm_disconnect() move the QP into SQ_DRAIN state"

int iw_cm_disconnect(struct iw_cm_id *cm_id, int abrupt)

if (qp) {
if (abrupt)
ret = iwcm_modify_qp_err(qp);
ret = iwcm_modify_qp_sqd(qp);

* If both sides are disconnecting the QP could
* already be in ERR or SQD states
ret = 0;

IFAIK, SQD state allows the ULP to post work requests on the send
queue and expect these work requests to FLUSH.

So Maybe we should have:
void ib_drain_qp(struct ib_qp *qp)
struct ib_qp_attr attr = { };
struct ib_stop_cqe rstop, sstop;
struct ib_recv_wr rwr = {}, *bad_rwr;
struct ib_send_wr swr = {}, *bad_swr;
enum ib_qp_state state;
int ret;

if rdma_cap_ib_cm(id->device, id->port_num) {
state = IB_QPS_ERR;
else if rdma_cap_iw_cm(id->device, id->port_num)
state = IB_QPS_SQD;

rwr.wr_cqe = &rstop.cqe;
rstop.cqe.done = ib_stop_done;

swr.wr_cqe = &sstop.cqe;
sstop.cqe.done = ib_stop_done;
swr.send_flags = IB_SEND_SIGNALED;

attr.qp_state = state;
ret = ib_modify_qp(qp, &attr, IB_QP_STATE);
if (ret) {
WARN_ONCE(ret, "failed to drain QP: %d\n", ret);

ret = ib_post_recv(qp, &rwr, &bad_rwr);
if (ret) {
WARN_ONCE(ret, "failed to drain recv queue: %d\n", ret);

ret = ib_post_send(qp, &swr, &bad_swr);
if (ret) {
WARN_ONCE(ret, "failed to drain send queue: %d\n", ret);


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at