1
1
openmpi/ompi/mca/coll/libnbc/nbc_ineighbor_alltoall.c
Nathan Hjelm d42e0968b1 coll/libnbc: rewrite parts of libnbc
This commit rewrites parts of libnbc to fix issues identified by
coverity and myself. The changes are as follows:

 - libnbc function would return invalid error codes (internal to
   libnbc) to the mpi layer. These codes names are of the form
   NBC_. They do not match up with the error codes expected by the mpi
   layer. I purged the use of all these error codes with the exception
   of NBC_OK and NBC_CONTINUE in progress. These codes are used to
   identify when a request handle is complete.

 - Handles and schedules were leaked by all collective routines on
   error. A new routine was added to return a collective handle
   (NBC_Return_handle).

 - Temporary buffers containting in/out neighbors for neighborhood
   collectives were always leaked.

 - Neigborhood collectives contained code to handle MPI_IN_PLACE which
   is never a valid input for the send or receive buffer. Stipped this
   code out.

 - Files were inconsistently named. Most are nbc_isomething.c but one
   was named coll_libnbc_ireduce_scatter_block.c.

 - Made the NBC_Schedule "structure" and object so it can be
   retained/released. This may enable the use of schedule caching at a
   later time. More testing will be needed to ensure the caching code
   works. If it doesn't the code should be stripped out completely.

 - Added code to simply common case of scheduling send/recv +
   barrier.

 - Code cleanup for readability.

The code now passes the clang static analyzer.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-08-10 11:53:25 -06:00

172 строки
5.2 KiB
C

/* -*- Mode: C; c-basic-offset:2 ; indent-tabs-mode:nil -*- */
/*
* Copyright (c) 2006 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2006 The Technical University of Chemnitz. All
* rights reserved.
* Copyright (c) 2014 Research Organization for Information Science
* and Technology (RIST). All rights reserved.
* Copyright (c) 2015 Los Alamos National Security, LLC. All rights
* reserved.
*
* Author(s): Torsten Hoefler <htor@cs.indiana.edu>
*
*/
#include "nbc_internal.h"
/* cannot cache schedules because one cannot check locally if the pattern is the same!! */
#undef NBC_CACHE_SCHEDULE
#ifdef NBC_CACHE_SCHEDULE
/* tree comparison function for schedule cache */
int NBC_Ineighbor_alltoall_args_compare(NBC_Ineighbor_alltoall_args *a, NBC_Ineighbor_alltoall_args *b, void *param) {
if( (a->sbuf == b->sbuf) &&
(a->scount == b->scount) &&
(a->stype == b->stype) &&
(a->rbuf == b->rbuf) &&
(a->rcount == b->rcount) &&
(a->rtype == b->rtype) ) {
return 0;
}
if( a->sbuf < b->sbuf ) {
return -1;
}
return +1;
}
#endif
int ompi_coll_libnbc_ineighbor_alltoall(void *sbuf, int scount, MPI_Datatype stype, void *rbuf,
int rcount, MPI_Datatype rtype, struct ompi_communicator_t *comm,
ompi_request_t ** request, struct mca_coll_base_module_2_1_0_t *module) {
int res, indegree, outdegree, *srcs, *dsts;
MPI_Aint sndext, rcvext;
NBC_Handle *handle;
ompi_coll_libnbc_module_t *libnbc_module = (ompi_coll_libnbc_module_t*) module;
NBC_Schedule *schedule;
res = MPI_Type_extent(stype, &sndext);
if (MPI_SUCCESS != res) {
NBC_Error("MPI Error in MPI_Type_extent() (%i)", res);
return res;
}
res = MPI_Type_extent(rtype, &rcvext);
if (MPI_SUCCESS != res) {
NBC_Error("MPI Error in MPI_Type_extent() (%i)", res);
return res;
}
#ifdef NBC_CACHE_SCHEDULE
NBC_Ineighbor_alltoall_args *args, *found, search;
/* search schedule in communicator specific tree */
search.sbuf = sbuf;
search.scount = scount;
search.stype = stype;
search.rbuf = rbuf;
search.rcount = rcount;
search.rtype = rtype;
found = (NBC_Ineighbor_alltoall_args *) hb_tree_search ((hb_tree *) libnbc_module->NBC_Dict[NBC_NEIGHBOR_ALLTOALL],
&search);
if (NULL == found) {
#endif
schedule = OBJ_NEW(NBC_Schedule);
if (OPAL_UNLIKELY(NULL == schedule)) {
return OMPI_ERR_OUT_OF_RESOURCE;
}
res = NBC_Comm_neighbors(comm, &srcs, &indegree, &dsts, &outdegree);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
OBJ_RELEASE(schedule);
return res;
}
for (int i = 0 ; i < indegree ; ++i) {
if (MPI_PROC_NULL != srcs[i]) {
res = NBC_Sched_recv ((char *) rbuf + i * rcount * rcvext, true, rcount, rtype, srcs[i], schedule, false);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
break;
}
}
}
free (srcs);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
OBJ_RELEASE(schedule);
free (dsts);
return res;
}
for (int i = 0 ; i < outdegree ; ++i) {
if (MPI_PROC_NULL != dsts[i]) {
res = NBC_Sched_send ((char *) sbuf + i * scount * sndext, false, scount, stype, dsts[i], schedule, false);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
break;
}
}
}
free (dsts);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
OBJ_RELEASE(schedule);
return res;
}
res = NBC_Sched_commit (schedule);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
OBJ_RELEASE(schedule);
return res;
}
#ifdef NBC_CACHE_SCHEDULE
/* save schedule to tree */
args = (NBC_Ineighbor_alltoall_args *) malloc (sizeof (args));
if (NULL != args) {
args->sbuf = sbuf;
args->scount = scount;
args->stype = stype;
args->rbuf = rbuf;
args->rcount = rcount;
args->rtype = rtype;
args->schedule = schedule;
res = hb_tree_insert ((hb_tree *) libnbc_module->NBC_Dict[NBC_NEIGHBOR_ALLTOALL], args, args, 0);
if (0 == res) {
OBJ_RETAIN(schedule);
/* increase number of elements for A2A */
if (++libnbc_module->NBC_Dict_size[NBC_NEIGHBOR_ALLTOALL] > NBC_SCHED_DICT_UPPER) {
NBC_SchedCache_dictwipe ((hb_tree *) libnbc_module->NBC_Dict[NBC_NEIGHBOR_ALLTOALL],
&libnbc_module->NBC_Dict_size[NBC_NEIGHBOR_ALLTOALL]);
}
} else {
NBC_Error("error in dict_insert() (%i)", res);
free (args);
}
}
} else {
/* found schedule */
schedule = found->schedule;
OBJ_RETAIN(schedule);
}
#endif
res = NBC_Init_handle (comm, &handle, libnbc_module);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
OBJ_RELEASE(schedule);
return res;
}
res = NBC_Start (handle, schedule);
if (OPAL_UNLIKELY(OMPI_SUCCESS != res)) {
NBC_Return_handle (handle);
return OMPI_ERR_OUT_OF_RESOURCE;
}
*request = (ompi_request_t *) handle;
return OMPI_SUCCESS;
}