552c9ca5a0
WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
108 строки
2.6 KiB
C
108 строки
2.6 KiB
C
/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
|
|
/*
|
|
* Copyright (c) 2011-2013 Los Alamos National Security, LLC. All rights
|
|
* reserved.
|
|
* Copyright (c) 2011 UT-Battelle, LLC. All rights reserved.
|
|
* $COPYRIGHT$
|
|
*
|
|
* Additional copyrights may follow
|
|
*
|
|
* $HEADER$
|
|
*/
|
|
|
|
#include "common_ugni.h"
|
|
|
|
OBJ_CLASS_INSTANCE(opal_common_ugni_endpoint_t, opal_object_t, NULL, NULL);
|
|
|
|
int opal_common_ugni_endpoint_for_proc (opal_common_ugni_device_t *dev, opal_proc_t *peer_proc,
|
|
opal_common_ugni_endpoint_t **ep)
|
|
{
|
|
opal_common_ugni_endpoint_t *endpoint;
|
|
opal_common_ugni_modex_t *modex;
|
|
size_t msg_size;
|
|
int rc;
|
|
|
|
assert (NULL != dev && NULL != ep && peer_proc);
|
|
|
|
endpoint = OBJ_NEW(opal_common_ugni_endpoint_t);
|
|
if (OPAL_UNLIKELY(NULL == endpoint)) {
|
|
assert (0);
|
|
return OPAL_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
|
|
/* Receive the modex */
|
|
rc = opal_modex_recv(&opal_common_ugni_component, peer_proc,
|
|
(void *) &modex, &msg_size);
|
|
if (OPAL_UNLIKELY(OPAL_SUCCESS != rc)) {
|
|
OPAL_OUTPUT((-1, "btl/ugni error receiving modex"));
|
|
return rc;
|
|
}
|
|
|
|
endpoint->ep_rem_addr = modex->addr;
|
|
endpoint->ep_rem_id = modex->id;
|
|
|
|
endpoint->dev = dev;
|
|
|
|
*ep = endpoint;
|
|
|
|
free (modex);
|
|
|
|
return OPAL_SUCCESS;
|
|
}
|
|
|
|
void opal_common_ugni_endpoint_return (opal_common_ugni_endpoint_t *ep)
|
|
{
|
|
assert(NULL != ep);
|
|
|
|
OBJ_RELEASE(ep);
|
|
}
|
|
|
|
int opal_common_ugni_ep_create (opal_common_ugni_endpoint_t *cep, gni_cq_handle_t cq,
|
|
gni_ep_handle_t *ep_handle)
|
|
{
|
|
gni_return_t grc;
|
|
|
|
if (OPAL_UNLIKELY(NULL == cep)) {
|
|
assert (0);
|
|
return OPAL_ERR_BAD_PARAM;
|
|
}
|
|
|
|
/* create a uGNI endpoint handle and bind it to the remote peer */
|
|
grc = GNI_EpCreate (cep->dev->dev_handle, cq, ep_handle);
|
|
if (OPAL_UNLIKELY(GNI_RC_SUCCESS != grc)) {
|
|
return opal_common_rc_ugni_to_opal (grc);
|
|
}
|
|
|
|
grc = GNI_EpBind (*ep_handle, cep->ep_rem_addr, cep->ep_rem_id);
|
|
if (GNI_RC_SUCCESS != grc) {
|
|
GNI_EpDestroy (*ep_handle);
|
|
return opal_common_rc_ugni_to_opal (grc);
|
|
}
|
|
|
|
return OPAL_SUCCESS;
|
|
}
|
|
|
|
int opal_common_ugni_ep_destroy (gni_ep_handle_t *ep)
|
|
{
|
|
int rc;
|
|
|
|
if (NULL == ep || 0 == *ep) {
|
|
return OPAL_SUCCESS;
|
|
}
|
|
|
|
rc = GNI_EpUnbind (*ep);
|
|
if (OPAL_UNLIKELY(GNI_RC_SUCCESS != rc)) {
|
|
/* should warn */
|
|
}
|
|
|
|
GNI_EpDestroy (*ep);
|
|
if (OPAL_UNLIKELY(GNI_RC_SUCCESS != rc)) {
|
|
/* should warn */
|
|
}
|
|
|
|
*ep = 0;
|
|
|
|
return OPAL_SUCCESS;
|
|
}
|
|
|