792 строки
26 KiB
C
792 строки
26 KiB
C
/*
|
|
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
|
|
* University Research and Technology
|
|
* Corporation. All rights reserved.
|
|
* Copyright (c) 2004-2005 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.
|
|
* $COPYRIGHT$
|
|
*
|
|
* Additional copyrights may follow
|
|
*
|
|
* $HEADER$
|
|
*/
|
|
/**
|
|
* @file
|
|
*
|
|
* Warning: this is not for the faint of heart -- don't even bother
|
|
* reading this source code if you don't have a strong understanding
|
|
* of nested data structures and pointer math (remember that
|
|
* associativity and order of C operations is *critical* in terms of
|
|
* pointer math!).
|
|
*/
|
|
|
|
#include "ompi_config.h"
|
|
|
|
#include <stdio.h>
|
|
#ifdef HAVE_SCHED_H
|
|
#include <sched.h>
|
|
#endif
|
|
#include <sys/types.h>
|
|
#include <sys/mman.h>
|
|
#include <unistd.h>
|
|
|
|
#include "mpi.h"
|
|
#include "opal/mca/maffinity/maffinity.h"
|
|
#include "opal/mca/maffinity/base/base.h"
|
|
#include "orte/mca/ns/ns.h"
|
|
#include "ompi/communicator/communicator.h"
|
|
#include "ompi/mca/coll/coll.h"
|
|
#include "ompi/mca/coll/base/base.h"
|
|
#include "ompi/mca/mpool/mpool.h"
|
|
#include "ompi/mca/mpool/base/base.h"
|
|
#include "ompi/proc/proc.h"
|
|
#include "coll_sm.h"
|
|
|
|
|
|
/*
|
|
* Global variables
|
|
*/
|
|
uint32_t mca_coll_sm_iov_size = 1;
|
|
int32_t mca_coll_sm_bogus_free_after = 0;
|
|
|
|
|
|
/*
|
|
* Local functions
|
|
*/
|
|
static const struct mca_coll_base_module_1_0_0_t *
|
|
sm_module_init(struct ompi_communicator_t *comm);
|
|
static int sm_module_finalize(struct ompi_communicator_t *comm);
|
|
static bool have_local_peers(ompi_proc_t **procs, size_t size);
|
|
static int bootstrap_init(void);
|
|
static int bootstrap_comm(ompi_communicator_t *comm);
|
|
|
|
|
|
/*
|
|
* Local variables
|
|
*/
|
|
static bool bootstrap_inited = false;
|
|
|
|
|
|
/*
|
|
* Linear set of collective algorithms
|
|
*/
|
|
static const mca_coll_base_module_1_0_0_t module = {
|
|
|
|
/* Initialization / finalization functions */
|
|
|
|
sm_module_init,
|
|
sm_module_finalize,
|
|
|
|
/* Collective function pointers */
|
|
|
|
NULL,
|
|
NULL,
|
|
mca_coll_sm_allreduce_intra,
|
|
NULL,
|
|
NULL,
|
|
NULL,
|
|
mca_coll_sm_barrier_intra,
|
|
mca_coll_sm_bcast_intra,
|
|
NULL,
|
|
NULL,
|
|
NULL,
|
|
mca_coll_sm_reduce_intra,
|
|
NULL,
|
|
NULL,
|
|
NULL,
|
|
NULL
|
|
};
|
|
|
|
|
|
/*
|
|
* Initial query function that is invoked during MPI_INIT, allowing
|
|
* this component to disqualify itself if it doesn't support the
|
|
* required level of thread support. This function is invoked exactly
|
|
* once.
|
|
*/
|
|
int mca_coll_sm_init_query(bool enable_progress_threads,
|
|
bool enable_mpi_threads)
|
|
{
|
|
#if 0
|
|
/* JMS: Arrgh. Unfortunately, we don't have this information by
|
|
the time this is invoked -- the GPR compound command doesn't
|
|
fire until after coll_base_open() in ompi_mpi_init(). */
|
|
|
|
ompi_proc_t **procs;
|
|
size_t size;
|
|
|
|
/* Check to see if anyone is local on this machine. If not, don't
|
|
bother doing anything else. */
|
|
|
|
procs = ompi_proc_all(&size);
|
|
if (NULL == procs || 0 == size) {
|
|
return OMPI_ERROR;
|
|
}
|
|
if (!have_local_peers(procs, size)) {
|
|
return OMPI_ERROR;
|
|
}
|
|
free(procs);
|
|
#endif
|
|
|
|
/* Don't do much here because we don't really want to allocate any
|
|
shared memory until this component is selected to be used. */
|
|
|
|
mca_coll_sm_component.sm_data_mpool_created = false;
|
|
mca_coll_sm_component.sm_component_setup = false;
|
|
opal_atomic_init(&mca_coll_sm_component.sm_component_setup_lock, 0);
|
|
|
|
/* Alles gut */
|
|
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
|
|
/*
|
|
* Invoked when there's a new communicator that has been created.
|
|
* Look at the communicator and decide which set of functions and
|
|
* priority we want to return.
|
|
*/
|
|
const mca_coll_base_module_1_0_0_t *
|
|
mca_coll_sm_comm_query(struct ompi_communicator_t *comm, int *priority,
|
|
struct mca_coll_base_comm_t **data)
|
|
{
|
|
/* See if someone has previously lazily initialized and failed */
|
|
|
|
if (mca_coll_sm_component.sm_component_setup &&
|
|
!mca_coll_sm_component.sm_component_setup_success) {
|
|
return NULL;
|
|
}
|
|
|
|
/* If we're intercomm, or if there's only one process in the
|
|
communicator, or if not all the processes in the communicator
|
|
are not on this node, then we don't want to run */
|
|
|
|
if (OMPI_COMM_IS_INTER(comm) || 1 == ompi_comm_size(comm) ||
|
|
!have_local_peers(comm->c_local_group->grp_proc_pointers,
|
|
ompi_comm_size(comm))) {
|
|
return NULL;
|
|
}
|
|
|
|
/* If the number of processes in this communicator is larger than
|
|
(mca_coll_sm_component.sm_control_size / sizeof(uint32_t)),
|
|
then we can't handle it. */
|
|
|
|
if (((unsigned) ompi_comm_size(comm)) >
|
|
mca_coll_sm_component.sm_control_size / sizeof(uint32_t)) {
|
|
return NULL;
|
|
}
|
|
|
|
/* Get our priority */
|
|
|
|
*priority = mca_coll_sm_component.sm_priority;
|
|
|
|
/* All is good -- return a module */
|
|
|
|
return &module;
|
|
}
|
|
|
|
|
|
/*
|
|
* Unquery the coll on comm
|
|
*/
|
|
int mca_coll_sm_comm_unquery(struct ompi_communicator_t *comm,
|
|
struct mca_coll_base_comm_t *data)
|
|
{
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
|
|
/*
|
|
* Init module on the communicator
|
|
*/
|
|
static const struct mca_coll_base_module_1_0_0_t *
|
|
sm_module_init(struct ompi_communicator_t *comm)
|
|
{
|
|
int i, j, root;
|
|
int rank = ompi_comm_rank(comm);
|
|
int size = ompi_comm_size(comm);
|
|
mca_coll_base_comm_t *data;
|
|
size_t control_size, frag_size;
|
|
mca_coll_sm_component_t *c = &mca_coll_sm_component;
|
|
opal_maffinity_base_segment_t *maffinity;
|
|
int parent, min_child, max_child, num_children;
|
|
char *base;
|
|
const int num_barrier_buffers = 2;
|
|
|
|
/* Once-per-component setup. This may happen at any time --
|
|
during MPI_INIT or later. So we must protect this with locks
|
|
to ensure that only one thread in the process actually does
|
|
this setup. */
|
|
|
|
opal_atomic_lock(&mca_coll_sm_component.sm_component_setup_lock);
|
|
if (!mca_coll_sm_component.sm_component_setup) {
|
|
mca_coll_sm_component.sm_component_setup = true;
|
|
|
|
if (OMPI_SUCCESS != bootstrap_init()) {
|
|
mca_coll_sm_component.sm_component_setup_success = false;
|
|
opal_atomic_unlock(&mca_coll_sm_component.sm_component_setup_lock);
|
|
return NULL;
|
|
}
|
|
|
|
/* Can we get an mpool allocation? See if there was one created
|
|
already. If not, try to make one. */
|
|
|
|
mca_coll_sm_component.sm_data_mpool =
|
|
mca_mpool_base_module_lookup(mca_coll_sm_component.sm_mpool_name);
|
|
if (NULL == mca_coll_sm_component.sm_data_mpool) {
|
|
mca_coll_sm_component.sm_data_mpool =
|
|
mca_mpool_base_module_create(mca_coll_sm_component.sm_mpool_name,
|
|
NULL, NULL);
|
|
if (NULL == mca_coll_sm_component.sm_data_mpool) {
|
|
mca_coll_sm_bootstrap_finalize();
|
|
mca_coll_sm_component.sm_component_setup_success = false;
|
|
opal_atomic_unlock(&mca_coll_sm_component.sm_component_setup_lock);
|
|
return NULL;
|
|
}
|
|
mca_coll_sm_component.sm_data_mpool_created = true;
|
|
} else {
|
|
mca_coll_sm_component.sm_data_mpool_created = false;
|
|
}
|
|
mca_coll_sm_component.sm_component_setup_success = true;
|
|
}
|
|
opal_atomic_unlock(&mca_coll_sm_component.sm_component_setup_lock);
|
|
|
|
/* Double check to see if some interleaved lazy init failed before
|
|
we got in here */
|
|
|
|
if (!mca_coll_sm_component.sm_component_setup_success) {
|
|
return NULL;
|
|
}
|
|
|
|
/* Get some space to setup memory affinity (just easier to try to
|
|
alloc here to handle the error case) */
|
|
|
|
maffinity = malloc(sizeof(opal_maffinity_base_segment_t) *
|
|
c->sm_comm_num_segments * 3);
|
|
if (NULL == maffinity) {
|
|
return NULL;
|
|
}
|
|
|
|
/* Allocate data to hang off the communicator. The memory we
|
|
alloc will be laid out as follows:
|
|
|
|
1. mca_coll_base_comm_t
|
|
2. array of num_segments mca_coll_base_mpool_index_t instances
|
|
(pointed to by the array in 2)
|
|
3. array of ompi_comm_size(comm) mca_coll_sm_tree_node_t
|
|
instances
|
|
4. array of sm_tree_degree pointers to other tree nodes (i.e.,
|
|
this nodes' children) for each instance of
|
|
mca_coll_sm_tree_node_t
|
|
*/
|
|
|
|
comm->c_coll_selected_data = data =
|
|
malloc(sizeof(mca_coll_base_comm_t) +
|
|
(c->sm_comm_num_segments *
|
|
sizeof(mca_coll_base_mpool_index_t)) +
|
|
(size *
|
|
(sizeof(mca_coll_sm_tree_node_t) +
|
|
(sizeof(mca_coll_sm_tree_node_t*) * c->sm_tree_degree))));
|
|
|
|
if (NULL == data) {
|
|
return NULL;
|
|
}
|
|
data->mcb_data_mpool_malloc_addr = NULL;
|
|
|
|
/* Setup #2: set the array to point immediately beyond the
|
|
mca_coll_base_comm_t */
|
|
data->mcb_mpool_index = (mca_coll_base_mpool_index_t*) (data + 1);
|
|
/* Setup array of pointers for #3 */
|
|
data->mcb_tree = (mca_coll_sm_tree_node_t*)
|
|
(data->mcb_mpool_index + c->sm_comm_num_segments);
|
|
/* Finally, setup the array of children pointers in the instances
|
|
in #5 to point to their corresponding arrays in #6 */
|
|
data->mcb_tree[0].mcstn_children = (mca_coll_sm_tree_node_t**)
|
|
(data->mcb_tree + size);
|
|
for (i = 1; i < size; ++i) {
|
|
data->mcb_tree[i].mcstn_children =
|
|
data->mcb_tree[i - 1].mcstn_children + c->sm_tree_degree;
|
|
}
|
|
|
|
/* Pre-compute a tree for a given number of processes and degree.
|
|
We'll re-use this tree for all possible values of root (i.e.,
|
|
shift everyone's process to be the "0"/root in this tree. */
|
|
for (root = 0; root < size; ++root) {
|
|
parent = (root - 1) / mca_coll_sm_component.sm_tree_degree;
|
|
num_children = mca_coll_sm_component.sm_tree_degree;
|
|
|
|
/* Do we have children? If so, how many? */
|
|
|
|
if ((root * num_children) + 1 >= size) {
|
|
/* Leaves */
|
|
min_child = -1;
|
|
max_child = -1;
|
|
num_children = 0;
|
|
} else {
|
|
/* Interior nodes */
|
|
min_child = root * num_children + 1;
|
|
max_child = root * num_children + num_children;
|
|
if (max_child >= size) {
|
|
max_child = size - 1;
|
|
}
|
|
num_children = max_child - min_child + 1;
|
|
}
|
|
|
|
/* Save the values */
|
|
data->mcb_tree[root].mcstn_id = root;
|
|
if (root == 0 && parent == 0) {
|
|
data->mcb_tree[root].mcstn_parent = NULL;
|
|
} else {
|
|
data->mcb_tree[root].mcstn_parent = &data->mcb_tree[parent];
|
|
}
|
|
data->mcb_tree[root].mcstn_num_children = num_children;
|
|
for (i = 0; i < c->sm_tree_degree; ++i) {
|
|
data->mcb_tree[root].mcstn_children[i] =
|
|
(i < num_children) ?
|
|
&data->mcb_tree[min_child + i] : NULL;
|
|
}
|
|
}
|
|
|
|
/* Bootstrap this communicator; find the shared memory in the main
|
|
mpool that has been allocated among my peers for this
|
|
communicator. */
|
|
|
|
if (OMPI_SUCCESS != bootstrap_comm(comm)) {
|
|
free(data);
|
|
comm->c_coll_selected_data = NULL;
|
|
return NULL;
|
|
}
|
|
|
|
/* Once the communicator is bootstrapped, setup the pointers into
|
|
the data mpool area. First, setup the barrier buffers. There
|
|
are 2 sets of barrier buffers (because there can never be more
|
|
than one outstanding barrier occuring at any timie). Setup
|
|
pointers to my control buffers, my parents, and [the beginning
|
|
of] my children (note that the children are contiguous, so
|
|
having the first pointer and the num_children from the mcb_tree
|
|
data is sufficient). */
|
|
|
|
control_size = c->sm_control_size;
|
|
base = (char*) (data->mcb_mpool_base + data->mcb_mpool_offset);
|
|
data->mcb_barrier_control_me = (uint32_t*)
|
|
(base + (rank * control_size * num_barrier_buffers * 2));
|
|
if (data->mcb_tree[rank].mcstn_parent) {
|
|
data->mcb_barrier_control_parent = (uint32_t*)
|
|
(base +
|
|
(data->mcb_tree[rank].mcstn_parent->mcstn_id * control_size *
|
|
num_barrier_buffers * 2));
|
|
} else {
|
|
data->mcb_barrier_control_parent = NULL;
|
|
}
|
|
if (data->mcb_tree[rank].mcstn_num_children > 0) {
|
|
data->mcb_barrier_control_children = (uint32_t*)
|
|
(base +
|
|
(data->mcb_tree[rank].mcstn_children[0]->mcstn_id * control_size *
|
|
num_barrier_buffers * 2));
|
|
} else {
|
|
data->mcb_barrier_control_children = NULL;
|
|
}
|
|
data->mcb_barrier_count = 0;
|
|
|
|
/* Next, setup the pointer to the in-use flags. The number of
|
|
segments will be an even multiple of the number of in-use
|
|
flags. */
|
|
|
|
base += (c->sm_control_size * size * num_barrier_buffers * 2);
|
|
data->mcb_in_use_flags = (mca_coll_sm_in_use_flag_t*) base;
|
|
|
|
/* All things being equal, if we're rank 0, then make the in-use
|
|
flags be local (memory affinity). Then zero them all out so
|
|
that they're marked as unused. */
|
|
|
|
j = 0;
|
|
if (0 == rank) {
|
|
maffinity[j].mbs_start_addr = base;
|
|
maffinity[j].mbs_len = c->sm_control_size *
|
|
c->sm_comm_num_in_use_flags;
|
|
/* Set the op counts to 1 (actually any nonzero value will do)
|
|
so that the first time children/leaf processes come
|
|
through, they don't see a value of 0 and think that the
|
|
root/parent has already set the count to their op number
|
|
(i.e., 0 is the first op count value). */
|
|
for (i = 0; i < mca_coll_sm_component.sm_comm_num_in_use_flags; ++i) {
|
|
((mca_coll_sm_in_use_flag_t *)base)[i].mcsiuf_operation_count = 1;
|
|
((mca_coll_sm_in_use_flag_t *)base)[i].mcsiuf_num_procs_using = 0;
|
|
}
|
|
D(("rank 0 zeroed in-use flags (num %d, len %d): %p - %p\n",
|
|
c->sm_comm_num_in_use_flags,
|
|
maffinity[j].mbs_len,
|
|
base, base + maffinity[j].mbs_len));
|
|
++j;
|
|
}
|
|
|
|
/* Next, setup pointers to the control and data portions of the
|
|
segments, as well as to the relevant in-use flags. */
|
|
|
|
base += (c->sm_comm_num_in_use_flags * c->sm_control_size);
|
|
control_size = size * c->sm_control_size;
|
|
frag_size = size * c->sm_fragment_size;
|
|
for (i = 0; i < c->sm_comm_num_segments; ++i) {
|
|
data->mcb_mpool_index[i].mcbmi_control = (uint32_t*)
|
|
(base + (i * (control_size + frag_size)));
|
|
data->mcb_mpool_index[i].mcbmi_data =
|
|
(((char*) data->mcb_mpool_index[i].mcbmi_control) +
|
|
control_size);
|
|
|
|
/* Memory affinity: control */
|
|
|
|
maffinity[j].mbs_len = c->sm_control_size;
|
|
maffinity[j].mbs_start_addr = (void *)
|
|
(data->mcb_mpool_index[i].mcbmi_control +
|
|
(rank * c->sm_control_size));
|
|
++j;
|
|
|
|
/* Memory affinity: data */
|
|
|
|
maffinity[j].mbs_len = c->sm_fragment_size;
|
|
maffinity[j].mbs_start_addr =
|
|
data->mcb_mpool_index[i].mcbmi_data +
|
|
(rank * c->sm_control_size);
|
|
++j;
|
|
}
|
|
|
|
/* Setup memory affinity so that the pages that belong to this
|
|
process are local to this process */
|
|
|
|
opal_maffinity_base_set(maffinity, j);
|
|
free(maffinity);
|
|
|
|
/* Zero out the control structures that belong to this process */
|
|
|
|
memset(data->mcb_barrier_control_me, 0,
|
|
num_barrier_buffers * 2 * c->sm_control_size);
|
|
for (i = 0; i < c->sm_comm_num_segments; ++i) {
|
|
memset((void *) data->mcb_mpool_index[i].mcbmi_control, 0,
|
|
c->sm_control_size);
|
|
}
|
|
|
|
/* All done */
|
|
|
|
return &module;
|
|
}
|
|
|
|
|
|
/*
|
|
* Finalize module on the communicator
|
|
*/
|
|
static int sm_module_finalize(struct ompi_communicator_t *comm)
|
|
{
|
|
mca_coll_base_comm_t *data;
|
|
|
|
/* Free the space in the data mpool and the data hanging off the
|
|
communicator */
|
|
|
|
data = comm->c_coll_selected_data;
|
|
if (NULL != data) {
|
|
/* If this was the process that allocated the space in the
|
|
data mpool, then this is the process that frees it */
|
|
|
|
if (NULL != data->mcb_data_mpool_malloc_addr) {
|
|
mca_coll_sm_component.sm_data_mpool->mpool_free(mca_coll_sm_component.sm_data_mpool,
|
|
data->mcb_data_mpool_malloc_addr, NULL);
|
|
}
|
|
|
|
/* Now free the data hanging off the communicator */
|
|
|
|
free(data);
|
|
}
|
|
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
|
|
static bool have_local_peers(ompi_proc_t **procs, size_t size)
|
|
{
|
|
size_t i;
|
|
|
|
for (i = 0; i < size; ++i) {
|
|
if (0 == (procs[i]->proc_flags & OMPI_PROC_FLAG_LOCAL)) {
|
|
return false;
|
|
}
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
|
|
static int bootstrap_init(void)
|
|
{
|
|
int i;
|
|
size_t size;
|
|
char *fullpath;
|
|
mca_common_sm_mmap_t *meta;
|
|
mca_coll_sm_bootstrap_header_extension_t *bshe;
|
|
|
|
/* Create/open the sm coll bootstrap mmap. Make it have enough
|
|
space for the top-level control structure and
|
|
sm_bootstrap_num_segments per-communicator setup struct's
|
|
(i.e., enough for sm_bootstrap_num_segments communicators to
|
|
simultaneously set themselves up) */
|
|
|
|
if (NULL == mca_coll_sm_component.sm_bootstrap_filename) {
|
|
return OMPI_ERROR;
|
|
}
|
|
orte_proc_info();
|
|
asprintf(&fullpath, "%s/%s", orte_process_info.job_session_dir,
|
|
mca_coll_sm_component.sm_bootstrap_filename);
|
|
if (NULL == fullpath) {
|
|
return OMPI_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
size =
|
|
sizeof(mca_coll_sm_bootstrap_header_extension_t) +
|
|
(mca_coll_sm_component.sm_bootstrap_num_segments *
|
|
sizeof(mca_coll_sm_bootstrap_comm_setup_t)) +
|
|
(sizeof(mca_coll_sm_bootstrap_comm_key_t) *
|
|
mca_coll_sm_component.sm_bootstrap_num_segments);
|
|
|
|
mca_coll_sm_component.sm_bootstrap_meta = meta =
|
|
mca_common_sm_mmap_init(size, fullpath,
|
|
sizeof(mca_coll_sm_bootstrap_header_extension_t),
|
|
8);
|
|
if (NULL == meta) {
|
|
return OMPI_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
free(fullpath);
|
|
|
|
/* set the pointer to the bootstrap control structure */
|
|
bshe = (mca_coll_sm_bootstrap_header_extension_t *) meta->map_seg;
|
|
|
|
/* Lock the bootstrap control structure. If it's not already
|
|
initialized, then we're the first one in and we setup the data
|
|
structures */
|
|
|
|
opal_atomic_lock(&bshe->super.seg_lock);
|
|
opal_atomic_wmb();
|
|
if (!bshe->super.seg_inited) {
|
|
bshe->smbhe_segments = (mca_coll_sm_bootstrap_comm_setup_t *)
|
|
(((char *) bshe) +
|
|
sizeof(mca_coll_sm_bootstrap_header_extension_t) +
|
|
(sizeof(uint32_t) *
|
|
mca_coll_sm_component.sm_bootstrap_num_segments));
|
|
bshe->smbhe_keys = (mca_coll_sm_bootstrap_comm_key_t *)
|
|
(((char *) bshe) + sizeof(*bshe));
|
|
for (i = 0; i < mca_coll_sm_component.sm_bootstrap_num_segments; ++i) {
|
|
bshe->smbhe_keys[i].mcsbck_cid = INT_MAX;
|
|
memset(&bshe->smbhe_keys[i].mcsbck_rank0_name, 0,
|
|
sizeof(orte_process_name_t));
|
|
}
|
|
|
|
bshe->super.seg_inited = true;
|
|
}
|
|
opal_atomic_unlock(&bshe->super.seg_lock);
|
|
|
|
/* All done */
|
|
|
|
bootstrap_inited = true;
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
|
|
static int bootstrap_comm(ompi_communicator_t *comm)
|
|
{
|
|
int i, empty_index, err;
|
|
bool found;
|
|
mca_coll_sm_component_t *c = &mca_coll_sm_component;
|
|
mca_coll_sm_bootstrap_header_extension_t *bshe;
|
|
mca_coll_sm_bootstrap_comm_setup_t *bscs;
|
|
mca_coll_base_comm_t *data = comm->c_coll_selected_data;
|
|
int comm_size = ompi_comm_size(comm);
|
|
int num_segments = c->sm_comm_num_segments;
|
|
int num_in_use = c->sm_comm_num_in_use_flags;
|
|
int frag_size = c->sm_fragment_size;
|
|
int control_size = c->sm_control_size;
|
|
orte_process_name_t *rank0;
|
|
|
|
/* Is our CID in the CIDs array? If not, loop until we can find
|
|
an open slot in the array to use in the bootstrap to setup our
|
|
communicator. */
|
|
|
|
bshe = (mca_coll_sm_bootstrap_header_extension_t *)
|
|
c->sm_bootstrap_meta->map_seg;
|
|
bscs = bshe->smbhe_segments;
|
|
opal_atomic_lock(&bshe->super.seg_lock);
|
|
rank0 = &(comm->c_local_group->grp_proc_pointers[0]->proc_name);
|
|
while (1) {
|
|
opal_atomic_wmb();
|
|
found = false;
|
|
empty_index = -1;
|
|
for (i = 0; i < mca_coll_sm_component.sm_bootstrap_num_segments; ++i) {
|
|
if (comm->c_contextid == bshe->smbhe_keys[i].mcsbck_cid &&
|
|
0 == orte_ns.compare(ORTE_NS_CMP_ALL,
|
|
rank0,
|
|
&bshe->smbhe_keys[i].mcsbck_rank0_name)) {
|
|
found = true;
|
|
break;
|
|
} else if (INT_MAX == bshe->smbhe_keys[i].mcsbck_cid &&
|
|
-1 == empty_index) {
|
|
empty_index = i;
|
|
}
|
|
}
|
|
|
|
/* Did we find our CID? */
|
|
|
|
if (found) {
|
|
break;
|
|
}
|
|
|
|
/* Nope. Did we find an empty slot? If so, initialize that
|
|
slot and its corresponding segment for our CID. Get an
|
|
mpool allocation big enough to handle all the shared memory
|
|
collective stuff. */
|
|
|
|
else if (-1 != empty_index) {
|
|
char *tmp;
|
|
size_t size;
|
|
|
|
i = empty_index;
|
|
bshe->smbhe_keys[i].mcsbck_cid = comm->c_contextid;
|
|
/* JMS better assignment? */
|
|
bshe->smbhe_keys[i].mcsbck_rank0_name = *rank0;
|
|
|
|
bscs[i].smbcs_count = comm_size;
|
|
|
|
/* Calculate how much space we need in the data mpool.
|
|
There are several values to add:
|
|
|
|
- size of the barrier data (2 of these):
|
|
- fan-in data (num_procs * control_size)
|
|
- fan-out data (num_procs * control_size)
|
|
- size of the "in use" buffers:
|
|
- num_in_use_buffers * control_size
|
|
- size of the message fragment area (one for each segment):
|
|
- control (num_procs * control_size)
|
|
- fragment data (num_procs * (frag_size))
|
|
|
|
So it's:
|
|
|
|
barrier: 2 * control_size + 2 * control_size
|
|
in use: num_in_use * control_size
|
|
control: num_segments * (num_procs * control_size * 2 +
|
|
num_procs * control_size)
|
|
message: num_segments * (num_procs * frag_size)
|
|
*/
|
|
|
|
size = 4 * control_size +
|
|
(num_in_use * control_size) +
|
|
(num_segments * (comm_size * control_size * 2)) +
|
|
(num_segments * (comm_size * frag_size));
|
|
|
|
data->mcb_data_mpool_malloc_addr = tmp =
|
|
c->sm_data_mpool->mpool_alloc(c->sm_data_mpool, size,
|
|
c->sm_control_size, 0, NULL);
|
|
if (NULL == tmp) {
|
|
/* Cleanup before returning; allow other processes in
|
|
this communicator to learn of the failure. Note
|
|
that by definition, bscs[i].smbcs_count won't be
|
|
zero after the decrement (because there must be >=2
|
|
processes in this communicator, or the self coll
|
|
component would have been chosen), so we don't need
|
|
to do that cleanup. */
|
|
bscs[i].smbcs_data_mpool_offset = 0;
|
|
bscs[i].smbcs_success = false;
|
|
--bscs[i].smbcs_count;
|
|
opal_atomic_unlock(&bshe->super.seg_lock);
|
|
return OMPI_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
bscs[i].smbcs_success = true;
|
|
|
|
/* Calculate the offset and put it in the bootstrap
|
|
area */
|
|
|
|
bscs[i].smbcs_data_mpool_offset = (size_t)
|
|
(tmp -
|
|
((char *) c->sm_data_mpool->mpool_base(c->sm_data_mpool)));
|
|
|
|
break;
|
|
}
|
|
|
|
/* Bad luck all around -- we didn't find our CID in the array
|
|
and there were no empty slots. So give up the lock and let
|
|
some other processes / threads in there to try to free up
|
|
some slots, and then try again once we have reacquired the
|
|
lock. */
|
|
|
|
else {
|
|
opal_atomic_unlock(&bshe->super.seg_lock);
|
|
#ifdef HAVE_SCHED_YIELD
|
|
sched_yield();
|
|
#endif
|
|
opal_atomic_lock(&bshe->super.seg_lock);
|
|
}
|
|
}
|
|
|
|
/* Check to see if there was an error while allocating the shared
|
|
memory */
|
|
if (!bscs[i].smbcs_success) {
|
|
err = OMPI_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
|
|
/* Look at the comm_setup_t section (in the data segment of the
|
|
bootstrap) and fill in the values on our communicator */
|
|
|
|
else {
|
|
err = OMPI_SUCCESS;
|
|
data->mcb_mpool_base = c->sm_data_mpool->mpool_base(c->sm_data_mpool);
|
|
data->mcb_mpool_offset = bscs[i].smbcs_data_mpool_offset;
|
|
data->mcb_mpool_area = data->mcb_mpool_base + data->mcb_mpool_offset;
|
|
data->mcb_operation_count = 0;
|
|
}
|
|
|
|
/* If the count is now zero, then we're finished with this section
|
|
in the bootstrap segment, and we should release it for others
|
|
to use */
|
|
|
|
--bscs[i].smbcs_count;
|
|
if (0 == bscs[i].smbcs_count) {
|
|
bscs[i].smbcs_data_mpool_offset = 0;
|
|
bshe->smbhe_keys[i].mcsbck_cid = INT_MAX;
|
|
memset(&bshe->smbhe_keys[i].mcsbck_rank0_name, 0,
|
|
sizeof(orte_process_name_t));
|
|
}
|
|
|
|
/* All done */
|
|
|
|
opal_atomic_unlock(&bshe->super.seg_lock);
|
|
return err;
|
|
}
|
|
|
|
|
|
/*
|
|
* This function is not static and has a prefix-rule-enabled name
|
|
* because it gets called from the component (but may also be called
|
|
* from above). This is only called once -- no need for reference
|
|
* counting or thread protection.
|
|
*/
|
|
int mca_coll_sm_bootstrap_finalize(void)
|
|
{
|
|
mca_common_sm_mmap_t *meta;
|
|
|
|
if (bootstrap_inited) {
|
|
meta = mca_coll_sm_component.sm_bootstrap_meta;
|
|
|
|
/* Free the area in the mpool that we were using */
|
|
if (mca_coll_sm_component.sm_data_mpool_created) {
|
|
mca_mpool_base_module_destroy(mca_coll_sm_component.sm_data_mpool);
|
|
}
|
|
|
|
/* Free the entire bootstrap area (no need to zero out
|
|
anything in here -- all data structures are referencing
|
|
within the bootstrap area, so the one top-level unmap does
|
|
it all) */
|
|
|
|
munmap(meta->map_seg, meta->map_size);
|
|
}
|
|
|
|
return OMPI_SUCCESS;
|
|
}
|