/* called by the create qp verb */
int rxe_qp_from_init(struct rxe_dev *rxe, struct rxe_qp *qp, struct rxe_pd *pd,
struct ib_qp_init_attr *init, struct ib_udata *udata,
struct ib_pd *ibpd)
{
int err;
struct rxe_cq *rcq = to_rcq(init->recv_cq);
struct rxe_cq *scq = to_rcq(init->send_cq);
struct rxe_srq *srq = init->srq ? to_rsrq(init->srq) : NULL;
struct ib_ucontext *context = udata ? ibpd->uobject->context : NULL;
rxe_add_ref(pd);
rxe_add_ref(rcq);
rxe_add_ref(scq);
if (srq)
rxe_add_ref(srq);
qp->pd = pd;
qp->rcq = rcq;
qp->scq = scq;
qp->srq = srq;
qp->udata = udata;
rxe_qp_init_misc(rxe, qp, init);
err = rxe_qp_init_req(rxe, qp, init, context, udata);
if (err)
goto err1;
err = rxe_qp_init_resp(rxe, qp, init, context, udata);
if (err)
goto err2;
qp->attr.qp_state = IB_QPS_RESET;
qp->valid = 1;
return 0;
err2:
rxe_queue_cleanup(qp->sq.queue);
err1:
if (srq)
rxe_drop_ref(srq);
rxe_drop_ref(scq);
rxe_drop_ref(rcq);
rxe_drop_ref(pd);
return err;
}
int rxe_send(struct rxe_pkt_info *pkt, struct sk_buff *skb)
{
int err;
skb->destructor = rxe_skb_tx_dtor;
skb->sk = pkt->qp->sk->sk;
rxe_add_ref(pkt->qp);
atomic_inc(&pkt->qp->skb_out);
if (skb->protocol == htons(ETH_P_IP)) {
err = ip_local_out(dev_net(skb_dst(skb)->dev), skb->sk, skb);
} else if (skb->protocol == htons(ETH_P_IPV6)) {
err = ip6_local_out(dev_net(skb_dst(skb)->dev), skb->sk, skb);
} else {
pr_err("Unknown layer 3 protocol: %d\n", skb->protocol);
atomic_dec(&pkt->qp->skb_out);
rxe_drop_ref(pkt->qp);
kfree_skb(skb);
return -EINVAL;
}
if (unlikely(net_xmit_eval(err))) {
pr_debug("error sending packet: %d\n", err);
return -EAGAIN;
}
return 0;
}
static struct ib_mr *rxe_reg_user_mr(struct ib_pd *ibpd,
u64 start,
u64 length,
u64 iova,
int access, struct ib_udata *udata)
{
int err;
struct rxe_dev *rxe = to_rdev(ibpd->device);
struct rxe_pd *pd = to_rpd(ibpd);
struct rxe_mem *mr;
mr = rxe_alloc(&rxe->mr_pool);
if (!mr) {
err = -ENOMEM;
goto err2;
}
rxe_add_index(mr);
rxe_add_ref(pd);
err = rxe_mem_init_user(pd, start, length, iova,
access, udata, mr);
if (err)
goto err3;
return &mr->ibmr;
err3:
rxe_drop_ref(pd);
rxe_drop_index(mr);
rxe_drop_ref(mr);
err2:
return ERR_PTR(err);
}
static struct ib_mr *rxe_alloc_mr(struct ib_pd *ibpd, enum ib_mr_type mr_type,
u32 max_num_sg, struct ib_udata *udata)
{
struct rxe_dev *rxe = to_rdev(ibpd->device);
struct rxe_pd *pd = to_rpd(ibpd);
struct rxe_mem *mr;
int err;
if (mr_type != IB_MR_TYPE_MEM_REG)
return ERR_PTR(-EINVAL);
mr = rxe_alloc(&rxe->mr_pool);
if (!mr) {
err = -ENOMEM;
goto err1;
}
rxe_add_index(mr);
rxe_add_ref(pd);
err = rxe_mem_init_fast(pd, max_num_sg, mr);
if (err)
goto err2;
return &mr->ibmr;
err2:
rxe_drop_ref(pd);
rxe_drop_index(mr);
rxe_drop_ref(mr);
err1:
return ERR_PTR(err);
}
static struct ib_mr *rxe_get_dma_mr(struct ib_pd *ibpd, int access)
{
struct rxe_dev *rxe = to_rdev(ibpd->device);
struct rxe_pd *pd = to_rpd(ibpd);
struct rxe_mem *mr;
int err;
mr = rxe_alloc(&rxe->mr_pool);
if (!mr) {
err = -ENOMEM;
goto err1;
}
rxe_add_index(mr);
rxe_add_ref(pd);
err = rxe_mem_init_dma(pd, access, mr);
if (err)
goto err2;
return &mr->ibmr;
err2:
rxe_drop_ref(pd);
rxe_drop_index(mr);
rxe_drop_ref(mr);
err1:
return ERR_PTR(err);
}
static struct ib_ah *rxe_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr)
{
int err;
struct rxe_dev *rxe = to_rdev(ibpd->device);
struct rxe_pd *pd = to_rpd(ibpd);
struct rxe_ah *ah;
err = rxe_av_chk_attr(rxe, attr);
if (err)
goto err1;
ah = rxe_alloc(&rxe->ah_pool);
if (!ah) {
err = -ENOMEM;
goto err1;
}
rxe_add_ref(pd);
ah->pd = pd;
err = rxe_init_av(rxe, attr, &ah->av);
if (err)
goto err2;
return &ah->ibah;
err2:
rxe_drop_ref(pd);
rxe_drop_ref(ah);
err1:
return ERR_PTR(err);
}
static int rxe_create_srq(struct ib_srq *ibsrq, struct ib_srq_init_attr *init,
struct ib_udata *udata)
{
int err;
struct rxe_dev *rxe = to_rdev(ibsrq->device);
struct rxe_pd *pd = to_rpd(ibsrq->pd);
struct rxe_srq *srq = to_rsrq(ibsrq);
struct rxe_create_srq_resp __user *uresp = NULL;
if (udata) {
if (udata->outlen < sizeof(*uresp))
return -EINVAL;
uresp = udata->outbuf;
}
err = rxe_srq_chk_attr(rxe, NULL, &init->attr, IB_SRQ_INIT_MASK);
if (err)
goto err1;
err = rxe_add_to_pool(&rxe->srq_pool, &srq->pelem);
if (err)
goto err1;
rxe_add_ref(pd);
srq->pd = pd;
err = rxe_srq_from_init(rxe, srq, init, udata, uresp);
if (err)
goto err2;
return 0;
err2:
rxe_drop_ref(pd);
rxe_drop_ref(srq);
err1:
return err;
}
static struct ib_srq *rxe_create_srq(struct ib_pd *ibpd,
struct ib_srq_init_attr *init,
struct ib_udata *udata)
{
int err;
struct rxe_dev *rxe = to_rdev(ibpd->device);
struct rxe_pd *pd = to_rpd(ibpd);
struct rxe_srq *srq;
struct ib_ucontext *context = udata ? ibpd->uobject->context : NULL;
err = rxe_srq_chk_attr(rxe, NULL, &init->attr, IB_SRQ_INIT_MASK);
if (err)
goto err1;
srq = rxe_alloc(&rxe->srq_pool);
if (!srq) {
err = -ENOMEM;
goto err1;
}
rxe_add_index(srq);
rxe_add_ref(pd);
srq->pd = pd;
err = rxe_srq_from_init(rxe, srq, init, context, udata);
if (err)
goto err2;
return &srq->ibsrq;
err2:
rxe_drop_ref(pd);
rxe_drop_index(srq);
rxe_drop_ref(srq);
err1:
return ERR_PTR(err);
}
int rxe_requester(void *arg)
{
struct rxe_qp *qp = (struct rxe_qp *)arg;
struct rxe_pkt_info pkt;
struct sk_buff *skb;
struct rxe_send_wqe *wqe;
enum rxe_hdr_mask mask;
int payload;
int mtu;
int opcode;
int ret;
struct rxe_send_wqe rollback_wqe;
u32 rollback_psn;
rxe_add_ref(qp);
next_wqe:
if (unlikely(!qp->valid || qp->req.state == QP_STATE_ERROR))
goto exit;
if (unlikely(qp->req.state == QP_STATE_RESET)) {
qp->req.wqe_index = consumer_index(qp->sq.queue);
qp->req.opcode = -1;
qp->req.need_rd_atomic = 0;
qp->req.wait_psn = 0;
qp->req.need_retry = 0;
goto exit;
}
if (unlikely(qp->req.need_retry)) {
req_retry(qp);
qp->req.need_retry = 0;
}
wqe = req_next_wqe(qp);
if (unlikely(!wqe))
goto exit;
if (wqe->mask & WR_REG_MASK) {
if (wqe->wr.opcode == IB_WR_LOCAL_INV) {
struct rxe_dev *rxe = to_rdev(qp->ibqp.device);
struct rxe_mem *rmr;
rmr = rxe_pool_get_index(&rxe->mr_pool,
wqe->wr.ex.invalidate_rkey >> 8);
if (!rmr) {
pr_err("No mr for key %#x\n",
wqe->wr.ex.invalidate_rkey);
wqe->state = wqe_state_error;
wqe->status = IB_WC_MW_BIND_ERR;
goto exit;
}
rmr->state = RXE_MEM_STATE_FREE;
rxe_drop_ref(rmr);
wqe->state = wqe_state_done;
wqe->status = IB_WC_SUCCESS;
} else if (wqe->wr.opcode == IB_WR_REG_MR) {
struct rxe_mem *rmr = to_rmr(wqe->wr.wr.reg.mr);
rmr->state = RXE_MEM_STATE_VALID;
rmr->access = wqe->wr.wr.reg.access;
rmr->lkey = wqe->wr.wr.reg.key;
rmr->rkey = wqe->wr.wr.reg.key;
wqe->state = wqe_state_done;
wqe->status = IB_WC_SUCCESS;
} else {
goto exit;
}
qp->req.wqe_index = next_index(qp->sq.queue,
qp->req.wqe_index);
goto next_wqe;
}
int rxe_completer(void *arg)
{
struct rxe_qp *qp = (struct rxe_qp *)arg;
struct rxe_dev *rxe = to_rdev(qp->ibqp.device);
struct rxe_send_wqe *wqe = wqe;
struct sk_buff *skb = NULL;
struct rxe_pkt_info *pkt = NULL;
enum comp_state state;
rxe_add_ref(qp);
if (!qp->valid || qp->req.state == QP_STATE_ERROR ||
qp->req.state == QP_STATE_RESET) {
rxe_drain_resp_pkts(qp, qp->valid &&
qp->req.state == QP_STATE_ERROR);
goto exit;
}
if (qp->comp.timeout) {
qp->comp.timeout_retry = 1;
qp->comp.timeout = 0;
} else {
qp->comp.timeout_retry = 0;
}
if (qp->req.need_retry)
goto exit;
state = COMPST_GET_ACK;
while (1) {
pr_debug("qp#%d state = %s\n", qp_num(qp),
comp_state_name[state]);
switch (state) {
case COMPST_GET_ACK:
skb = skb_dequeue(&qp->resp_pkts);
if (skb) {
pkt = SKB_TO_PKT(skb);
qp->comp.timeout_retry = 0;
}
state = COMPST_GET_WQE;
break;
case COMPST_GET_WQE:
state = get_wqe(qp, pkt, &wqe);
break;
case COMPST_CHECK_PSN:
state = check_psn(qp, pkt, wqe);
break;
case COMPST_CHECK_ACK:
state = check_ack(qp, pkt, wqe);
break;
case COMPST_READ:
state = do_read(qp, pkt, wqe);
break;
case COMPST_ATOMIC:
state = do_atomic(qp, pkt, wqe);
break;
case COMPST_WRITE_SEND:
if (wqe->state == wqe_state_pending &&
wqe->last_psn == pkt->psn)
state = COMPST_COMP_ACK;
else
state = COMPST_UPDATE_COMP;
break;
case COMPST_COMP_ACK:
state = complete_ack(qp, pkt, wqe);
break;
case COMPST_COMP_WQE:
state = complete_wqe(qp, pkt, wqe);
break;
case COMPST_UPDATE_COMP:
if (pkt->mask & RXE_END_MASK)
qp->comp.opcode = -1;
else
qp->comp.opcode = pkt->opcode;
if (psn_compare(pkt->psn, qp->comp.psn) >= 0)
qp->comp.psn = (pkt->psn + 1) & BTH_PSN_MASK;
if (qp->req.wait_psn) {
qp->req.wait_psn = 0;
rxe_run_task(&qp->req.task, 1);
}
state = COMPST_DONE;
break;
case COMPST_DONE:
if (pkt) {
rxe_drop_ref(pkt->qp);
kfree_skb(skb);
skb = NULL;
}
goto done;
case COMPST_EXIT:
if (qp->comp.timeout_retry && wqe) {
state = COMPST_ERROR_RETRY;
break;
}
/* re reset the timeout counter if
* (1) QP is type RC
* (2) the QP is alive
* (3) there is a packet sent by the requester that
* might be acked (we still might get spurious
* timeouts but try to keep them as few as possible)
* (4) the timeout parameter is set
*/
if ((qp_type(qp) == IB_QPT_RC) &&
(qp->req.state == QP_STATE_READY) &&
(psn_compare(qp->req.psn, qp->comp.psn) > 0) &&
qp->qp_timeout_jiffies)
mod_timer(&qp->retrans_timer,
jiffies + qp->qp_timeout_jiffies);
WARN_ON_ONCE(skb);
goto exit;
case COMPST_ERROR_RETRY:
/* we come here if the retry timer fired and we did
* not receive a response packet. try to retry the send
* queue if that makes sense and the limits have not
* been exceeded. remember that some timeouts are
* spurious since we do not reset the timer but kick
* it down the road or let it expire
*/
/* there is nothing to retry in this case */
if (!wqe || (wqe->state == wqe_state_posted)) {
WARN_ON_ONCE(skb);
goto exit;
}
if (qp->comp.retry_cnt > 0) {
if (qp->comp.retry_cnt != 7)
qp->comp.retry_cnt--;
/* no point in retrying if we have already
* seen the last ack that the requester could
* have caused
*/
if (psn_compare(qp->req.psn,
qp->comp.psn) > 0) {
/* tell the requester to retry the
* send queue next time around
*/
rxe_counter_inc(rxe,
RXE_CNT_COMP_RETRY);
qp->req.need_retry = 1;
rxe_run_task(&qp->req.task, 1);
}
if (pkt) {
rxe_drop_ref(pkt->qp);
kfree_skb(skb);
skb = NULL;
}
WARN_ON_ONCE(skb);
goto exit;
} else {
rxe_counter_inc(rxe, RXE_CNT_RETRY_EXCEEDED);
wqe->status = IB_WC_RETRY_EXC_ERR;
state = COMPST_ERROR;
}
break;
case COMPST_RNR_RETRY:
if (qp->comp.rnr_retry > 0) {
if (qp->comp.rnr_retry != 7)
qp->comp.rnr_retry--;
qp->req.need_retry = 1;
pr_debug("qp#%d set rnr nak timer\n",
qp_num(qp));
mod_timer(&qp->rnr_nak_timer,
jiffies + rnrnak_jiffies(aeth_syn(pkt)
& ~AETH_TYPE_MASK));
rxe_drop_ref(pkt->qp);
kfree_skb(skb);
skb = NULL;
goto exit;
} else {
rxe_counter_inc(rxe,
RXE_CNT_RNR_RETRY_EXCEEDED);
wqe->status = IB_WC_RNR_RETRY_EXC_ERR;
state = COMPST_ERROR;
}
break;
case COMPST_ERROR:
WARN_ON_ONCE(wqe->status == IB_WC_SUCCESS);
do_complete(qp, wqe);
rxe_qp_error(qp);
if (pkt) {
rxe_drop_ref(pkt->qp);
kfree_skb(skb);
skb = NULL;
}
WARN_ON_ONCE(skb);
goto exit;
}
}
exit:
/* we come here if we are done with processing and want the task to
* exit from the loop calling us
*/
WARN_ON_ONCE(skb);
rxe_drop_ref(qp);
return -EAGAIN;
done:
/* we come here if we have processed a packet we want the task to call
* us again to see if there is anything else to do
*/
WARN_ON_ONCE(skb);
rxe_drop_ref(qp);
return 0;
}