ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
openmpi/oshmem/mca/spml/ikrit/spml_ikrit.c
Mike Dubman ff384daab4 Added new project: oshmem. 
This commit was SVN r28048.
2013-02-12 15:33:21 +00:00

1344 строки
43 KiB
C
Исходник Ответственный История

/*
* Copyright (c) 2012 Mellanox Technologies, Inc.
* All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#define _GNU_SOURCE
#include <stdio.h>
#include <sys/types.h>
#include <unistd.h>
#include <stdint.h>
#include "oshmem_config.h"
#include "opal/datatype/opal_convertor.h"
#include "orte/include/orte/types.h"
#include "orte/runtime/orte_globals.h"
#include "oshmem/mca/spml/ikrit/spml_ikrit.h"
#include "oshmem/include/shmem.h"
#include "oshmem/mca/memheap/memheap.h"
#include "oshmem/mca/memheap/base/base.h"
#include "oshmem/proc/proc.h"
#include "oshmem/mca/spml/base/base.h"
#include "oshmem/mca/spml/base/spml_base_putreq.h"
#include "oshmem/runtime/runtime.h"
#include "orte/util/show_help.h"
#include "ompi/runtime/ompi_module_exchange.h"
#include "oshmem/mca/spml/ikrit/spml_ikrit_component.h"
#include "ompi/communicator/communicator.h" /*TODO: ompi_communicator_t */
#include "ompi/patterns/comm/coll_ops.h" /*TODO: comm_bcast_pml */
/* use zcopy for put/get via sysv shared memory */
#define SPML_IKRIT_USE_SHM_ZCOPY
//#define SPML_IKRIT_DEBUG_PUT
typedef struct spml_ikrit_am_hdr {
uint64_t va;
} spml_ikrit_am_hdr_t;
struct mca_spml_ikrit_put_request {
mca_spml_base_put_request_t req_put;
mxm_send_req_t mxm_req;
int pe;
mxm_req_buffer_t iov[2];
spml_ikrit_am_hdr_t am_pkt;
};
typedef struct mca_spml_ikrit_put_request mca_spml_ikrit_put_request_t;
OBJ_CLASS_DECLARATION(mca_spml_ikrit_put_request_t);
static int spml_ikrit_get_ep_address(spml_ikrit_mxm_ep_conn_info_t *ep_info, mxm_ptl_id_t ptlid)
{
size_t addrlen;
mxm_error_t err;
addrlen = sizeof(ep_info->ptl_addr[ptlid]);
err = mxm_ep_address(mca_spml_ikrit.mxm_ep, ptlid,
(struct sockaddr *) &ep_info->ptl_addr[ptlid], &addrlen);
if (MXM_OK != err) {
orte_show_help("help-spml-ikrit.txt", "unable to extract endpoint address",
true, mxm_error_string(err));
return OSHMEM_ERROR;
}
return OSHMEM_SUCCESS;
}
static inline void mca_spml_irkit_req_wait(mxm_req_base_t *req)
{
while (!mxm_req_test(req))
opal_progress();
}
static int mca_spml_ikrit_put_request_free(struct oshmem_request_t** request)
{
mca_spml_ikrit_put_request_t *put_req = *(mca_spml_ikrit_put_request_t **)request;
assert(false == put_req->req_put.req_base.req_free_called);
OPAL_THREAD_LOCK(&oshmem_request_lock);
put_req->req_put.req_base.req_free_called = true;
OMPI_FREE_LIST_RETURN( &mca_spml_base_put_requests,
(ompi_free_list_item_t*)put_req);
OPAL_THREAD_UNLOCK(&oshmem_request_lock);
*request = SHMEM_REQUEST_NULL; /*MPI_REQUEST_NULL;*/
return OSHMEM_SUCCESS;
}
static int mca_spml_ikrit_put_request_cancel(struct oshmem_request_t * request, int complete)
{
return OSHMEM_SUCCESS;
}
static void mca_spml_ikrit_put_request_construct(mca_spml_ikrit_put_request_t* req)
{
req->req_put.req_base.req_type = MCA_SPML_REQUEST_PUT;
req->req_put.req_base.req_oshmem.req_free = mca_spml_ikrit_put_request_free;
req->req_put.req_base.req_oshmem.req_cancel = mca_spml_ikrit_put_request_cancel;
}
static void mca_spml_ikrit_put_request_destruct(mca_spml_ikrit_put_request_t* req)
{
}
OBJ_CLASS_INSTANCE( mca_spml_ikrit_put_request_t,
mca_spml_base_put_request_t,
mca_spml_ikrit_put_request_construct,
mca_spml_ikrit_put_request_destruct );
struct mca_spml_ikrit_get_request {
mca_spml_base_get_request_t req_get;
mxm_send_req_t mxm_req;
};
typedef struct mca_spml_ikrit_get_request mca_spml_ikrit_get_request_t;
OBJ_CLASS_DECLARATION(mca_spml_ikrit_get_request_t);
static int mca_spml_ikrit_get_request_free(struct oshmem_request_t** request)
{
mca_spml_ikrit_get_request_t *get_req = *(mca_spml_ikrit_get_request_t **)request;
assert(false == get_req->req_get.req_base.req_free_called);
OPAL_THREAD_LOCK(&oshmem_request_lock);
get_req->req_get.req_base.req_free_called = true;
OMPI_FREE_LIST_RETURN( &mca_spml_base_get_requests,
(ompi_free_list_item_t*)get_req);
OPAL_THREAD_UNLOCK(&oshmem_request_lock);
*request = SHMEM_REQUEST_NULL; /*MPI_REQUEST_NULL;*/
return OSHMEM_SUCCESS;
}
static int mca_spml_ikrit_get_request_cancel(struct oshmem_request_t * request, int complete)
{
return OSHMEM_SUCCESS;
}
static void mca_spml_ikrit_get_request_construct(mca_spml_ikrit_get_request_t* req)
{
req->req_get.req_base.req_type = MCA_SPML_REQUEST_PUT;
req->req_get.req_base.req_oshmem.req_free = mca_spml_ikrit_get_request_free;
req->req_get.req_base.req_oshmem.req_cancel = mca_spml_ikrit_get_request_cancel;
}
static void mca_spml_ikrit_get_request_destruct(mca_spml_ikrit_get_request_t* req)
{
}
OBJ_CLASS_INSTANCE( mca_spml_ikrit_get_request_t,
mca_spml_base_get_request_t,
mca_spml_ikrit_get_request_construct,
mca_spml_ikrit_get_request_destruct );
int mca_spml_ikrit_put_simple(void* dst_addr, size_t size, void* src_addr, int dst);
static void mxm_setup_relays(oshmem_proc_t **procs, size_t nprocs);
mca_spml_ikrit_t mca_spml_ikrit = {
{
/* Init mca_spml_base_module_t */
mca_spml_ikrit_add_procs,
mca_spml_ikrit_del_procs,
mca_spml_ikrit_enable,
mca_spml_ikrit_register,
mca_spml_ikrit_deregister,
mca_spml_ikrit_oob_get_mkeys,
mca_spml_ikrit_put,
mca_spml_ikrit_put_nb,
mca_spml_ikrit_get,
mca_spml_ikrit_recv ,
mca_spml_ikrit_send,
mca_spml_base_wait,
mca_spml_base_wait_nb,
mca_spml_ikrit_fence
}
};
void mca_spml_ikrit_dump_stats(void);
void mca_spml_ikrit_dump_stats()
{
int num_procs;
int i;
char sbuf[1024];
FILE *fp;
fp = fmemopen(sbuf, sizeof(sbuf), "rw");
num_procs = oshmem_num_procs();
for (i = 0; i < num_procs; i++) {
mxm_print_conn_state(mca_spml_ikrit.mxm_peers[i]->mxm_conn, MXM_STATE_DETAIL_LEVEL_DATA, "", fp);
printf("=========== pe:%d conn:%p stats:\n %s==================\n", i, mca_spml_ikrit.mxm_peers[i]->mxm_conn, sbuf);
rewind(fp);
}
fclose(fp);
}
static inline mca_spml_ikrit_put_request_t *alloc_put_req(void)
{
mca_spml_ikrit_put_request_t *req;
ompi_free_list_item_t* item;
int rc;
rc = OSHMEM_SUCCESS;
OMPI_FREE_LIST_WAIT(&mca_spml_base_put_requests, item, rc);
if (OMPI_SUCCESS != rc)
return NULL;
req = (mca_spml_ikrit_put_request_t *)item;
req->req_put.req_base.req_free_called = false;
req->req_put.req_base.req_oshmem.req_complete = false;
return req;
}
static inline mca_spml_ikrit_get_request_t *alloc_get_req(void)
{
mca_spml_ikrit_get_request_t *req;
ompi_free_list_item_t* item;
int rc;
rc = OSHMEM_SUCCESS;
OMPI_FREE_LIST_WAIT(&mca_spml_base_get_requests, item, rc);
if (OMPI_SUCCESS != rc)
return NULL;
req = (mca_spml_ikrit_get_request_t *)item;
req->req_get.req_base.req_free_called = false;
req->req_get.req_base.req_oshmem.req_complete = false;
return req;
}
int mca_spml_ikrit_enable(bool enable)
{
SPML_VERBOSE(50, "*** ikrit ENABLED ****");
ompi_free_list_init_new( &mca_spml_base_put_requests,
sizeof(mca_spml_ikrit_put_request_t),
opal_cache_line_size,
OBJ_CLASS(mca_spml_ikrit_put_request_t),
0,opal_cache_line_size,
mca_spml_ikrit.free_list_num,
mca_spml_ikrit.free_list_max,
mca_spml_ikrit.free_list_inc,
NULL );
ompi_free_list_init_new( &mca_spml_base_get_requests,
sizeof(mca_spml_ikrit_get_request_t),
opal_cache_line_size,
OBJ_CLASS(mca_spml_ikrit_get_request_t),
0,opal_cache_line_size,
mca_spml_ikrit.free_list_num,
mca_spml_ikrit.free_list_max,
mca_spml_ikrit.free_list_inc,
NULL );
return OSHMEM_SUCCESS;
}
static int create_ptl_idx(int dst_pe)
{
oshmem_proc_t *proc;
proc = oshmem_proc_group_find(oshmem_group_all, dst_pe);
proc->transport_ids = (char *)malloc(MXM_PTL_LAST * sizeof(char));
if (!proc->transport_ids)
return OSHMEM_ERROR;
proc->num_transports = 1;
if (oshmem_my_proc_id() == dst_pe)
proc->transport_ids[0] = MXM_PTL_SELF;
else
proc->transport_ids[0] = MXM_PTL_RDMA;
return OSHMEM_SUCCESS;
}
static void destroy_ptl_idx(int dst_pe)
{
oshmem_proc_t *proc;
proc = oshmem_proc_group_find(oshmem_group_all, dst_pe);
if (proc->transport_ids)
free(proc->transport_ids);
}
static void mxm_peer_construct(mxm_peer_t *p)
{
p->pe = -1;
p->n_active_puts = 0;
p->need_fence = 0;
p->pe_relay = -1;
p->n_slaves = 0;
}
static void mxm_peer_destruct(mxm_peer_t *p)
{
/* may be we need to remov item from list */
}
OBJ_CLASS_INSTANCE(
mxm_peer_t,
opal_list_item_t,
mxm_peer_construct,
mxm_peer_destruct
);
int mca_spml_ikrit_del_procs(oshmem_proc_t** procs, size_t nprocs)
{
size_t i;
opal_list_item_t *item;
if (mca_spml_ikrit.mxm_ep) {
mxm_ep_destroy(mca_spml_ikrit.mxm_ep);
mca_spml_ikrit.mxm_ep = 0;
}
while (NULL != (item = opal_list_remove_first(&mca_spml_ikrit.active_peers))) {};
OBJ_DESTRUCT(&mca_spml_ikrit.active_peers);
for (i = 0; i < nprocs; i++) {
destroy_ptl_idx(i);
if (mca_spml_ikrit.mxm_peers[i]) {
//mxm_ep_disconnect(mca_spml_ikrit.mxm_peers[i]->mxm_conn);
OBJ_RELEASE(mca_spml_ikrit.mxm_peers[i]);
}
}
if (mca_spml_ikrit.mxm_peers)
free(mca_spml_ikrit.mxm_peers);
return OSHMEM_SUCCESS;
}
int mca_spml_ikrit_add_procs(oshmem_proc_t** procs, size_t nprocs)
{
spml_ikrit_mxm_ep_conn_info_t* ep_info;
mxm_conn_req_t *conn_reqs;
mxm_error_t err;
size_t i;
int rc = OSHMEM_ERROR;
oshmem_proc_t *proc_self;
int my_rank = oshmem_my_proc_id();
int timeout;
OBJ_CONSTRUCT(&mca_spml_ikrit.active_peers, opal_list_t);
/* Allocate connection requests */
conn_reqs = malloc(nprocs * sizeof(mxm_conn_req_t));
ep_info = malloc(nprocs * sizeof(spml_ikrit_mxm_ep_conn_info_t));
if (NULL == conn_reqs || NULL == ep_info) {
rc = OSHMEM_ERR_OUT_OF_RESOURCE;
goto bail;
}
memset(conn_reqs, 0x0, sizeof(mxm_conn_req_t));
memset(ep_info, 0x0, sizeof(spml_ikrit_mxm_ep_conn_info_t));
mca_spml_ikrit.mxm_peers = (mxm_peer_t **)malloc(nprocs*sizeof(*(mca_spml_ikrit.mxm_peers)));
if (NULL == mca_spml_ikrit.mxm_peers) {
rc = OSHMEM_ERR_OUT_OF_RESOURCE;
goto bail;
}
int* ranks_in_comm;
ranks_in_comm = (int *) malloc(nprocs * sizeof(int));
for (i = 0; i < nprocs; ++i) {
ranks_in_comm[i] = i;
}
if (OSHMEM_SUCCESS != spml_ikrit_get_ep_address(ep_info + my_rank, MXM_PTL_SELF)) {
return OSHMEM_ERROR;
}
if (OSHMEM_SUCCESS != spml_ikrit_get_ep_address(ep_info + my_rank, MXM_PTL_RDMA)) {
return OSHMEM_ERROR;
}
opal_progress_register(spml_ikrit_progress);
for (i = 0; i < nprocs; ++i)
{
comm_bcast_pml(ep_info + i, i, sizeof(spml_ikrit_mxm_ep_conn_info_t),
MPI_BYTE, my_rank, nprocs,
ranks_in_comm, (ompi_communicator_t *)&ompi_mpi_comm_world);
}
if (ranks_in_comm)
free(ranks_in_comm);
/* Get the EP connection requests for all the processes from modex */
for (i = 0; i < nprocs; ++i) {
mca_spml_ikrit.mxm_peers[i] = OBJ_NEW(mxm_peer_t);
if (NULL == mca_spml_ikrit.mxm_peers[i]) {
rc = OSHMEM_ERR_OUT_OF_RESOURCE;
goto bail;
}
mca_spml_ikrit.mxm_peers[i]->pe = i;
conn_reqs[i].ptl_addr[MXM_PTL_SELF] = (struct sockaddr *)&ep_info[i].ptl_addr[MXM_PTL_SELF];
conn_reqs[i].ptl_addr[MXM_PTL_SHM] = NULL;
conn_reqs[i].ptl_addr[MXM_PTL_RDMA] = (struct sockaddr *)&ep_info[i].ptl_addr[MXM_PTL_RDMA];
}
/* Connect to remote peers */
if (mxm_get_version() < MXM_VERSION(1,5)) {
timeout = 1000;
} else {
timeout = -1;
}
err = mxm_ep_connect(mca_spml_ikrit.mxm_ep, conn_reqs, nprocs, timeout);
if (MXM_OK != err) {
SPML_ERROR("MXM returned connect error: %s\n", mxm_error_string(err));
for (i = 0; i < nprocs; ++i) {
if (MXM_OK != conn_reqs[i].error) {
SPML_ERROR("MXM EP connect to %s error: %s\n", procs[i]->proc_hostname,
mxm_error_string(conn_reqs[i].error));
}
}
rc = OSHMEM_ERR_CONNECTION_FAILED;
goto bail;
}
/* Save returned connections */
for (i = 0; i < nprocs; ++i) {
mca_spml_ikrit.mxm_peers[i]->mxm_conn = conn_reqs[i].conn;
if (OSHMEM_SUCCESS != create_ptl_idx(i))
goto bail;
//printf("proc=%d name=%s jobid = %u vpid = %u flags = %x\n", (int)i, procs[i]->proc_hostname, procs[i]->proc_name.jobid, procs[i]->proc_name.vpid, procs[i]->proc_flags);
mxm_conn_ctx_set(conn_reqs[i].conn, mca_spml_ikrit.mxm_peers[i]);
}
if (ep_info)
free(ep_info);
if (conn_reqs)
free(conn_reqs);
proc_self = oshmem_proc_group_find(oshmem_group_all, my_rank);
/* identify local processes and change transport to SHM */
for (i = 0; i < nprocs; i++) {
if (procs[i]->proc_name.jobid != proc_self->proc_name.jobid ||
!OPAL_PROC_ON_LOCAL_NODE(procs[i]->proc_flags)) {
continue;
}
if (procs[i] == proc_self)
continue;
#ifdef SPML_IKRIT_USE_SHM_ZCOPY
procs[i]->transport_ids[0] = MXM_PTL_SHM;
procs[i]->transport_ids[1] = MXM_PTL_RDMA;
procs[i]->num_transports = 2;
#endif
}
mxm_setup_relays(procs, nprocs);
SPML_VERBOSE(50, "*** ADDED PROCS ***");
return OSHMEM_SUCCESS;
bail:
if (ep_info)
free(ep_info);
if (conn_reqs)
free(conn_reqs);
SPML_ERROR("add procs FAILED rc=%d", rc);
return rc;
}
mca_spml_mkey_t *mca_spml_ikrit_register(void* addr, size_t size, uint64_t shmid, int *count)
{
//mxm_error_t err;
int i;
mca_spml_mkey_t *mkeys;
*count = 0;
mkeys = (mca_spml_mkey_t *)calloc(1, MXM_PTL_LAST * sizeof(*mkeys));
if(!mkeys){
return NULL;
}
for (i = 0; i < MXM_PTL_LAST; i++) {
switch(i) {
case MXM_PTL_SHM:
if ((int)MEMHEAP_SHM_GET_ID(shmid) != MEMHEAP_SHM_INVALID) {
mkeys[i].key = shmid;
mkeys[i].va_base = 0;
}
else {
mkeys[i].key = 0;
mkeys[i].va_base = (unsigned long)addr;
}
mkeys[i].spml_context = 0;
break;
case MXM_PTL_SELF:
mkeys[i].key = 0;
mkeys[i].spml_context = 0;
mkeys[i].va_base = (unsigned long)addr;
break;
case MXM_PTL_RDMA:
#if MXM_API < MXM_VERSION(1,5)
mkeys[i].ib.lkey = mkeys[i].ib.rkey = MXM_MKEY_NONE;
#else
mkeys[i].ib.lkey = mkeys[i].ib.rkey = 0;
#endif
mkeys[i].spml_context = 0;
#if 0
/* don't register memheap if zcopy support is not enabled */
err = mxm_reg_mr(mca_spml_ikrit.mxm_ep, MXM_PTL_RDMA, addr, size, &mkeys[i].ib.lkey, &mkeys[i].ib.rkey);
if (MXM_OK != err) {
SPML_VERBOSE(1, "failed to register memory: %s", mxm_error_string(err));
goto err;
}
#endif
mkeys[i].va_base = (unsigned long)addr;
//mkeys[i].spml_context = (void *)(unsigned long)size;
break;
default:
SPML_ERROR("unsupported PTL: %d", i);
goto err;
}
SPML_VERBOSE(5,"rank %d ptl %d rkey %x lkey %x key %llx address 0x%llX len %llu shmid 0x%X|0x%X",
oshmem_proc_local_proc->proc_name.vpid,
i,
mkeys[i].ib.rkey,
mkeys[i].ib.lkey,
(unsigned long long)mkeys[i].key,
(unsigned long long)mkeys[i].va_base,
(unsigned long long)size,
MEMHEAP_SHM_GET_TYPE(shmid), MEMHEAP_SHM_GET_ID(shmid)
);
}
*count = MXM_PTL_LAST;
return mkeys;
err:
mca_spml_ikrit_deregister(mkeys);
return NULL;
}
int mca_spml_ikrit_deregister(mca_spml_mkey_t *mkeys)
{
int i;
if (!mkeys)
return OSHMEM_SUCCESS;
for (i = 0; i < MXM_PTL_LAST; i++) {
switch(i) {
case MXM_PTL_SELF:
case MXM_PTL_SHM:
break;
case MXM_PTL_RDMA:
/* dereg memory */
if (!mkeys[i].spml_context)
break;
#if MXM_API < MXM_VERSION(1,5)
mxm_dereg_mr(mca_spml_ikrit.mxm_ep, MXM_PTL_RDMA,
(void *)mkeys[i].va_base,
(unsigned long)mkeys[i].spml_context
);
#endif
break;
}
}
return OSHMEM_SUCCESS;
}
static inline int get_ptl_id(int dst)
{
oshmem_proc_t *proc;
/* get endpoint and btl */
proc = oshmem_proc_group_all(dst);
if (!proc) {
SPML_ERROR("Can not find destination proc for pe=%d", dst);
oshmem_shmem_abort(-1);
return -1;
}
return proc->transport_ids[0];
}
int mca_spml_ikrit_oob_get_mkeys(int pe, uint32_t seg, mca_spml_mkey_t *mkeys)
{
int ptl;
ptl = get_ptl_id(pe);
if (ptl < 0)
return OSHMEM_ERROR;
if (ptl != MXM_PTL_RDMA)
return OSHMEM_ERROR;
if (seg > 1)
return OSHMEM_ERROR;
#if MXM_API < MXM_VERSION(1,5)
mkeys[ptl].ib.rkey = MXM_MKEY_NONE;
#endif
return OSHMEM_SUCCESS;
}
static int mca_spml_ikrit_get_helper(mxm_send_req_t *sreq, void *src_addr, size_t size, void *dst_addr, int src)
{
/* shmem spec states that get() operations are blocking. So it is enough
to have single mxm request. Also we count on mxm doing copy */
uint64_t rva;
mca_spml_mkey_t *r_mkey;
int ptl_id;
ptl_id = get_ptl_id(src);
/* already tried to send via shm and failed. go via rdma */
if (ptl_id == MXM_PTL_SHM)
ptl_id = MXM_PTL_RDMA;
/**
* Get the address to the remote rkey.
**/
r_mkey = mca_memheap.memheap_get_cached_mkey(src, (unsigned long)src_addr, ptl_id, &rva);
if (!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", src, src_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
SPML_VERBOSE(100, "get: pe:%d ptl=%d src=%p -> dst: %p sz=%d. src_rva=%p, src_rkey=0x%lx",
src, ptl_id, src_addr, dst_addr, (int)size, (void *)rva, r_mkey->key);
/* mxm does not really cares for get lkey */
sreq->base.mq = mca_spml_ikrit.mxm_mq;
sreq->base.conn = mca_spml_ikrit.mxm_peers[src]->mxm_conn;
sreq->base.data_type = MXM_REQ_DATA_BUFFER;
sreq->base.data.buffer.ptr = dst_addr;
sreq->base.data.buffer.length = size;
#if MXM_API < MXM_VERSION(1,5)
sreq->base.data.buffer.mkey = MXM_MKEY_NONE;
sreq->op.mem.remote_mkey = r_mkey->ib.rkey;
#else
sreq->base.data.buffer.memh = NULL;
sreq->op.mem.remote_memh = NULL;
#endif
sreq->opcode = MXM_REQ_OP_GET;
sreq->op.mem.remote_vaddr = (intptr_t)rva;
sreq->base.state = MXM_REQ_NEW;
return OSHMEM_SUCCESS;
}
static inline int mca_spml_ikrit_get_shm(void *src_addr, size_t size, void *dst_addr, int src)
{
int ptl_id;
uint64_t rva;
mca_spml_mkey_t *r_mkey;
ptl_id = get_ptl_id(src);
/**
* Get the address to the remote rkey.
**/
if (ptl_id != MXM_PTL_SHM)
return OSHMEM_ERROR;
r_mkey = mca_memheap.memheap_get_cached_mkey(src, (unsigned long)src_addr, ptl_id, &rva);
if (!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", src, src_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
if (OPAL_UNLIKELY(!mca_memheap.memheap_is_symmetric_addr((unsigned long)src_addr) || (unsigned long)src_addr == rva))
return OSHMEM_ERROR;
SPML_VERBOSE(100, "shm get: pe:%d src=%p -> dst: %p sz=%d. src_rva=%p, src_rkey=0x%lx",
src, src_addr, dst_addr, (int)size, (void *)rva, r_mkey->key);
memcpy(dst_addr, (void *)(unsigned long)rva, size);
opal_progress();
return OSHMEM_SUCCESS;
}
int mca_spml_ikrit_get(void *src_addr, size_t size, void *dst_addr, int src)
{
mxm_send_req_t sreq;
if (OSHMEM_SUCCESS == mca_spml_ikrit_get_shm(src_addr, size, dst_addr, src))
return OSHMEM_SUCCESS;
if (OSHMEM_SUCCESS != mca_spml_ikrit_get_helper(&sreq, src_addr, size, dst_addr, src)) {
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
sreq.base.flags = MXM_REQ_FLAG_BLOCKING;
sreq.base.completed_cb = NULL;
mxm_req_send(&sreq);
opal_progress();
mca_spml_irkit_req_wait(&sreq.base);
if (MXM_OK != sreq.base.error) {
SPML_ERROR("get request failed: %s - aborting", mxm_error_string(sreq.base.error));
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
return OSHMEM_SUCCESS;
}
static inline void get_completion_cb(void *ctx)
{
mca_spml_ikrit_get_request_t *get_req = (mca_spml_ikrit_get_request_t *)ctx;
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_active_gets, -1);
get_req->req_get.req_base.req_spml_complete = true;
get_req->req_get.req_base.req_oshmem.req_status.SHMEM_ERROR = OSHMEM_SUCCESS;
oshmem_request_complete( &get_req->req_get.req_base.req_oshmem, 1);
oshmem_request_free( (oshmem_request_t**)&get_req );
}
/* extension. used 4 fence implementation b4 fence was added to mxm */
int mca_spml_ikrit_get_async(void *src_addr, size_t size, void *dst_addr, int src)
{
mca_spml_ikrit_get_request_t *get_req;
if (OSHMEM_SUCCESS == mca_spml_ikrit_get_shm(src_addr, size, dst_addr, src))
return OSHMEM_SUCCESS;
get_req = alloc_get_req();
if (NULL == get_req) {
SPML_ERROR("out of get requests - aborting");
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
if (OSHMEM_SUCCESS != mca_spml_ikrit_get_helper(&get_req->mxm_req, src_addr, size, dst_addr, src)) {
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
get_req->mxm_req.base.flags = 0;
get_req->mxm_req.base.completed_cb = get_completion_cb;
get_req->mxm_req.base.context = get_req;
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_active_gets, 1);
mxm_req_send(&get_req->mxm_req);
if (MXM_OK != get_req->mxm_req.base.error) {
SPML_ERROR("get request failed: %s - aborting", mxm_error_string(get_req->mxm_req.base.error));
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
return OSHMEM_SUCCESS;
}
static inline void fence_completion_cb(void *ctx)
{
mca_spml_ikrit_get_request_t *fence_req = (mca_spml_ikrit_get_request_t *)ctx;
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_mxm_fences, -1);
fence_req->req_get.req_base.req_spml_complete = true;
fence_req->req_get.req_base.req_oshmem.req_status.SHMEM_ERROR = OSHMEM_SUCCESS;
oshmem_request_complete( &fence_req->req_get.req_base.req_oshmem, 1);
oshmem_request_free( (oshmem_request_t**)&fence_req );
}
static int mca_spml_ikrit_mxm_fence(int dst)
{
mca_spml_ikrit_get_request_t *fence_req;
fence_req = alloc_get_req();
if (NULL == fence_req) {
SPML_ERROR("out of get requests - aborting");
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
fence_req->mxm_req.base.mq = mca_spml_ikrit.mxm_mq;
fence_req->mxm_req.base.conn = mca_spml_ikrit.mxm_peers[dst]->mxm_conn;
fence_req->mxm_req.opcode = MXM_REQ_OP_FENCE;
fence_req->mxm_req.base.state = MXM_REQ_NEW;
fence_req->mxm_req.base.flags = MXM_REQ_FLAG_SEND_SYNC;
fence_req->mxm_req.base.completed_cb = fence_completion_cb;
fence_req->mxm_req.base.context = fence_req;
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_mxm_fences, 1);
mxm_req_send(&fence_req->mxm_req);
return OSHMEM_SUCCESS;
}
static inline void put_completion_cb(void *ctx)
{
mca_spml_ikrit_put_request_t *put_req = (mca_spml_ikrit_put_request_t *)ctx;
mxm_peer_t *peer;
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_active_puts, -1);
peer = mca_spml_ikrit.mxm_peers[put_req->pe];
/* this was last put in progress. Remove peer from the list so that we do not need explicit fence */
#ifdef SPML_IKRIT_DEBUG_PUT
if (peer) {
if (peer->n_active_puts <= 0) {
/* actually this can happen because fence forces ref count to 0 while puts still may be in flight */
SPML_VERBOSE(1, "pe %d n_active_puts %d", put_req->pe, peer->n_active_puts);
}
}
if (put_req->mxm_req.base.state != MXM_REQ_COMPLETED)
SPML_ERROR("oops: pe %d uncompleted request state %d", put_req->pe, put_req->mxm_req.base.state);
#endif
if (0 < peer->n_active_puts) {
peer->n_active_puts--;
if (0 == peer->n_active_puts && (put_req->mxm_req.base.flags & MXM_REQ_FLAG_SEND_SYNC)) {
//SPML_VERBOSE(20, "removed pe %d from active list", put_req->pe);
opal_list_remove_item(&mca_spml_ikrit.active_peers, &peer->super);
peer->need_fence = 0;
}
}
put_req->req_put.req_base.req_spml_complete = true;
put_req->req_put.req_base.req_oshmem.req_status.SHMEM_ERROR = OSHMEM_SUCCESS;
oshmem_request_complete( &put_req->req_put.req_base.req_oshmem, 1);
oshmem_request_free( (oshmem_request_t**)&put_req );
}
/**
* TODO: using put request as handle is not good.
*/
static inline int mca_spml_ikrit_put_internal(void* dst_addr, size_t size, void* src_addr, int dst, void **handle, int zcopy)
{
uint64_t rva;
mca_spml_ikrit_put_request_t *put_req;
int ptl_id;
mca_spml_mkey_t *l_mkey, *r_mkey;
uint32_t lkey;
static int count;
int need_progress = 0;
ptl_id = get_ptl_id(dst);
/* Get rkey of remote PE (dst proc) which must be on memheap */
r_mkey = mca_memheap.memheap_get_cached_mkey(dst, (unsigned long)dst_addr, ptl_id, &rva);
if(!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", dst, dst_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
#ifdef SPML_IKRIT_DEBUG_PUT
SPML_VERBOSE(100, "put: pe:%d ptl=%d dst=%p <- src: %p sz=%d. dst_rva=%p, dst_rkey=0x%lx",
dst, ptl_id, dst_addr, src_addr, (int)size, (void *)rva, r_mkey->key);
#endif
if (ptl_id == MXM_PTL_SHM) {
if (OPAL_LIKELY(mca_memheap.memheap_is_symmetric_addr((unsigned long)dst_addr) && (unsigned long)dst_addr != rva)) {
memcpy((void *)(unsigned long)rva, src_addr, size);
// call progress as often as we would have with regular put
if (++count % SPML_IKRIT_PACKETS_PER_SYNC == 0)
mxm_progress(mca_spml_ikrit.mxm_context);
return OSHMEM_SUCCESS;
}
// segment not mapped - fallback to rmda
ptl_id = MXM_PTL_RDMA;
r_mkey = mca_memheap.memheap_get_cached_mkey(dst, (unsigned long)dst_addr, ptl_id, &rva);
if(!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", dst, dst_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
}
#ifdef SPML_IKRIT_DEBUG_PUT
SPML_VERBOSE(100, "put: pe:%d ptl=%d dst=%p <- src: %p sz=%d. dst_rva=%p, dst_rkey=0x%lx",
dst, ptl_id, dst_addr, src_addr, (int)size, (void *)rva, r_mkey->key);
#endif
l_mkey = mca_memheap.memheap_get_local_mkey((unsigned long)src_addr, ptl_id);
if (zcopy == 0 || !l_mkey) {
/* local memory is not registered - pass proper flag to mxm */
#if MXM_API < MXM_VERSION(1,5)
lkey = MXM_MKEY_NONE;
#else
lkey = 0;
#endif
}
else {
lkey = l_mkey->ib.lkey;
}
put_req = alloc_put_req();
if (NULL == put_req) {
SPML_ERROR("out of put requests - aborting");
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
if (handle)
*handle = put_req;
/* fill out request */
put_req->mxm_req.base.mq = mca_spml_ikrit.mxm_mq;
/* request immediate responce if we are getting low on send buffers. We only get responce from remote on ack timeout.
* Also request explicit ack once in a while */
if (mca_spml_ikrit.free_list_max - mca_spml_ikrit.n_active_puts <= SPML_IKRIT_PUT_LOW_WATER ||
(mca_spml_ikrit.mxm_peers[dst]->n_active_puts + 1) % SPML_IKRIT_PACKETS_PER_SYNC == 0) {
put_req->mxm_req.base.flags = MXM_REQ_FLAG_SEND_SYNC;
need_progress = 1;
} else {
put_req->mxm_req.base.flags = MXM_REQ_FLAG_SEND_LAZY|MXM_REQ_FLAG_SEND_SYNC;
}
put_req->mxm_req.base.conn = mca_spml_ikrit.mxm_peers[dst]->mxm_conn;
put_req->mxm_req.base.data_type = MXM_REQ_DATA_BUFFER;
put_req->mxm_req.base.data.buffer.ptr = src_addr;
put_req->mxm_req.base.data.buffer.length = size;
put_req->mxm_req.base.completed_cb = put_completion_cb;
put_req->mxm_req.base.context = put_req;
put_req->mxm_req.opcode = MXM_REQ_OP_PUT;
put_req->mxm_req.op.mem.remote_vaddr = (intptr_t)rva;
put_req->mxm_req.base.state = MXM_REQ_NEW;
put_req->pe = dst;
#if MXM_API < MXM_VERSION(1,5)
put_req->mxm_req.base.data.buffer.mkey = lkey;
put_req->mxm_req.op.mem.remote_mkey = r_mkey->ib.rkey;
#else
put_req->mxm_req.base.data.buffer.memh = NULL;
put_req->mxm_req.op.mem.remote_memh = NULL;
#endif
if (mca_spml_ikrit.mxm_peers[dst]->pe_relay >= 0 &&
mca_memheap_base_detect_addr_type((unsigned long)dst_addr) == ADDR_USER ) {
put_req->mxm_req.op.am.hid = 0;
put_req->mxm_req.op.am.imm_data = dst;
put_req->pe = mca_spml_ikrit.mxm_peers[dst]->pe_relay;
put_req->mxm_req.base.conn = mca_spml_ikrit.mxm_peers[put_req->pe]->mxm_conn;
put_req->mxm_req.opcode = MXM_REQ_OP_AM;
/* set up iov */
put_req->mxm_req.base.data_type = MXM_REQ_DATA_IOV;
put_req->mxm_req.base.data.iov.count = 2;
put_req->mxm_req.base.data.iov.vector = put_req->iov;
put_req->iov[0].ptr = &put_req->am_pkt.va;
put_req->iov[0].length = sizeof(uint64_t);
put_req->am_pkt.va = (uint64_t)rva;
put_req->iov[1].ptr = src_addr;
put_req->iov[1].length = size;
#if MXM_API < MXM_VERSION(1,5)
put_req->iov[0].mkey = MXM_MKEY_NONE;
put_req->iov[1].mkey = lkey;
#else
put_req->iov[0].memh = NULL;
put_req->iov[1].memh = NULL;
#endif
}
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_active_puts, 1);
if (mca_spml_ikrit.mxm_peers[dst]->need_fence == 0) {
//SPML_VERBOSE(20, "added pe %d to active list", dst);
opal_list_append(&mca_spml_ikrit.active_peers, &mca_spml_ikrit.mxm_peers[dst]->super);
mca_spml_ikrit.mxm_peers[dst]->need_fence = 1;
}
mca_spml_ikrit.mxm_peers[dst]->n_active_puts++;
mxm_req_send(&put_req->mxm_req);
if (MXM_OK != put_req->mxm_req.base.error) {
OPAL_THREAD_ADD32(&mca_spml_ikrit.n_active_puts, -1);
SPML_ERROR("put request %p failed: %s - aborting", put_req, mxm_error_string(put_req->mxm_req.base.error));
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
//put_completion_cb(put_req);
if (need_progress)
mxm_progress(mca_spml_ikrit.mxm_context);
return OSHMEM_SUCCESS;
}
/* simple buffered put implementation. NOT IN USE
* Problems:
* - slighly worse performance than impl based on non buffered put
* - fence complexity is O(n_active_connections) instead of O(n_connections_with_outstanding_puts).
* Later is bounded by the network RTT & mxm ack timer.
*/
int mca_spml_ikrit_put_simple(void* dst_addr, size_t size, void* src_addr, int dst)
{
uint64_t rva;
mxm_send_req_t mxm_req;
mxm_wait_t wait;
int ptl_id;
mca_spml_mkey_t *r_mkey;
static int count;
ptl_id = get_ptl_id(dst);
/* Get rkey of remote PE (dst proc) which must be on memheap */
r_mkey = mca_memheap.memheap_get_cached_mkey(dst, (unsigned long)dst_addr, ptl_id, &rva);
if(!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", dst, dst_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
#ifdef SPML_IKRIT_DEBUG_PUT
SPML_VERBOSE(100, "put: pe:%d ptl=%d dst=%p <- src: %p sz=%d. dst_rva=%p, dst_rkey=0x%lx",
dst, ptl_id, dst_addr, src_addr, (int)size, (void *)rva, r_mkey->key);
#endif
if (ptl_id == MXM_PTL_SHM) {
if (OPAL_LIKELY(mca_memheap.memheap_is_symmetric_addr((unsigned long)dst_addr) && (unsigned long)dst_addr != rva)) {
memcpy((void *)(unsigned long)rva, src_addr, size);
// call progress as often as we would have with regular put
if (++count % SPML_IKRIT_PACKETS_PER_SYNC == 0)
mxm_progress(mca_spml_ikrit.mxm_context);
return OSHMEM_SUCCESS;
}
// segment not mapped - fallback to rmda
ptl_id = MXM_PTL_RDMA;
r_mkey = mca_memheap.memheap_get_cached_mkey(dst, (unsigned long)dst_addr, ptl_id, &rva);
if(!r_mkey) {
SPML_ERROR("pe=%d: %p is not address of shared variable", dst, dst_addr);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
}
#ifdef SPML_IKRIT_DEBUG_PUT
SPML_VERBOSE(100, "put: pe:%d ptl=%d dst=%p <- src: %p sz=%d. dst_rva=%p, dst_rkey=0x%lx",
dst, ptl_id, dst_addr, src_addr, (int)size, (void *)rva, r_mkey->key);
#endif
/* fill out request */
mxm_req.base.mq = mca_spml_ikrit.mxm_mq;
mxm_req.base.flags = MXM_REQ_FLAG_BLOCKING;
mxm_req.base.conn = mca_spml_ikrit.mxm_peers[dst]->mxm_conn;
mxm_req.base.data_type = MXM_REQ_DATA_BUFFER;
mxm_req.base.data.buffer.ptr = src_addr;
mxm_req.base.data.buffer.length = size;
mxm_req.base.completed_cb = 0;
mxm_req.base.context = 0;
mxm_req.opcode = MXM_REQ_OP_PUT;
mxm_req.op.mem.remote_vaddr = (intptr_t)rva;
mxm_req.base.state = MXM_REQ_NEW;
mxm_req.base.error = MXM_OK;
#if MXM_API < MXM_VERSION(1,5)
mxm_req.base.data.buffer.mkey = MXM_MKEY_NONE;
mxm_req.op.mem.remote_mkey = MXM_MKEY_NONE;
#else
mxm_req.base.data.buffer.memh = NULL;
mxm_req.op.mem.remote_memh = NULL;
#endif
if (mca_spml_ikrit.mxm_peers[dst]->need_fence == 0) {
//SPML_VERBOSE(20, "added pe %d to active list", dst);
opal_list_append(&mca_spml_ikrit.active_peers, &mca_spml_ikrit.mxm_peers[dst]->super);
mca_spml_ikrit.mxm_peers[dst]->need_fence = 1;
}
mxm_req_send(&mxm_req);
if (MXM_OK != mxm_req.base.error) {
SPML_ERROR("put request failed: %s(%d) - aborting", mxm_error_string(mxm_req.base.error), mxm_req.base.error);
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
wait.req = &mxm_req.base;
wait.state = (mxm_req_state_t)(MXM_REQ_SENT|MXM_REQ_COMPLETED);
wait.progress_cb = NULL;
wait.progress_arg = NULL;
mxm_wait(&wait);
return OSHMEM_SUCCESS;
}
int mca_spml_ikrit_put_nb(void* dst_addr, size_t size, void* src_addr, int dst, void **handle)
{
int err;
err = mca_spml_ikrit_put_internal(dst_addr, size, src_addr, dst, handle, 1);
if (OSHMEM_SUCCESS != err) {
SPML_ERROR("put failed - aborting");
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
return OSHMEM_SUCCESS;
}
int mca_spml_ikrit_put(void* dst_addr, size_t size, void* src_addr, int dst)
{
int err;
mca_spml_ikrit_put_request_t *put_req;
mxm_wait_t wait;
put_req = 0;
err = mca_spml_ikrit_put_internal(dst_addr, size, src_addr, dst, (void **)&put_req, 0);
if (OSHMEM_SUCCESS != err) {
SPML_ERROR("put failed - aborting");
oshmem_shmem_abort(-1);
return OSHMEM_ERROR;
}
if (!put_req)
return OSHMEM_SUCCESS;
wait.req = &put_req->mxm_req.base;
wait.state = (mxm_req_state_t)(MXM_REQ_SENT|MXM_REQ_COMPLETED);
wait.progress_cb = NULL;
wait.progress_arg = NULL;
mxm_wait(&wait);
return OSHMEM_SUCCESS;
}
static void mxm_relay_handler(mxm_conn_h conn, mxm_imm_t imm, void *data, size_t len, size_t offset, int is_lf)
{
uint64_t va, rva;
char *pkt_data;
mca_spml_mkey_t *r_mkey;
int ptl_id;
mxm_peer_t *peer;
//printf("relay req recvd conn=%p data=%p dst=%d len=%d offset=%d is_lf=%d - aborting\n",
// conn, data, (int)imm, (int)len, (int)offset, is_lf);
if (offset == 0) {
va = *(uint64_t *)data;
pkt_data = (char *)data + sizeof(va);
len -= sizeof(va);
if (!is_lf) {
// we expect more fragments: save destination virtual address
peer = mxm_conn_ctx_get(conn);
peer->dst_va = va;
}
}
else {
// next fragment: use saved va and offset to compute va
pkt_data = data;
peer = mxm_conn_ctx_get(conn);
va = peer->dst_va + offset - sizeof(va);
}
ptl_id = get_ptl_id(imm);
if (ptl_id != MXM_PTL_SHM) {
SPML_ERROR("relay req to non local PE recvd dst=%d va=%llx len=%d - aborting", (int)imm, (unsigned long long)va, (int)len);
oshmem_shmem_abort(-1);
return;
}
/* Get rkey of remote PE (dst proc) which must be on memheap */
r_mkey = mca_memheap.memheap_get_cached_mkey(imm, va, ptl_id, &rva);
if(!r_mkey) {
SPML_ERROR("relay to PE=%d: %p is not address of shared variable", imm, (void *)va);
oshmem_shmem_abort(-1);
return;
}
memcpy((void *)(unsigned long)rva, pkt_data, len);
}
static void mxm_setup_relays(oshmem_proc_t **procs, size_t nprocs)
{
size_t i;
//long i;
opal_hash_table_t h;
int pe_relay;
int ret;
int r_i, r;
if (mca_spml_ikrit.n_relays <= 0)
return;
OBJ_CONSTRUCT(&h, opal_hash_table_t);
opal_hash_table_init(&h, 128);
/* lowest rank on node will be used to relay to everyone on that node */
for (i = 0; i < nprocs; i++) {
//for (i = nprocs-1; i >= 0; i--) {
if (OPAL_PROC_ON_LOCAL_NODE(procs[i]->proc_flags))
continue;
ret = opal_hash_table_get_value_ptr(&h, procs[i]->proc_hostname, strlen(procs[i]->proc_hostname), (void **)&pe_relay);
if (ret != OPAL_SUCCESS) {
opal_hash_table_set_value_ptr(&h, procs[i]->proc_hostname, strlen(procs[i]->proc_hostname), (void *)i);
mca_spml_ikrit.mxm_peers[i]->n_relays = 1;
mca_spml_ikrit.mxm_peers[i]->pe_relays[0] = i;
continue;
}
#if 0
if (mca_spml_ikrit.mxm_peers[pe_relay]->n_slaves >= 15) {
opal_hash_table_set_value_ptr(&h, procs[i]->proc_hostname, strlen(procs[i]->proc_hostname), (void *)i);
}
else {
mca_spml_ikrit.mxm_peers[pe_relay]->n_slaves++;
mca_spml_ikrit.mxm_peers[i]->pe_relay = pe_relay;
// printf("dst %d relay %d\n", (int)i, pe_relay);
}
#endif
/* first allocate relays */
if (mca_spml_ikrit.mxm_peers[pe_relay]->n_relays < mca_spml_ikrit.n_relays) {
//printf("r_i=%d assigned relay %d\n", mca_spml_ikrit.mxm_peers[pe_relay]->n_relays, (int)i);
mca_spml_ikrit.mxm_peers[pe_relay]->pe_relays[mca_spml_ikrit.mxm_peers[pe_relay]->n_relays] = i;
mca_spml_ikrit.mxm_peers[pe_relay]->n_relays++;
continue;
}
/* now assign slave to relay */
r_i = mca_spml_ikrit.mxm_peers[pe_relay]->n_relays - 1;
while (r_i >= 0) {
r = mca_spml_ikrit.mxm_peers[pe_relay]->pe_relays[r_i];
if (mca_spml_ikrit.mxm_peers[r]->n_slaves >= 1) {
r_i--;
continue;
}
mca_spml_ikrit.mxm_peers[r]->n_slaves++;
mca_spml_ikrit.mxm_peers[i]->pe_relay = r;
//printf("dst %d relay %d r_idx %d master %d\n", (int)i, r, r_i, pe_relay);
break;
}
//mca_spml_ikrit.mxm_peers[pe_relay]->n_relays = r_i;
}
OBJ_DESTRUCT(&h);
mxm_set_am_handler(mca_spml_ikrit.mxm_context, 0, mxm_relay_handler, MXM_AM_FLAG_THREAD_SAFE);
}
int mca_spml_ikrit_fence(void)
{
mxm_peer_t *peer;
opal_list_item_t *item;
SPML_VERBOSE(20, "Into fence with %d active puts on %d pes",
mca_spml_ikrit.n_active_puts,
(int)opal_list_get_size(&mca_spml_ikrit.active_peers)
);
/* puts(unless are send sync) are completed by remote side lazily. That is either when remote decides to
* ack window which can take hundreds of ms. So speed things up by doing fence */
while (NULL != (item = opal_list_remove_first(&mca_spml_ikrit.active_peers))) {
peer = (mxm_peer_t *)item;
peer->n_active_puts = 0;
peer->need_fence = 0;
mca_spml_ikrit_mxm_fence(peer->pe);
}
while (0 < mca_spml_ikrit.n_mxm_fences) {
oshmem_request_wait_any_completion();
}
SPML_VERBOSE(20, "fence completed");
return OSHMEM_SUCCESS;
}
/* blocking receive */
int mca_spml_ikrit_recv(void* buf, size_t size, int src)
{
mxm_recv_req_t req;
char dummy_buf[1];
/* tag mask 0 matches any tag */
SPML_VERBOSE(100, "want to recv from src %d, size %d buf %p", src, (int)size, buf);
req.tag = src == SHMEM_ANY_SOURCE ? 0 : src;
req.tag_mask = src == SHMEM_ANY_SOURCE ? 0 : 0xFFFFFFFF;
req.base.state = MXM_REQ_NEW;
req.base.mq = mca_spml_ikrit.mxm_mq;
req.base.conn = NULL;
req.base.flags = MXM_REQ_FLAG_BLOCKING;
req.base.completed_cb = NULL;
req.base.data_type = MXM_REQ_DATA_BUFFER;
req.base.data.buffer.ptr = buf == NULL ? dummy_buf : buf;
req.base.data.buffer.length = size == 0 ? sizeof(dummy_buf) : size;
#if MXM_API < MXM_VERSION(1,5)
req.base.data.buffer.mkey = MXM_MKEY_NONE;
#else
req.base.data.buffer.memh = NULL;
#endif
mxm_req_recv(&req);
mca_spml_irkit_req_wait(&req.base);
if (req.base.error != MXM_OK) {
return OSHMEM_ERROR;
}
SPML_VERBOSE(100, "recvd from tag %d len %d", req.completion.sender_tag,
(int)req.completion.actual_len);
return OSHMEM_SUCCESS;
}
/* for now only do blocking copy send */
int mca_spml_ikrit_send(void* buf, size_t size, int dst, mca_spml_base_put_mode_t mode)
{
mxm_send_req_t req;
char dummy_buf[1];
SPML_VERBOSE(100, "sending %p size %d to %d, mode %d", buf, (int)size, dst, (int)mode);
req.opcode = MXM_REQ_OP_SEND;
// FIXME: doing tagging like this can cause a conflict with MPI mtl
req.op.send.tag = oshmem_my_proc_id();
req.base.state = MXM_REQ_NEW;
req.base.mq = mca_spml_ikrit.mxm_mq;
req.base.conn = mca_spml_ikrit.mxm_peers[dst]->mxm_conn;
req.base.flags = MXM_REQ_FLAG_BLOCKING;
req.base.completed_cb = NULL;
req.base.data_type = MXM_REQ_DATA_BUFFER;
req.base.data.buffer.ptr = buf == NULL ? dummy_buf : buf;
req.base.data.buffer.length = size == 0 ? sizeof(dummy_buf) : size;
#if MXM_API < MXM_VERSION(1,5)
req.base.data.buffer.mkey = MXM_MKEY_NONE;
#else
req.base.data.buffer.memh = NULL;
#endif
mxm_req_send(&req);
mca_spml_irkit_req_wait(&req.base);
if (req.base.error != MXM_OK) {
return OSHMEM_ERROR;
}
return OSHMEM_SUCCESS;
}
Показать git blame Копировать постоянную ссылку