ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
938e7cd8d9
openmpi/ompi/mca/coll/tuned/coll_tuned_allreduce.c

230 строки
10 KiB
C
Исходник Обычный вид История

Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
/*
Update the copyright notices for IU and UTK. This commit was SVN r7999.
2005-11-05 22:57:48 +03:00
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
* Copyright (c) 2004-2006 The University of Tennessee and The University
Update the copyright notices for IU and UTK. This commit was SVN r7999.
2005-11-05 22:57:48 +03:00
* of Tennessee Research Foundation. All rights
* reserved.
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
* 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.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include "mpi.h"
Next step in the project split, mainly source code re-arranging - move files out of toplevel include/ and etc/, moving it into the sub-projects - rather than including config headers with <project>/include, have them as <project> - require all headers to be included with a project prefix, with the exception of the config headers ({opal,orte,ompi}_config.h mpi.h, and mpif.h) This commit was SVN r8985.
2006-02-12 04:33:29 +03:00
#include "ompi/constants.h"
#include "ompi/datatype/datatype.h"
#include "ompi/communicator/communicator.h"
#include "ompi/mca/coll/coll.h"
#include "ompi/mca/coll/base/coll_tags.h"
#include "ompi/mca/pml/pml.h"
#include "ompi/op/op.h"
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
#include "coll_tuned.h"
#include "coll_tuned_topo.h"
/*
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
* ompi_coll_tuned_allreduce_intra_nonoverlapping
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
*
* This function just calls a reduce followed by a broadcast
* both called functions are tuned but they complete sequentially,
* i.e. no additional overlapping
* meaning if the number of segments used is greater than the topo depth
* then once the first segment of data is fully 'reduced' it is not broadcast
* while the reduce continues (cost = cost-reduce + cost-bcast + decision x 3)
*
*/
int
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
ompi_coll_tuned_allreduce_intra_nonoverlapping(void *sbuf, void *rbuf, int count,
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm)
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
{
int err;
int rank;
rank = ompi_comm_rank(comm);
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_nonoverlapping rank %d", rank));
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
/* Reduce to 0 and broadcast. */
if (MPI_IN_PLACE == sbuf) {
if (0 == ompi_comm_rank(comm)) {
err = comm->c_coll.coll_reduce (MPI_IN_PLACE, rbuf, count, dtype, op, 0, comm);
} else {
err = comm->c_coll.coll_reduce (rbuf, NULL, count, dtype, op, 0, comm);
}
} else {
err = comm->c_coll.coll_reduce (sbuf, rbuf, count, dtype, op, 0, comm);
}
if (MPI_SUCCESS != err) {
return err;
}
return comm->c_coll.coll_bcast (rbuf, count, dtype, 0, comm);
}
/*
* Linear functions are copied from the BASIC coll module
* they do not segment the message and are simple implementations
* but for some small number of nodes and/or small data sizes they
* are just as fast as tuned/tree based segmenting operations
* and as such may be selected by the decision functions
* These are copied into this module due to the way we select modules
* in V1. i.e. in V2 we will handle this differently and so will not
* have to duplicate code.
* GEF Oct05 after asking Jeff.
*/
/* copied function (with appropriate renaming) starts here */
/*
* allreduce_intra
*
* Function: - allreduce using other MPI collectives
* Accepts: - same as MPI_Allreduce()
* Returns: - MPI_SUCCESS or error code
*/
int
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
ompi_coll_tuned_allreduce_intra_basic_linear(void *sbuf, void *rbuf, int count,
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm)
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
{
int err;
int rank;
rank = ompi_comm_rank(comm);
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_basic_linear rank %d", rank));
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
/* Reduce to 0 and broadcast. */
if (MPI_IN_PLACE == sbuf) {
if (0 == ompi_comm_rank(comm)) {
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
err = ompi_coll_tuned_reduce_intra_basic_linear (MPI_IN_PLACE, rbuf, count, dtype, op, 0, comm);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
} else {
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
err = ompi_coll_tuned_reduce_intra_basic_linear(rbuf, NULL, count, dtype, op, 0, comm);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
}
} else {
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
err = ompi_coll_tuned_reduce_intra_basic_linear(sbuf, rbuf, count, dtype, op, 0, comm);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
}
if (MPI_SUCCESS != err) {
return err;
}
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
return ompi_coll_tuned_bcast_intra_basic_linear(rbuf, count, dtype, 0, comm);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
}
/* copied function (with appropriate renaming) ends here */
/* The following are used by dynamic and forced rules */
/* publish details of each algorithm and if its forced/fixed/locked in */
/* as you add methods/algorithms you must update this and the query/map routines */
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
/* this routine is called by the component only */
/* this makes sure that the mca parameters are set to their initial values and perms */
/* module does not call this they call the forced_getvalues routine instead */
int ompi_coll_tuned_allreduce_intra_check_forced_init (coll_tuned_force_algorithm_mca_param_indices_t *mca_param_indices)
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
{
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
int rc;
int max_alg = 2;
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
ompi_coll_tuned_forced_max_algorithms[ALLREDUCE] = max_alg;
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
rc = mca_base_param_reg_int (&mca_coll_tuned_component.super.collm_version,
"allreduce_algorithm_count",
"Number of allreduce algorithms available",
false, true, max_alg, NULL);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
mca_param_indices->algorithm_param_index = mca_base_param_reg_int(
&mca_coll_tuned_component.super.collm_version,
"allreduce_algorithm",
"Which allreduce algorithm is used. Can be locked down to any of: 0 ignore, 1 basic linear, 2 nonoverlapping (tuned reduce + tuned bcast)",
false, false, 0, NULL);
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
A few small changes that just expanded in the name of neatness... (1) As pointed out by Torsten after Jeff comment that there are 15 collectives yesterday.. nope.. I have 16 but miss counted them in my ifdefs (I had two #11s). Replaces with enum... (2) Added a readonly MCA param for how many backend algorithms are available per collective (used by benchmarker/STS) This allowed me to remove the tuned query internal functions and replace them with ompi_coll_tuned_forced_max_algorithms[COLL]. (3) I was reading the user forced MCA params for the collectives on each comm create (module init) but I then put the values into a global set of variables (like ompi_coll_tuned_reduce_forced_algorithm). To fix this and make the code neater: (a) The component looks up the MCA param indices on Open if dynamic_rules is set via the ompi_coll_tuned_COLLECTIVE_intra_check_forced_init () call. (b) Got rid of the ompi_coll_ompi_coll_tuned_COLLECTIVE_forced_algorithm/segmentsize/etc globals with a struct that is now cached on the module data hung off the communicator. i.e. done right. (c) On module init if dynamic rules enabled we call a general getvalues routine (in coll_tuned_forced.c) to get the CURRENT values using the MCA param indices and then put them on the modules data segment. A shorter version of getvalues exists for barrier which only needs the algorithm choice This commit was SVN r9663.
2006-04-20 03:42:06 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
mca_param_indices->segsize_param_index = mca_base_param_reg_int(
&mca_coll_tuned_component.super.collm_version,
"allreduce_algorithm_segmentsize",
"Segment size in bytes used by default for allreduce algorithms. Only has meaning if algorithm is forced and supports segmenting. 0 bytes means no segmentation.",
false, false, 0, NULL);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
mca_param_indices->tree_fanout_param_index = mca_base_param_reg_int(
&mca_coll_tuned_component.super.collm_version,
"allreduce_algorithm_tree_fanout",
"Fanout for n-tree used for allreduce algorithms. Only has meaning if algorithm is forced and supports n-tree topo based operation.",
false, false, ompi_coll_tuned_init_tree_fanout, /* get system wide default */
NULL);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
mca_param_indices->chain_fanout_param_index = mca_base_param_reg_int(
&mca_coll_tuned_component.super.collm_version,
"allreduce_algorithm_chain_fanout",
"Fanout for chains used for allreduce algorithms. Only has meaning if algorithm is forced and supports chain topo based operation.",
false, false,
ompi_coll_tuned_init_chain_fanout, /* get system wide default */
NULL);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
return (MPI_SUCCESS);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
}
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
int ompi_coll_tuned_allreduce_intra_do_forced(void *sbuf, void *rbuf, int count,
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm)
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
{
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_do_forced selected algorithm %d",
comm->c_coll_selected_data->user_forced[ALLREDUCE].algorithm));
allow forced algorithms (where the user or *test* suite knows better) to go through the dynamic decision rule interface. (forced algorithms are set with MCA params) fixed some silly verbose output with wrong func name in it etc updates to fixed dec rules. This commit was SVN r7940.
2005-10-31 23:45:50 +03:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
switch (comm->c_coll_selected_data->user_forced[ALLREDUCE].algorithm) {
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
case (0): return ompi_coll_tuned_allreduce_intra_dec_fixed (sbuf, rbuf, count, dtype, op, comm);
case (1): return ompi_coll_tuned_allreduce_intra_basic_linear (sbuf, rbuf, count, dtype, op, comm);
case (2): return ompi_coll_tuned_allreduce_intra_nonoverlapping (sbuf, rbuf, count, dtype, op, comm);
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
default:
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_do_forced attempt to select algorithm %d when only 0-%d is valid?",
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
comm->c_coll_selected_data->user_forced[ALLREDUCE].algorithm,
ompi_coll_tuned_forced_max_algorithms[ALLREDUCE]));
Started to add basic linear functions Also started to add the allreduce algorithms as I test them (i.e. if it goes in its after testing from now on) This commit was SVN r7886.
2005-10-27 03:11:32 +04:00
return (MPI_ERR_ARG);
} /* switch */
}
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
int ompi_coll_tuned_allreduce_intra_do_this(void *sbuf, void *rbuf, int count,
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
int algorithm, int faninout, int segsize)
File based dynamic up and tested... Lots of misc fixes: printfs->opal_output, handles fanin/out correctly for forced ops unused vars, correct calculations on meaning of 'msgsize' for decision functions (varies depending on algorithm), etc This commit was SVN r8113.
2005-11-11 07:49:29 +03:00
{
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_do_this algorithm %d topo fan in/out %d segsize %d",
algorithm, faninout, segsize));
File based dynamic up and tested... Lots of misc fixes: printfs->opal_output, handles fanin/out correctly for forced ops unused vars, correct calculations on meaning of 'msgsize' for decision functions (varies depending on algorithm), etc This commit was SVN r8113.
2005-11-11 07:49:29 +03:00
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
switch (algorithm) {
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
case (0): return ompi_coll_tuned_allreduce_intra_dec_fixed (sbuf, rbuf, count, dtype, op, comm);
case (1): return ompi_coll_tuned_allreduce_intra_basic_linear (sbuf, rbuf, count, dtype, op, comm);
case (2): return ompi_coll_tuned_allreduce_intra_nonoverlapping (sbuf, rbuf, count, dtype, op, comm);
File based dynamic up and tested... Lots of misc fixes: printfs->opal_output, handles fanin/out correctly for forced ops unused vars, correct calculations on meaning of 'msgsize' for decision functions (varies depending on algorithm), etc This commit was SVN r8113.
2005-11-11 07:49:29 +03:00
default:
Update to have public symbols be consistent; use new prefix rule (apparently we've been doing this in opal and orte, but not in ompi yet). All public symbols begin with "ompi_coll_tuned_" (not mca_coll_tuned_) except the component struct. Now this component passes the illegal symbol report with no hits. This commit was SVN r8589.
2005-12-22 16:49:33 +03:00
OPAL_OUTPUT((ompi_coll_tuned_stream,"coll:tuned:allreduce_intra_do_this attempt to select algorithm %d when only 0-%d is valid?",
Correct the bcast problem where we always did a bcast with segzise of 0. Activate the reduce decision function. Others small updates (mostly TAB to spaces). This commit was SVN r12161.
2006-10-18 06:00:46 +04:00
algorithm, ompi_coll_tuned_forced_max_algorithms[ALLREDUCE]));
File based dynamic up and tested... Lots of misc fixes: printfs->opal_output, handles fanin/out correctly for forced ops unused vars, correct calculations on meaning of 'msgsize' for decision functions (varies depending on algorithm), etc This commit was SVN r8113.
2005-11-11 07:49:29 +03:00
return (MPI_ERR_ARG);
} /* switch */
}
Копировать постоянную ссылку