1
1
openmpi/ompi/mca/coll/basic/coll_basic.h
Nathan Hjelm 9c788ff940 coll/basic: fix segmentation fault in neighborhood collectives if the degree
of the topology is higher than the communicator size

It is possible to have a topology degree higher than the size of the communicator.
For example, a periodic cartesian communicator on MPI_COMM_SELF. This will leave
the neighborhood collectives with a request buffer that is too small. This commit
adds a call that will dynamically increase the size of the request buffer if it
is too small.

A better fix would be to create the topology *before* calling the coll_select
routine on a communicator. This will take some discussion and the solution will
not likely be ready anytime soon.

Thanks to Lisandro Dalcin for reporting this.

Original thread: http://www.open-mpi.org/community/lists/devel/2014/08/15713.php

cmr=v1.8.3:reviewer=jsquyres

This commit was SVN r32796.
2014-09-25 17:43:29 +00:00

381 строка
20 KiB
C

/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
/*
* Copyright (c) 2004-2007 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2006 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2008 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2012 Sandia National Laboratories. All rights reserved.
* Copyright (c) 2013-2014 Los Alamos National Security, LLC. All rights
* reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#ifndef MCA_COLL_BASIC_EXPORT_H
#define MCA_COLL_BASIC_EXPORT_H
#include "ompi_config.h"
#include "mpi.h"
#include "opal/mca/mca.h"
#include "ompi/mca/coll/coll.h"
#include "ompi/request/request.h"
#include "ompi/communicator/communicator.h"
BEGIN_C_DECLS
/* Globally exported variables */
OMPI_MODULE_DECLSPEC extern const mca_coll_base_component_2_0_0_t
mca_coll_basic_component;
extern int mca_coll_basic_priority;
extern int mca_coll_basic_crossover;
/* API functions */
int mca_coll_basic_init_query(bool enable_progress_threads,
bool enable_mpi_threads);
mca_coll_base_module_t
*mca_coll_basic_comm_query(struct ompi_communicator_t *comm,
int *priority);
int mca_coll_basic_module_enable(mca_coll_base_module_t *module,
struct ompi_communicator_t *comm);
int mca_coll_basic_allgather_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_allgather_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_allgatherv_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts,
int *disps,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_allgatherv_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts,
int *disps,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_allreduce_intra(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_allreduce_inter(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoall_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoall_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoallv_intra(void *sbuf, int *scounts,
int *sdisps,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts,
int *rdisps,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoallv_inter(void *sbuf, int *scounts,
int *sdisps,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts,
int *rdisps,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoallw_intra(void *sbuf, int *scounts,
int *sdisps,
struct ompi_datatype_t **sdtypes,
void *rbuf, int *rcounts,
int *rdisps,
struct ompi_datatype_t **rdtypes,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_alltoallw_inter(void *sbuf, int *scounts,
int *sdisps,
struct ompi_datatype_t **sdtypes,
void *rbuf, int *rcounts,
int *rdisps,
struct ompi_datatype_t **rdtypes,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_barrier_intra_lin(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_barrier_inter_lin(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_barrier_intra_log(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_bcast_lin_intra(void *buff, int count,
struct ompi_datatype_t *datatype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_bcast_lin_inter(void *buff, int count,
struct ompi_datatype_t *datatype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_bcast_log_intra(void *buff, int count,
struct ompi_datatype_t *datatype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_bcast_log_inter(void *buff, int count,
struct ompi_datatype_t *datatype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_exscan_intra(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_exscan_inter(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_gather_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_gather_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_gatherv_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts, int *disps,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_gatherv_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int *rcounts, int *disps,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_lin_intra(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_lin_inter(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_log_intra(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_log_inter(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_scatter_block_intra(void *sbuf, void *rbuf,
int rcount,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_scatter_block_inter(void *sbuf, void *rbuf,
int rcount,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_scatter_intra(void *sbuf, void *rbuf,
int *rcounts,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_reduce_scatter_inter(void *sbuf, void *rbuf,
int *rcounts,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scan_intra(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scan_inter(void *sbuf, void *rbuf, int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scatter_intra(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scatter_inter(void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scatterv_intra(void *sbuf, int *scounts, int *disps,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_scatterv_inter(void *sbuf, int *scounts, int *disps,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_neighbor_allgather(void *sbuf, int scount,
struct ompi_datatype_t *sdtype, void *rbuf,
int rcount, struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_neighbor_allgatherv(void *sbuf, int scount, struct ompi_datatype_t *sdtype,
void *rbuf, int rcounts[], int disps[], struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm, mca_coll_base_module_t *module);
int mca_coll_basic_neighbor_alltoall(void *sbuf, int scount, struct ompi_datatype_t *sdtype, void *rbuf,
int rcount, struct ompi_datatype_t *rdtype, struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_basic_neighbor_alltoallv(void *sbuf, int scounts[], int sdisps[],
struct ompi_datatype_t *sdtype, void *rbuf, int rcounts[],
int rdisps[], struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm, mca_coll_base_module_t *module);
int mca_coll_basic_neighbor_alltoallw(void *sbuf, int scounts[], MPI_Aint sdisps[],
struct ompi_datatype_t *sdtypes[], void *rbuf, int rcounts[],
MPI_Aint rdisps[], struct ompi_datatype_t *rdtypes[],
struct ompi_communicator_t *comm, mca_coll_base_module_t *module);
int mca_coll_basic_ft_event(int status);
/* Utility functions */
static inline void mca_coll_basic_free_reqs(ompi_request_t ** reqs,
int count)
{
int i;
for (i = 0; i < count; ++i)
ompi_request_free(&reqs[i]);
}
struct mca_coll_basic_module_t {
mca_coll_base_module_t super;
ompi_request_t **mccb_reqs;
int mccb_num_reqs;
};
typedef struct mca_coll_basic_module_t mca_coll_basic_module_t;
static inline int mca_coll_basic_check_for_requests (mca_coll_basic_module_t *basic_module, int max_reqs)
{
if (basic_module->mccb_num_reqs < max_reqs) {
void *tmp;
basic_module->mccb_num_reqs = max_reqs;
tmp = realloc (basic_module->mccb_reqs, sizeof(ompi_request_t *) * basic_module->mccb_num_reqs);
if (NULL == tmp) {
return OMPI_ERR_OUT_OF_RESOURCE;
}
basic_module->mccb_reqs = tmp;
}
return OMPI_SUCCESS;
}
OBJ_CLASS_DECLARATION(mca_coll_basic_module_t);
END_C_DECLS
#endif /* MCA_COLL_BASIC_EXPORT_H */