sizeof(qp_cmd),
(struct efa_admin_acq_entry *)&cmd_completion,
sizeof(cmd_completion));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev, "Failed to destroy qp-%u [%d]\n",
qp_cmd.qp_handle, err);
+ return err;
+ }
return 0;
}
(struct efa_admin_acq_entry *)&destroy_resp,
sizeof(destroy_resp));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev, "Failed to destroy CQ-%u [%d]\n",
params->cq_idx, err);
+ return err;
+ }
return 0;
}
sizeof(mr_cmd),
(struct efa_admin_acq_entry *)&cmd_completion,
sizeof(cmd_completion));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev,
"Failed to de-register mr(lkey-%u) [%d]\n",
mr_cmd.l_key, err);
+ return err;
+ }
return 0;
}
sizeof(ah_cmd),
(struct efa_admin_acq_entry *)&cmd_completion,
sizeof(cmd_completion));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev, "Failed to destroy ah-%d pd-%d [%d]\n",
ah_cmd.ah, ah_cmd.pd, err);
+ return err;
+ }
return 0;
}
get_resp,
sizeof(*get_resp));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev,
"Failed to submit get_feature command %d [%d]\n",
feature_id, err);
+ return err;
+ }
return 0;
}
(struct efa_admin_acq_entry *)set_resp,
sizeof(*set_resp));
- if (err)
+ if (err) {
ibdev_err(edev->efa_dev,
"Failed to submit set_feature command %d error: %d\n",
feature_id, err);
+ return err;
+ }
return 0;
}
void *obj, u64 address, u64 length, u8 mmap_flag)
{
struct efa_mmap_entry *entry;
+ u32 next_mmap_page;
int err;
entry = kmalloc(sizeof(*entry), GFP_KERNEL);
entry->mmap_flag = mmap_flag;
xa_lock(&ucontext->mmap_xa);
+ if (check_add_overflow(ucontext->mmap_xa_page,
+ (u32)(length >> PAGE_SHIFT),
+ &next_mmap_page))
+ goto err_unlock;
+
entry->mmap_page = ucontext->mmap_xa_page;
- ucontext->mmap_xa_page += DIV_ROUND_UP(length, PAGE_SIZE);
+ ucontext->mmap_xa_page = next_mmap_page;
err = __xa_insert(&ucontext->mmap_xa, entry->mmap_page, entry,
GFP_KERNEL);
+ if (err)
+ goto err_unlock;
+
xa_unlock(&ucontext->mmap_xa);
- if (err){
- kfree(entry);
- return EFA_MMAP_INVALID;
- }
ibdev_dbg(
&dev->ibdev,
entry->obj, entry->address, entry->length, get_mmap_key(entry));
return get_mmap_key(entry);
+
+err_unlock:
+ xa_unlock(&ucontext->mmap_xa);
+ kfree(entry);
+ return EFA_MMAP_INVALID;
+
}
int efa_query_device(struct ib_device *ibdev,
}
/**
+ * hfi1_get_qp_map
+ * @dd: device data
+ * @idx: index to read
+ */
+u8 hfi1_get_qp_map(struct hfi1_devdata *dd, u8 idx)
+{
+ u64 reg = read_csr(dd, RCV_QP_MAP_TABLE + (idx / 8) * 8);
+
+ reg >>= (idx % 8) * 8;
+ return reg;
+}
+
+/**
* init_qpmap_table
* @dd - device data
* @first_ctxt - first context
void remap_intr(struct hfi1_devdata *dd, int isrc, int msix_intr);
void remap_sdma_interrupts(struct hfi1_devdata *dd, int engine, int msix_intr);
void reset_interrupts(struct hfi1_devdata *dd);
+u8 hfi1_get_qp_map(struct hfi1_devdata *dd, u8 idx);
/*
* Interrupt source table.
char *dash;
unsigned long range_start, range_end, i;
bool remove = false;
+ unsigned long bound = 1U << BITS_PER_BYTE;
end = strchr(ptr, ',');
if (end)
BITS_PER_BYTE);
break;
}
+ /* Check the inputs */
+ if (range_start >= bound || range_end >= bound)
+ break;
+
for (i = range_start; i <= range_end; i++) {
if (remove)
clear_bit(i, fault->opcodes);
mgmt->src_qpn = cpu_to_be32(src_qp & OPA_16B_MGMT_QPN_MASK);
}
+/**
+ * hfi1_get_rc_ohdr - get extended header
+ * @opah - the opaheader
+ */
+static inline struct ib_other_headers *
+hfi1_get_rc_ohdr(struct hfi1_opa_header *opah)
+{
+ struct ib_other_headers *ohdr;
+ struct ib_header *hdr = NULL;
+ struct hfi1_16b_header *hdr_16b = NULL;
+
+ /* Find out where the BTH is */
+ if (opah->hdr_type == HFI1_PKT_TYPE_9B) {
+ hdr = &opah->ibh;
+ if (ib_get_lnh(hdr) == HFI1_LRH_BTH)
+ ohdr = &hdr->u.oth;
+ else
+ ohdr = &hdr->u.l.oth;
+ } else {
+ u8 l4;
+
+ hdr_16b = &opah->opah;
+ l4 = hfi1_16B_get_l4(hdr_16b);
+ if (l4 == OPA_16B_L4_IB_LOCAL)
+ ohdr = &hdr_16b->u.oth;
+ else
+ ohdr = &hdr_16b->u.l.oth;
+ }
+ return ohdr;
+}
+
struct rvt_sge_state;
/*
}
}
spin_unlock(&sc->release_lock);
+
+ write_seqlock(&sc->waitlock);
+ while (!list_empty(&sc->piowait)) {
+ struct iowait *wait;
+ struct rvt_qp *qp;
+ struct hfi1_qp_priv *priv;
+
+ wait = list_first_entry(&sc->piowait, struct iowait, list);
+ qp = iowait_to_qp(wait);
+ priv = qp->priv;
+ list_del_init(&priv->s_iowait.list);
+ priv->s_iowait.lock = NULL;
+ hfi1_qp_wakeup(qp, RVT_S_WAIT_PIO | HFI1_S_WAIT_PIO_DRAIN);
+ }
+ write_sequnlock(&sc->waitlock);
+
spin_unlock_irq(&sc->alloc_lock);
}
* @cb: optional callback to call when the buffer is finished sending
* @arg: argument for cb
*
- * Return a pointer to a PIO buffer if successful, NULL if not enough room.
+ * Return a pointer to a PIO buffer, NULL if not enough room, -ECOMM
+ * when link is down.
*/
struct pio_buf *sc_buffer_alloc(struct send_context *sc, u32 dw_len,
pio_release_cb cb, void *arg)
spin_lock_irqsave(&sc->alloc_lock, flags);
if (!(sc->flags & SCF_ENABLED)) {
spin_unlock_irqrestore(&sc->alloc_lock, flags);
- goto done;
+ return ERR_PTR(-ECOMM);
}
retry:
pbc = create_pbc(ppd, pbc_flags, qp->srate_mbps,
sc_to_vlt(ppd->dd, sc5), plen);
pbuf = sc_buffer_alloc(rcd->sc, plen, NULL, NULL);
- if (!pbuf) {
+ if (IS_ERR_OR_NULL(pbuf)) {
/*
* We have no room to send at the moment. Pass
* responsibility for sending the ACK to the send engine
}
}
+/**
+ * hfi1_rc_verbs_aborted - handle abort status
+ * @qp: the QP
+ * @opah: the opa header
+ *
+ * This code modifies both ACK bit in BTH[2]
+ * and the s_flags to go into send one mode.
+ *
+ * This serves to throttle the send engine to only
+ * send a single packet in the likely case the
+ * a link has gone down.
+ */
+void hfi1_rc_verbs_aborted(struct rvt_qp *qp, struct hfi1_opa_header *opah)
+{
+ struct ib_other_headers *ohdr = hfi1_get_rc_ohdr(opah);
+ u8 opcode = ib_bth_get_opcode(ohdr);
+ u32 psn;
+
+ /* ignore responses */
+ if ((opcode >= OP(RDMA_READ_RESPONSE_FIRST) &&
+ opcode <= OP(ATOMIC_ACKNOWLEDGE)) ||
+ opcode == TID_OP(READ_RESP) ||
+ opcode == TID_OP(WRITE_RESP))
+ return;
+
+ psn = ib_bth_get_psn(ohdr) | IB_BTH_REQ_ACK;
+ ohdr->bth[2] = cpu_to_be32(psn);
+ qp->s_flags |= RVT_S_SEND_ONE;
+}
+
/*
* This should be called with the QP s_lock held and interrupts disabled.
*/
struct ib_other_headers *ohdr;
struct hfi1_qp_priv *priv = qp->priv;
struct rvt_swqe *wqe;
- struct ib_header *hdr = NULL;
- struct hfi1_16b_header *hdr_16b = NULL;
u32 opcode, head, tail;
u32 psn;
struct tid_rdma_request *req;
if (!(ib_rvt_state_ops[qp->state] & RVT_SEND_OR_FLUSH_OR_RECV_OK))
return;
- /* Find out where the BTH is */
- if (priv->hdr_type == HFI1_PKT_TYPE_9B) {
- hdr = &opah->ibh;
- if (ib_get_lnh(hdr) == HFI1_LRH_BTH)
- ohdr = &hdr->u.oth;
- else
- ohdr = &hdr->u.l.oth;
- } else {
- u8 l4;
-
- hdr_16b = &opah->opah;
- l4 = hfi1_16B_get_l4(hdr_16b);
- if (l4 == OPA_16B_L4_IB_LOCAL)
- ohdr = &hdr_16b->u.oth;
- else
- ohdr = &hdr_16b->u.l.oth;
- }
-
+ ohdr = hfi1_get_rc_ohdr(opah);
opcode = ib_bth_get_opcode(ohdr);
if ((opcode >= OP(RDMA_READ_RESPONSE_FIRST) &&
opcode <= OP(ATOMIC_ACKNOWLEDGE)) ||
struct sdma_txreq *txp, *txp_next;
LIST_HEAD(flushlist);
unsigned long flags;
+ uint seq;
/* flush from head to tail */
sdma_flush_descq(sde);
spin_lock_irqsave(&sde->flushlist_lock, flags);
/* copy flush list */
- list_for_each_entry_safe(txp, txp_next, &sde->flushlist, list) {
- list_del_init(&txp->list);
- list_add_tail(&txp->list, &flushlist);
- }
+ list_splice_init(&sde->flushlist, &flushlist);
spin_unlock_irqrestore(&sde->flushlist_lock, flags);
/* flush from flush list */
list_for_each_entry_safe(txp, txp_next, &flushlist, list)
complete_tx(sde, txp, SDMA_TXREQ_S_ABORTED);
+ /* wakeup QPs orphaned on the dmawait list */
+ do {
+ struct iowait *w, *nw;
+
+ seq = read_seqbegin(&sde->waitlock);
+ if (!list_empty(&sde->dmawait)) {
+ write_seqlock(&sde->waitlock);
+ list_for_each_entry_safe(w, nw, &sde->dmawait, list) {
+ if (w->wakeup) {
+ w->wakeup(w, SDMA_AVAIL_REASON);
+ list_del_init(&w->list);
+ }
+ }
+ write_sequnlock(&sde->waitlock);
+ }
+ } while (read_seqretry(&sde->waitlock, seq));
}
/*
list_add_tail(&tx->list, &sde->flushlist);
spin_unlock(&sde->flushlist_lock);
iowait_inc_wait_count(wait, tx->num_desc);
- schedule_work(&sde->flush_worker);
+ queue_work_on(sde->cpu, system_highpri_wq, &sde->flush_worker);
ret = -ECOMM;
goto unlock;
nodesc:
iowait_inc_wait_count(wait, tx->num_desc);
}
spin_unlock(&sde->flushlist_lock);
- schedule_work(&sde->flush_worker);
+ queue_work_on(sde->cpu, system_highpri_wq, &sde->flush_worker);
ret = -ECOMM;
goto update_tail;
nodesc:
if (qp->ibqp.qp_num == 0)
ctxt = 0;
else
- ctxt = ((qp->ibqp.qp_num >> dd->qos_shift) %
- (dd->n_krcv_queues - 1)) + 1;
-
+ ctxt = hfi1_get_qp_map(dd, qp->ibqp.qp_num >> dd->qos_shift);
return dd->rcd[ctxt];
}
pbc = create_pbc(ppd, pbc_flags, qp->srate_mbps, vl, plen);
if (ctxt) {
pbuf = sc_buffer_alloc(ctxt, plen, NULL, NULL);
- if (pbuf) {
+ if (!IS_ERR_OR_NULL(pbuf)) {
trace_pio_output_ibhdr(ppd->dd, &hdr, sc5);
ppd->dd->pio_inline_send(ppd->dd, pbuf, pbc,
&hdr, hwords);
pbc = create_pbc(ppd, pbc_flags, qp->srate_mbps, vl, plen);
if (ctxt) {
pbuf = sc_buffer_alloc(ctxt, plen, NULL, NULL);
- if (pbuf) {
+ if (!IS_ERR_OR_NULL(pbuf)) {
trace_pio_output_ibhdr(ppd->dd, &hdr, sc5);
ppd->dd->pio_inline_send(ppd->dd, pbuf, pbc,
&hdr, hwords);
{
struct hfi1_user_sdma_pkt_q *pq =
container_of(wait->iow, struct hfi1_user_sdma_pkt_q, busy);
- struct user_sdma_txreq *tx =
- container_of(txreq, struct user_sdma_txreq, txreq);
- if (sdma_progress(sde, seq, txreq)) {
- if (tx->busycount++ < MAX_DEFER_RETRY_COUNT)
- goto eagain;
- }
+ write_seqlock(&sde->waitlock);
+ if (sdma_progress(sde, seq, txreq))
+ goto eagain;
/*
* We are assuming that if the list is enqueued somewhere, it
* is to the dmawait list since that is the only place where
* it is supposed to be enqueued.
*/
xchg(&pq->state, SDMA_PKT_Q_DEFERRED);
- write_seqlock(&sde->waitlock);
if (list_empty(&pq->busy.list)) {
iowait_get_priority(&pq->busy);
iowait_queue(pkts_sent, &pq->busy, &sde->dmawait);
write_sequnlock(&sde->waitlock);
return -EBUSY;
eagain:
+ write_sequnlock(&sde->waitlock);
return -EAGAIN;
}
tx->flags = 0;
tx->req = req;
- tx->busycount = 0;
INIT_LIST_HEAD(&tx->list);
/*
struct list_head list;
struct user_sdma_request *req;
u16 flags;
- unsigned int busycount;
u16 seqnum;
};
struct hfi1_opa_header *hdr;
hdr = &tx->phdr.hdr;
+ if (unlikely(status == SDMA_TXREQ_S_ABORTED))
+ hfi1_rc_verbs_aborted(qp, hdr);
hfi1_rc_send_complete(qp, hdr);
}
spin_unlock(&qp->s_lock);
if (cb)
iowait_pio_inc(&priv->s_iowait);
pbuf = sc_buffer_alloc(sc, plen, cb, qp);
- if (unlikely(!pbuf)) {
+ if (unlikely(IS_ERR_OR_NULL(pbuf))) {
if (cb)
verbs_pio_complete(qp, 0);
- if (ppd->host_link_state != HLS_UP_ACTIVE) {
+ if (IS_ERR(pbuf)) {
/*
* If we have filled the PIO buffers to capacity and are
* not in an active state this request is not going to
&ps->s_txreq->phdr.hdr, ib_is_sc5(sc5));
pio_bail:
+ spin_lock_irqsave(&qp->s_lock, flags);
if (qp->s_wqe) {
- spin_lock_irqsave(&qp->s_lock, flags);
rvt_send_complete(qp, qp->s_wqe, wc_status);
- spin_unlock_irqrestore(&qp->s_lock, flags);
} else if (qp->ibqp.qp_type == IB_QPT_RC) {
- spin_lock_irqsave(&qp->s_lock, flags);
+ if (unlikely(wc_status == IB_WC_GENERAL_ERR))
+ hfi1_rc_verbs_aborted(qp, &ps->s_txreq->phdr.hdr);
hfi1_rc_send_complete(qp, &ps->s_txreq->phdr.hdr);
- spin_unlock_irqrestore(&qp->s_lock, flags);
}
+ spin_unlock_irqrestore(&qp->s_lock, flags);
ret = 0;
u8 ah_to_sc(struct ib_device *ibdev, struct rdma_ah_attr *ah_attr);
+void hfi1_rc_verbs_aborted(struct rvt_qp *qp, struct hfi1_opa_header *opah);
void hfi1_rc_send_complete(struct rvt_qp *qp, struct hfi1_opa_header *opah);
void hfi1_ud_rcv(struct hfi1_packet *packet);
if (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK) {
struct hfi1_qp_priv *priv;
- tx = kmem_cache_alloc(dev->verbs_txreq_cache, GFP_ATOMIC);
+ tx = kmem_cache_alloc(dev->verbs_txreq_cache, VERBS_TXREQ_GFP);
if (tx)
goto out;
priv = qp->priv;
struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev,
struct rvt_qp *qp);
+#define VERBS_TXREQ_GFP (GFP_ATOMIC | __GFP_NOWARN)
static inline struct verbs_txreq *get_txreq(struct hfi1_ibdev *dev,
struct rvt_qp *qp)
__must_hold(&qp->slock)
struct verbs_txreq *tx;
struct hfi1_qp_priv *priv = qp->priv;
- tx = kmem_cache_alloc(dev->verbs_txreq_cache, GFP_ATOMIC);
+ tx = kmem_cache_alloc(dev->verbs_txreq_cache, VERBS_TXREQ_GFP);
if (unlikely(!tx)) {
/* call slow path to get the lock */
tx = __get_txreq(dev, qp);