static int mca_oob_ud_recv_copy (mca_oob_ud_req_t *dst, mca_oob_ud_req_t *src)
{
int rc, i;
dst->req_rem_data_len = src->req_rem_data_len;
rc = mca_oob_ud_recv_alloc (dst);
if (ORTE_SUCCESS == rc) {
unsigned char *dptr = src->req_uiov[0].iov_base;
for (i = 0 ; i < dst->req_count ; ++i) {
memcpy (dst->req_uiov[i].iov_base, dptr, dst->req_uiov[i].iov_len);
dptr += dst->req_uiov[i].iov_len;
}
}
mca_oob_ud_req_complete (dst, (ORTE_SUCCESS == rc) ? dst->req_rem_data_len : rc);
/* free io vector data */
free (src->req_uiov[0].iov_base);
free (src->req_uiov);
src->req_uiov = NULL;
OBJ_RELEASE(src);
return rc;
}
void mca_oob_ud_req_abort (mca_oob_ud_req_t *req)
{
/* caller should have removed this request from any owner list */
req->req_list = NULL;
if (NULL != req->req_qp) {
mca_oob_ud_qp_data_release (req->req_qp);
req->req_qp = NULL;
}
/* free up request resources */
mca_oob_ud_req_complete (req, ORTE_ERR_INTERUPTED);
}
static void mca_oob_ud_peer_msg_timeout (int fd, short event, void *ctx)
{
mca_oob_ud_peer_t *peer = (mca_oob_ud_peer_t *) ctx;
mca_oob_ud_msg_t *msg = (mca_oob_ud_msg_t *) opal_list_get_first (&peer->peer_flying_messages);
OPAL_THREAD_LOCK(&peer->peer_lock);
if (false == peer->peer_timer.active) {
return;
}
peer->peer_timer.active = false;
OPAL_OUTPUT_VERBOSE((10, mca_oob_base_output, "%s oob:ud:peer_msg_timeout timeout sending to peer "
"%s. first message = %" PRIu64 " which has length %d" , ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&peer->peer_name), msg->hdr->msg_id, msg->wr.sg_list[0].length));
if (peer->peer_timer.tries == 0) {
opal_list_item_t *item;
while (NULL != (item = opal_list_remove_first (&peer->peer_flying_messages))) {
msg = (mca_oob_ud_msg_t *) item;
mca_oob_ud_msg_status_update (msg, MCA_OOB_UD_MSG_STATUS_TIMEOUT);
if (msg->req) {
mca_oob_ud_req_complete (msg->req, ORTE_ERR_TIMEOUT);
}
}
OPAL_THREAD_UNLOCK(&peer->peer_lock);
mca_oob_ud_peer_lost (peer);
return;
}
peer->peer_timer.tries--;
mca_oob_ud_peer_post_all (peer);
mca_oob_ud_peer_start_timer (peer);
OPAL_THREAD_UNLOCK(&peer->peer_lock);
}
int mca_oob_ud_recv_complete (mca_oob_ud_req_t *recv_req)
{
mca_oob_ud_msg_t *dataok;
int i, j, rc = ORTE_SUCCESS;
uint32_t expected;
bool error = false, out_of_order = false;
#if defined(HAVE_VALGRIND)
int iov_index;
#endif
OPAL_OUTPUT_VERBOSE((5, mca_oob_base_output, "%s oob:ud:recv_complete req = %p",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), (void *) recv_req));
if (false == recv_req->req_is_eager) {
for (i = 0, expected = 0 ; i < recv_req->req_packet_count ; ) {
struct ibv_wc wc[10];
rc = ibv_poll_cq (recv_req->req_qp->ib_recv_cq, 10, wc);
for (j = 0 ; j < rc ; ++j) {
if (wc[j].imm_data != expected) {
out_of_order = true;
}
if (IBV_WC_SUCCESS != wc[j].status) {
error = true;
}
OPAL_OUTPUT_VERBOSE((5, mca_oob_base_output, "%s oob:ud:recv_complete wc status = %d. imm data = %d. "
"len = %d", ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), wc[j].status, wc[j].imm_data,
wc[j].byte_len));
expected++;
}
if (rc <= 0) {
break;
}
i += rc;
}
if (i != recv_req->req_packet_count || error || out_of_order) {
/* retry */
recv_req->state = MCA_OOB_UD_REQ_PENDING;
OPAL_OUTPUT_VERBOSE((5, mca_oob_base_output, "%s oob:ud:recv_complete receive incomplete. error: %d, "
"out_of_order: %d packets: %d/%d. rc = %d, errno = %d. flags = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), error, out_of_order, i,
recv_req->req_packet_count, rc, errno, recv_req->req_flags));
mca_oob_ud_recv_try (recv_req);
return ORTE_SUCCESS;
}
OPAL_OUTPUT_VERBOSE((10, mca_oob_base_output, "%s oob:ud:recv_complete data received ok!",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME)));
/* send data ok and wait for ack */
rc = mca_oob_ud_msg_get (recv_req->req_port, recv_req, &recv_req->req_port->listen_qp,
recv_req->req_peer, false, &dataok);
if (ORTE_SUCCESS != rc) {
return rc;
}
dataok->hdr->msg_type = MCA_OOB_UD_MSG_DATA_OK;
dataok->hdr->msg_lcl_ctx = recv_req->req_rem_ctx;
rc = mca_oob_ud_msg_post_send (dataok);
if (ORTE_SUCCESS != rc) {
return rc;
}
}
#if defined(HAVE_VALGRIND)
for (iov_index = 0 ; iov_index < recv_req->req_count ; ++iov_index) {
VALGRIND_MAKE_MEM_DEFINED(recv_req->req_uiov[iov_index].iov_base,
recv_req->req_uiov[iov_index].iov_len);
}
#endif
if (MCA_OOB_UD_REQ_UNEX != recv_req->type) {
mca_oob_ud_req_complete (recv_req, (ORTE_SUCCESS == rc) ? recv_req->req_rem_data_len : rc);
} else {
mca_oob_ud_recv_unex_complete (recv_req);
}
return ORTE_SUCCESS;
}
int mca_oob_ud_recv_try (mca_oob_ud_req_t *recv_req)
{
int rc, data_len;
int wr_count, sge_count, wr_index, sge_index, iov_index;
unsigned int iov_left, iov_offset, packet_size;
const unsigned int mtu = recv_req->req_mtu;
struct timeval aquire_timeout = {0, 500000};
mca_oob_ud_msg_t *rep_msg = NULL;
OPAL_OUTPUT_VERBOSE((10, mca_oob_base_output, "%s oob:ud:recv_try receiving from %s. rem ctx = %p",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&recv_req->req_peer->peer_name),
recv_req->req_rem_ctx));
do {
if (NULL == recv_req->req_qp) {
rc = mca_oob_ud_qp_data_aquire (recv_req->req_port, &recv_req->req_qp);
if (ORTE_SUCCESS != rc) {
break;
}
}
(void) mca_oob_ud_qp_purge (recv_req->req_qp);
rc = mca_oob_ud_msg_get (recv_req->req_port, recv_req, &recv_req->req_port->listen_qp,
recv_req->req_peer, NULL, &rep_msg);
if (ORTE_SUCCESS != rc) {
break;
}
if (NULL == recv_req->req_mr) {
/* allocate space for memory registers */
recv_req->req_mr = (struct ibv_mr **) calloc (recv_req->req_count, sizeof (struct ibv_mr *));
if (NULL == recv_req->req_mr) {
opal_output (0, "%s oob:ud:recv_try error allocating space for memory registers. errno = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), errno);
rc = ORTE_ERR_OUT_OF_RESOURCE;
break;
}
}
rc = mca_oob_ud_register_iov (recv_req->req_uiov, recv_req->req_count,
recv_req->req_mr, recv_req->req_port->device->ib_pd,
mtu, &sge_count, &wr_count, &data_len);
if (ORTE_SUCCESS != rc) {
break;
}
data_len = min(data_len, recv_req->req_rem_data_len);
if (data_len < recv_req->req_rem_data_len && !(recv_req->req_flags & ORTE_RML_TRUNC)) {
/* receive buffers are not big enough and ORTE_RML_TRUNC was not specified.
this is probably an error condition */
rc = ORTE_ERR_BAD_PARAM;
break;
}
wr_count = (data_len + mtu - 1) / mtu;
sge_count += wr_count;
OPAL_OUTPUT_VERBOSE((5, mca_oob_base_output, "%s oob:ud:recv_try receiving %d bytes in %d "
"work requests, %d sges", ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), data_len,
wr_count, sge_count));
recv_req->req_packet_count = wr_count;
if (NULL == recv_req->req_wr.recv) {
/* allocate work requests */
recv_req->req_wr.recv = (struct ibv_recv_wr *) calloc (wr_count, sizeof (struct ibv_recv_wr));
if (NULL == recv_req->req_wr.recv) {
opal_output (0, "%s oob:ud:recv_try error allocating work requests. errno = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), errno);
rc = ORTE_ERR_OUT_OF_RESOURCE;
break;
}
}
if (NULL == recv_req->req_sge) {
/* allocate scatter-gather lists. we need more to hold the grh */
recv_req->req_sge = (struct ibv_sge *) calloc (sge_count, sizeof (struct ibv_sge));
if (NULL == recv_req->req_sge) {
opal_output (0, "%s oob:ud:recv_try error allocating sges. errno = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), errno);
rc = ORTE_ERR_OUT_OF_RESOURCE;
break;
}
}
if (NULL == recv_req->req_grh) {
/* allocate grh buffers */
recv_req->req_grh = (struct ibv_grh *) calloc (wr_count, sizeof (struct ibv_grh));
if (NULL == recv_req->req_grh) {
opal_output (0, "%s oob:ud:recv_try error allocating space for GRHs. errno = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), errno);
rc = ORTE_ERR_OUT_OF_RESOURCE;
break;
}
}
if (NULL == recv_req->req_grh_mr) {
/* register grh buffers */
recv_req->req_grh_mr = ibv_reg_mr (recv_req->req_port->device->ib_pd, recv_req->req_grh,
wr_count * sizeof (struct ibv_grh),
IBV_ACCESS_LOCAL_WRITE);
if (NULL == recv_req->req_grh_mr) {
opal_output (0, "%s oob:ud:recv_try error allocating registering GRH memory. errno = %d",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), errno);
/* could not register memory */
rc = ORTE_ERR_OUT_OF_RESOURCE;
break;
}
}
rc = ORTE_SUCCESS;
iov_left = recv_req->req_uiov[0].iov_len;
iov_offset = 0;
iov_index = 0;
for (wr_index = 0, sge_index = 0 ; wr_index < wr_count ; ++wr_index) {
int sge_first = sge_index;
packet_size = 0;
/* grh */
mca_oob_ud_fill_sge(recv_req->req_sge + sge_index++,
recv_req->req_grh + wr_index,
sizeof (struct ibv_grh),
recv_req->req_grh_mr->lkey);
do {
int to_recv = min (iov_left, mtu - packet_size);
mca_oob_ud_fill_sge(recv_req->req_sge + sge_index++,
(char *)recv_req->req_uiov[iov_index].iov_base + iov_offset,
to_recv, recv_req->req_mr[iov_index]->lkey);
iov_offset += to_recv;
iov_left -= to_recv;
packet_size += to_recv;
if (0 == iov_left) {
iov_index++;
iov_offset = 0;
if (iov_index < recv_req->req_count) {
iov_left = recv_req->req_uiov[iov_index].iov_len;
}
}
} while ((packet_size < mtu) && (iov_left > 0));
mca_oob_ud_fill_recv_wr(recv_req->req_wr.recv + wr_index,
recv_req->req_sge + sge_first,
sge_index - sge_first);
if (wr_index + 1 < wr_count) {
recv_req->req_wr.recv[wr_index].next = recv_req->req_wr.recv + wr_index + 1;
}
}
rc = mca_oob_ud_qp_post_recv (recv_req->req_qp, recv_req->req_wr.recv);
if (ORTE_SUCCESS != rc) {
break;
}
OPAL_OUTPUT_VERBOSE((5, mca_oob_base_output, "%s oob:ud:recv_try posting reply message",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME)));
/* ok, we have a data queue pair */
rep_msg->hdr->msg_type = MCA_OOB_UD_MSG_REPLY;
rep_msg->hdr->msg_lcl_ctx = recv_req->req_rem_ctx;
rep_msg->hdr->msg_rem_ctx = recv_req;
rep_msg->hdr->msg_data.rep.qpn = recv_req->req_qp->ib_qp->qp_num;
rep_msg->hdr->msg_data.rep.data_len = data_len;
rep_msg->hdr->msg_data.rep.mtu = mtu;
rc = mca_oob_ud_msg_post_send (rep_msg);
/* post send already returned the message */
rep_msg = NULL;
} while (0);
if (ORTE_ERR_TEMP_OUT_OF_RESOURCE == rc) {
mca_oob_ud_req_timer_set (recv_req, &aquire_timeout, 1, mca_oob_ud_recv_try_to);
rc = ORTE_SUCCESS;
}
if (ORTE_SUCCESS != rc) {
/* bad stuff happened */
if (MCA_OOB_UD_REQ_UNEX != recv_req->type) {
mca_oob_ud_req_complete (recv_req, rc);
}
OBJ_RELEASE(recv_req);
return rc;
}
recv_req->state = MCA_OOB_UD_REQ_ACTIVE;
return rc;
}
int mca_oob_ud_send_complete (mca_oob_ud_req_t *send_req, int rc)
{
mca_oob_ud_req_complete (send_req, rc);
return rc;
}
