1
1
openmpi/ompi/mca/btl/openib/btl_openib.c
Gleb Natapov 2b6cbd6299 Separate frag lists for RDMA descriptors to two, one for src descriptors
and another for dst descriptors. This provide partial solution to OB1 protocol
deadlock problem. We can limit number of RDMA descriptors (by setting
btl_openib_free_list_max to something different from -1) and if we will be
lucky to hit this limit before we fail to register more memory the protocol
will not deadlock. When we had only one list for src/dst descriptors we
deadlocked when we reached max limit for the list.

This commit was SVN r13844.
2007-02-28 13:43:38 +00:00

908 строки
32 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 (c) 2007 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2006-2007 Mellanox Technologies. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include <string.h>
#include <inttypes.h>
#include "opal/util/output.h"
#include "opal/util/if.h"
#include "opal/util/show_help.h"
#include "ompi/mca/pml/pml.h"
#include "ompi/mca/btl/btl.h"
#include "ompi/mca/btl/base/btl_base_error.h"
#include "btl_openib.h"
#include "btl_openib_frag.h"
#include "btl_openib_proc.h"
#include "btl_openib_endpoint.h"
#include "ompi/datatype/convertor.h"
#include "ompi/datatype/datatype.h"
#include "ompi/mca/mpool/base/base.h"
#include "ompi/mca/mpool/mpool.h"
#include "ompi/mca/mpool/rdma/mpool_rdma.h"
#include "orte/util/sys_info.h"
#include <errno.h>
#include <string.h>
#include <math.h>
#include <inttypes.h>
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif
#ifdef HAVE_SYS_TIME_H
#include <sys/time.h>
#endif
#ifdef HAVE_SYS_RESOURCE_H
#include <sys/resource.h>
#endif
mca_btl_openib_module_t mca_btl_openib_module = {
{
&mca_btl_openib_component.super,
0, /* max size of first fragment */
0, /* min send fragment size */
0, /* max send fragment size */
0, /* min rdma fragment size */
0, /* max rdma fragment size */
0, /* exclusivity */
0, /* latency */
0, /* bandwidth */
0, /* TODO this should be PUT btl flags */
mca_btl_openib_add_procs,
mca_btl_openib_del_procs,
mca_btl_openib_register,
mca_btl_openib_finalize,
/* we need alloc free, pack */
mca_btl_openib_alloc,
mca_btl_openib_free,
mca_btl_openib_prepare_src,
mca_btl_openib_prepare_dst,
mca_btl_openib_send,
mca_btl_openib_put,
mca_btl_openib_get,
mca_btl_base_dump,
NULL, /* mpool */
mca_btl_openib_register_error_cb /* error call back registration */
}
};
int mca_btl_openib_size_queues( struct mca_btl_openib_module_t* openib_btl, size_t nprocs);
static void show_init_error(const char *file, int line,
const char *func, const char *dev)
{
if (ENOMEM == errno) {
int ret;
struct rlimit limit;
char *str_limit = NULL;
ret = getrlimit(RLIMIT_MEMLOCK, &limit);
if (0 != ret) {
asprintf(&str_limit, "Unknown");
} else if (limit.rlim_cur == RLIM_INFINITY) {
asprintf(&str_limit, "unlimited");
} else {
asprintf(&str_limit, "%ld", (long)limit.rlim_cur);
}
opal_show_help("help-mpi-btl-openib.txt", "init-fail-no-mem",
true, orte_system_info.nodename,
file, line, func, dev, str_limit);
if (NULL != str_limit) free(str_limit);
} else {
opal_show_help("help-mpi-btl-openib.txt", "init-fail-create-q",
true, orte_system_info.nodename,
file, line, func, strerror(errno), errno, dev);
}
}
/*
* add a proc to this btl module
* creates an endpoint that is setup on the
* first send to the endpoint
*/
int mca_btl_openib_add_procs(
struct mca_btl_base_module_t* btl,
size_t nprocs,
struct ompi_proc_t **ompi_procs,
struct mca_btl_base_endpoint_t** peers,
ompi_bitmap_t* reachable)
{
mca_btl_openib_module_t* openib_btl = (mca_btl_openib_module_t*)btl;
int i,j, rc;
int rem_subnet_id_port_cnt;
int lcl_subnet_id_port_cnt = 0;
int btl_rank = 0;
for(j=0; j < mca_btl_openib_component.ib_num_btls; j++){
if(mca_btl_openib_component.openib_btls[j].port_info.subnet_id
== openib_btl->port_info.subnet_id) {
lcl_subnet_id_port_cnt++;
}
if(openib_btl == &(mca_btl_openib_component.openib_btls[j])) {
btl_rank = j;
}
}
for(i = 0; i < (int) nprocs; i++) {
struct ompi_proc_t* ompi_proc = ompi_procs[i];
mca_btl_openib_proc_t* ib_proc;
mca_btl_base_endpoint_t* endpoint;
if(NULL == (ib_proc = mca_btl_openib_proc_create(ompi_proc))) {
return OMPI_ERR_OUT_OF_RESOURCE;
}
rem_subnet_id_port_cnt = 0;
/* check if the remote proc has a reachable subnet first */
BTL_VERBOSE(("got %d port_infos \n", ib_proc->proc_port_count));
for(j = 0; j < (int) ib_proc->proc_port_count; j++){
BTL_VERBOSE(("got a subnet %016x\n",
ib_proc->proc_ports[j].subnet_id));
if(ib_proc->proc_ports[j].subnet_id ==
openib_btl->port_info.subnet_id) {
BTL_VERBOSE(("Got a matching subnet!\n"));
rem_subnet_id_port_cnt ++;
}
}
if(!rem_subnet_id_port_cnt ) {
/* no use trying to communicate with this endpointlater */
BTL_VERBOSE(("No matching subnet id was found, moving on.. \n"));
continue;
}
#if 0
num_endpoints = rem_subnet_id_port_cnt / lcl_subnet_id_port_cnt +
(btl_rank < (rem_subnet_id_port_cnt / lcl_subnet_id_port_cnt)) ? 1:0;
#endif
if(rem_subnet_id_port_cnt < lcl_subnet_id_port_cnt &&
btl_rank >= rem_subnet_id_port_cnt ) {
BTL_VERBOSE(("Not enough remote ports on this subnet id, moving on.. \n"));
continue;
}
OPAL_THREAD_LOCK(&ib_proc->proc_lock);
/* The btl_proc datastructure is shared by all IB PTL
* instances that are trying to reach this destination.
* Cache the peer instance on the btl_proc.
*/
endpoint = OBJ_NEW(mca_btl_openib_endpoint_t);
if(NULL == endpoint) {
OPAL_THREAD_UNLOCK(&ib_proc->proc_lock);
return OMPI_ERR_OUT_OF_RESOURCE;
}
endpoint->endpoint_btl = openib_btl;
endpoint->use_eager_rdma = openib_btl->hca->use_eager_rdma &
mca_btl_openib_component.use_eager_rdma;
endpoint->subnet_id = openib_btl->port_info.subnet_id;
rc = mca_btl_openib_proc_insert(ib_proc, endpoint);
if(rc != OMPI_SUCCESS) {
OBJ_RELEASE(endpoint);
OPAL_THREAD_UNLOCK(&ib_proc->proc_lock);
continue;
}
orte_pointer_array_add((orte_std_cntr_t*)&endpoint->index,
openib_btl->endpoints, (void*)endpoint);
ompi_bitmap_set_bit(reachable, i);
OPAL_THREAD_UNLOCK(&ib_proc->proc_lock);
peers[i] = endpoint;
}
return mca_btl_openib_size_queues(openib_btl, nprocs);
}
int mca_btl_openib_size_queues( struct mca_btl_openib_module_t* openib_btl, size_t nprocs)
{
int min_cq_size;
int first_time = (0 == openib_btl->num_peers);
int rc;
openib_btl->num_peers += nprocs;
if(mca_btl_openib_component.use_srq) {
openib_btl->rd_num = mca_btl_openib_component.rd_num + log2(nprocs) * mca_btl_openib_component.srq_rd_per_peer;
if(openib_btl->rd_num > mca_btl_openib_component.srq_rd_max)
openib_btl->rd_num = mca_btl_openib_component.srq_rd_max;
openib_btl->rd_low = openib_btl->rd_num - 1;
min_cq_size = openib_btl->rd_num * 2 * openib_btl->num_peers;
if(!first_time) {
struct ibv_srq_attr srq_attr;
srq_attr.max_wr = openib_btl->rd_num;
rc = ibv_modify_srq(openib_btl->srq[BTL_OPENIB_HP_QP],
&srq_attr, IBV_SRQ_MAX_WR);
if(rc) {
BTL_ERROR(("cannot resize high priority shared receive queue, error: %d", rc));
return OMPI_ERROR;
}
rc = ibv_modify_srq(openib_btl->srq[BTL_OPENIB_LP_QP],
&srq_attr, IBV_SRQ_MAX_WR);
if(rc) {
BTL_ERROR(("cannot resize low priority shared receive queue, error: %d", rc));
return OMPI_ERROR;
}
}
} else
{
min_cq_size = ( mca_btl_openib_component.rd_num > (int32_t) mca_btl_openib_component.eager_rdma_num ?
mca_btl_openib_component.rd_num : (int32_t) mca_btl_openib_component.eager_rdma_num ) *
2 * openib_btl->num_peers;
}
if(min_cq_size > (int32_t) mca_btl_openib_component.ib_cq_size) {
mca_btl_openib_component.ib_cq_size = min_cq_size > openib_btl->hca->ib_dev_attr.max_cq ?
openib_btl->hca->ib_dev_attr.max_cq : min_cq_size;
#if OMPI_MCA_BTL_OPENIB_HAVE_RESIZE_CQ
if(!first_time) {
rc = ibv_resize_cq(openib_btl->ib_cq[BTL_OPENIB_LP_QP], mca_btl_openib_component.ib_cq_size);
if(rc) {
BTL_ERROR(("cannot resize low priority completion queue, error: %d", rc));
return OMPI_ERROR;
}
rc = ibv_resize_cq(openib_btl->ib_cq[BTL_OPENIB_HP_QP],
mca_btl_openib_component.ib_cq_size);
if(rc) {
BTL_ERROR(("cannot resize high priority completion queue, error: %d", rc));
return OMPI_ERROR;
}
}
#endif
}
if(first_time) {
/* never been here before, setup cq and srq */
mca_btl_openib_component.ib_cq_size = (int) mca_btl_openib_component.ib_cq_size >
openib_btl->hca->ib_dev_attr.max_cq ?
openib_btl->hca->ib_dev_attr.max_cq :
(int) mca_btl_openib_component.ib_cq_size;
return mca_btl_openib_create_cq_srq(openib_btl);
}
return OMPI_SUCCESS;
}
/*
* delete the proc as reachable from this btl module
*/
int mca_btl_openib_del_procs(struct mca_btl_base_module_t* btl,
size_t nprocs,
struct ompi_proc_t **procs,
struct mca_btl_base_endpoint_t ** peers)
{
BTL_DEBUG(("TODO\n"));
return OMPI_SUCCESS;
}
/*
*Register callback function to support send/recv semantics
*/
int mca_btl_openib_register(
struct mca_btl_base_module_t* btl,
mca_btl_base_tag_t tag,
mca_btl_base_module_recv_cb_fn_t cbfunc,
void* cbdata)
{
mca_btl_openib_module_t* openib_btl = (mca_btl_openib_module_t*) btl;
OPAL_THREAD_LOCK(&openib_btl->ib_lock);
openib_btl->ib_reg[tag].cbfunc = cbfunc;
openib_btl->ib_reg[tag].cbdata = cbdata;
OPAL_THREAD_UNLOCK(&openib_btl->ib_lock);
return OMPI_SUCCESS;
}
/*
*Register callback function for error handling..
*/
int mca_btl_openib_register_error_cb(
struct mca_btl_base_module_t* btl,
mca_btl_base_module_error_cb_fn_t cbfunc)
{
mca_btl_openib_module_t* openib_btl = (mca_btl_openib_module_t*) btl;
openib_btl->error_cb = cbfunc; /* stash for later */
return OMPI_SUCCESS;
}
/**
* Allocate a segment.
*
* @param btl (IN) BTL module
* @param size (IN) Request segment size.
*
* When allocating a segment we pull a pre-alllocated segment
* from one of two free lists, an eager list and a max list
*/
mca_btl_base_descriptor_t* mca_btl_openib_alloc(
struct mca_btl_base_module_t* btl,
size_t size)
{
mca_btl_openib_frag_t* frag = NULL;
mca_btl_openib_module_t* openib_btl;
int rc;
openib_btl = (mca_btl_openib_module_t*) btl;
if(size <= mca_btl_openib_component.eager_limit){
MCA_BTL_IB_FRAG_ALLOC_EAGER(btl, frag, rc);
} else if(size <= mca_btl_openib_component.max_send_size) {
MCA_BTL_IB_FRAG_ALLOC_MAX(btl, frag, rc);
}
if(NULL == frag)
return NULL;
frag->segment.seg_len = size <= openib_btl->super.btl_eager_limit ? size : openib_btl->super.btl_eager_limit;
frag->base.des_flags = 0;
return (mca_btl_base_descriptor_t*)frag;
}
/**
* Return a segment
*
* Return the segment to the appropriate
* preallocated segment list
*/
int mca_btl_openib_free(
struct mca_btl_base_module_t* btl,
mca_btl_base_descriptor_t* des)
{
mca_btl_openib_frag_t* frag = (mca_btl_openib_frag_t*)des;
if(((MCA_BTL_OPENIB_SEND_FRAG_FRAG == frag->type) ||
(MCA_BTL_OPENIB_RECV_FRAG_FRAG == frag->type))
&& frag->registration != NULL) {
btl->btl_mpool->mpool_deregister(btl->btl_mpool,
(mca_mpool_base_registration_t*)
frag->registration);
frag->registration = NULL;
}
MCA_BTL_IB_FRAG_RETURN(((mca_btl_openib_module_t*) btl), frag);
return OMPI_SUCCESS;
}
/**
* register user buffer or pack
* data into pre-registered buffer and return a
* descriptor that can be
* used for send/put.
*
* @param btl (IN) BTL module
* @param peer (IN) BTL peer addressing
*
* prepare source's behavior depends on the following:
* Has a valid memory registration been passed to prepare_src?
* if so we attempt to use the pre-registred user-buffer, if the memory registration
* is to small (only a portion of the user buffer) then we must reregister the user buffer
* Has the user requested the memory to be left pinned?
* if so we insert the memory registration into a memory tree for later lookup, we
* may also remove a previous registration if a MRU (most recently used) list of
* registions is full, this prevents resources from being exhausted.
* Is the requested size larger than the btl's max send size?
* if so and we aren't asked to leave the registration pinned than we register the memory if
* the users buffer is contiguous
* Otherwise we choose from two free lists of pre-registered memory in which to pack the data into.
*
*/
mca_btl_base_descriptor_t* mca_btl_openib_prepare_src(
struct mca_btl_base_module_t* btl,
struct mca_btl_base_endpoint_t* endpoint,
mca_mpool_base_registration_t* registration,
struct ompi_convertor_t* convertor,
size_t reserve,
size_t* size
)
{
mca_btl_openib_module_t *openib_btl;
mca_btl_openib_frag_t *frag = NULL;
mca_btl_openib_reg_t *openib_reg;
struct iovec iov;
uint32_t iov_count = 1;
size_t max_data = *size;
int rc;
openib_btl = (mca_btl_openib_module_t*)btl;
if(ompi_convertor_need_buffers(convertor) == false && 0 == reserve) {
if(registration != NULL || max_data > btl->btl_max_send_size) {
MCA_BTL_IB_FRAG_ALLOC_SEND_FRAG(btl, frag, rc);
if(NULL == frag) {
return NULL;
}
iov.iov_len = max_data;
iov.iov_base = NULL;
ompi_convertor_pack(convertor, &iov, &iov_count, &max_data);
*size = max_data;
if(NULL == registration) {
rc = btl->btl_mpool->mpool_register(btl->btl_mpool,
iov.iov_base, max_data, 0, &registration);
if(OMPI_SUCCESS != rc || NULL == registration) {
MCA_BTL_IB_FRAG_RETURN(openib_btl, frag);
return NULL;
}
/* keep track of the registration we did */
frag->registration = (mca_btl_openib_reg_t*)registration;
}
openib_reg = (mca_btl_openib_reg_t*)registration;
frag->base.des_flags = 0;
frag->base.des_src = &frag->segment;
frag->base.des_src_cnt = 1;
frag->base.des_dst = NULL;
frag->base.des_dst_cnt = 0;
frag->base.des_flags = 0;
frag->sg_entry.length = max_data;
frag->sg_entry.lkey = openib_reg->mr->lkey;
frag->sg_entry.addr = (unsigned long)iov.iov_base;
frag->segment.seg_len = max_data;
frag->segment.seg_addr.pval = iov.iov_base;
frag->segment.seg_key.key32[0] = (uint32_t)frag->sg_entry.lkey;
BTL_VERBOSE(("frag->sg_entry.lkey = %lu .addr = %llu "
"frag->segment.seg_key.key32[0] = %lu",
frag->sg_entry.lkey, frag->sg_entry.addr,
frag->segment.seg_key.key32[0]));
return &frag->base;
}
}
if(max_data + reserve <= btl->btl_eager_limit) {
/* the data is small enough to fit in the eager frag and
* memory is not prepinned */
MCA_BTL_IB_FRAG_ALLOC_EAGER(btl, frag, rc);
}
if(NULL == frag) {
/* the data doesn't fit into eager frag or eger frag is
* not available */
MCA_BTL_IB_FRAG_ALLOC_MAX(btl, frag, rc);
if(NULL == frag) {
return NULL;
}
if(max_data + reserve > btl->btl_max_send_size) {
max_data = btl->btl_max_send_size - reserve;
}
}
iov.iov_len = max_data;
iov.iov_base = (unsigned char*)frag->segment.seg_addr.pval + reserve;
rc = ompi_convertor_pack(convertor, &iov, &iov_count, &max_data);
if(rc < 0) {
MCA_BTL_IB_FRAG_RETURN(openib_btl, frag);
return NULL;
}
*size = max_data;
frag->segment.seg_len = max_data + reserve;
frag->segment.seg_key.key32[0] = (uint32_t)frag->sg_entry.lkey;
frag->base.des_src = &frag->segment;
frag->base.des_src_cnt = 1;
frag->base.des_dst = NULL;
frag->base.des_dst_cnt = 0;
frag->base.des_flags = 0;
return &frag->base;
}
/**
* Prepare the dst buffer
*
* @param btl (IN) BTL module
* @param peer (IN) BTL peer addressing
* prepare dest's behavior depends on the following:
* Has a valid memory registration been passed to prepare_src?
* if so we attempt to use the pre-registred user-buffer, if the memory registration
* is to small (only a portion of the user buffer) then we must reregister the user buffer
* Has the user requested the memory to be left pinned?
* if so we insert the memory registration into a memory tree for later lookup, we
* may also remove a previous registration if a MRU (most recently used) list of
* registions is full, this prevents resources from being exhausted.
*/
mca_btl_base_descriptor_t* mca_btl_openib_prepare_dst(
struct mca_btl_base_module_t* btl,
struct mca_btl_base_endpoint_t* endpoint,
mca_mpool_base_registration_t* registration,
struct ompi_convertor_t* convertor,
size_t reserve,
size_t* size)
{
mca_btl_openib_module_t *openib_btl;
mca_btl_openib_frag_t *frag;
mca_btl_openib_reg_t *openib_reg;
int rc;
ptrdiff_t lb;
openib_btl = (mca_btl_openib_module_t*)btl;
MCA_BTL_IB_FRAG_ALLOC_RECV_FRAG(btl, frag, rc);
if(NULL == frag) {
return NULL;
}
ompi_ddt_type_lb(convertor->pDesc, &lb);
frag->segment.seg_addr.pval = convertor->pBaseBuf + lb +
convertor->bConverted;
if(NULL == registration){
/* we didn't get a memory registration passed in, so we have to
* register the region ourselves
*/
rc = btl->btl_mpool->mpool_register(btl->btl_mpool,
frag->segment.seg_addr.pval, *size, 0, &registration);
if(OMPI_SUCCESS != rc || NULL == registration) {
MCA_BTL_IB_FRAG_RETURN(openib_btl, frag);
return NULL;
}
/* keep track of the registration we did */
frag->registration = (mca_btl_openib_reg_t*)registration;
}
openib_reg = (mca_btl_openib_reg_t*)registration;
frag->sg_entry.length = *size;
frag->sg_entry.lkey = openib_reg->mr->lkey;
frag->sg_entry.addr = (unsigned long) frag->segment.seg_addr.pval;
frag->segment.seg_len = *size;
frag->segment.seg_key.key32[0] = openib_reg->mr->rkey;
frag->base.des_dst = &frag->segment;
frag->base.des_dst_cnt = 1;
frag->base.des_src = NULL;
frag->base.des_src_cnt = 0;
frag->base.des_flags = 0;
BTL_VERBOSE(("frag->sg_entry.lkey = %lu .addr = %llu "
"frag->segment.seg_key.key32[0] = %lu",
frag->sg_entry.lkey, frag->sg_entry.addr,
frag->segment.seg_key.key32[0]));
return &frag->base;
}
int mca_btl_openib_finalize(struct mca_btl_base_module_t* btl)
{
mca_btl_openib_module_t* openib_btl;
openib_btl = (mca_btl_openib_module_t*) btl;
#if 0
if(openib_btl->send_free_eager.fl_num_allocated !=
openib_btl->send_free_eager.super.opal_list_length){
opal_output(0, "btl ib send_free_eager frags: %d allocated %d returned \n",
openib_btl->send_free_eager.fl_num_allocated,
openib_btl->send_free_eager.super.opal_list_length);
}
if(openib_btl->send_free_max.fl_num_allocated !=
openib_btl->send_free_max.super.opal_list_length){
opal_output(0, "btl ib send_free_max frags: %d allocated %d returned \n",
openib_btl->send_free_max.fl_num_allocated,
openib_btl->send_free_max.super.opal_list_length);
}
if(openib_btl->send_free_frag.fl_num_allocated !=
openib_btl->send_free_frag.super.opal_list_length){
opal_output(0, "btl ib send_free_frag frags: %d allocated %d returned \n",
openib_btl->send_free_frag.fl_num_allocated,
openib_btl->send_free_frag.super.opal_list_length);
}
if(openib_btl->recv_free_eager.fl_num_allocated !=
openib_btl->recv_free_eager.super.opal_list_length){
opal_output(0, "btl ib recv_free_eager frags: %d allocated %d returned \n",
openib_btl->recv_free_eager.fl_num_allocated,
openib_btl->recv_free_eager.super.opal_list_length);
}
if(openib_btl->recv_free_max.fl_num_allocated !=
openib_btl->recv_free_max.super.opal_list_length){
opal_output(0, "btl ib recv_free_max frags: %d allocated %d returned \n",
openib_btl->recv_free_max.fl_num_allocated,
openib_btl->recv_free_max.super.opal_list_length);
}
#endif
#if OMPI_ENABLE_PROGRESS_THREADS == 1
if(openib_btl->hca->progress) {
openib_btl->hca->progress = false;
if (pthread_cancel(openib_btl->hca->thread.t_handle))
BTL_ERROR(("Failed to cancel OpenIB progress thread"));
opal_thread_join(&openib_btl->hca->thread, NULL);
}
#endif
return OMPI_SUCCESS;
}
/*
* Initiate a send.
*/
int mca_btl_openib_send(
struct mca_btl_base_module_t* btl,
struct mca_btl_base_endpoint_t* endpoint,
struct mca_btl_base_descriptor_t* descriptor,
mca_btl_base_tag_t tag)
{
mca_btl_openib_frag_t* frag = (mca_btl_openib_frag_t*)descriptor;
frag->endpoint = endpoint;
frag->hdr->tag = tag;
frag->wr_desc.sr_desc.opcode = IBV_WR_SEND;
return mca_btl_openib_endpoint_send(endpoint, frag);
}
/*
* RDMA WRITE local buffer to remote buffer address.
*/
int mca_btl_openib_put( mca_btl_base_module_t* btl,
mca_btl_base_endpoint_t* endpoint,
mca_btl_base_descriptor_t* descriptor)
{
int rc;
struct ibv_send_wr* bad_wr;
mca_btl_openib_frag_t* frag = (mca_btl_openib_frag_t*) descriptor;
mca_btl_openib_module_t* openib_btl = (mca_btl_openib_module_t*) btl;
/* setup for queued requests */
frag->endpoint = endpoint;
frag->wr_desc.sr_desc.opcode = IBV_WR_RDMA_WRITE;
/* check for a send wqe */
if (OPAL_THREAD_ADD32(&endpoint->sd_wqe[BTL_OPENIB_LP_QP],-1) < 0) {
OPAL_THREAD_ADD32(&endpoint->sd_wqe[BTL_OPENIB_LP_QP],1);
OPAL_THREAD_LOCK(&endpoint->endpoint_lock);
opal_list_append(&endpoint->pending_put_frags, (opal_list_item_t *)frag);
OPAL_THREAD_UNLOCK(&endpoint->endpoint_lock);
return OMPI_SUCCESS;
/* post descriptor */
} else {
frag->wr_desc.sr_desc.send_flags = IBV_SEND_SIGNALED;
frag->wr_desc.sr_desc.wr.rdma.remote_addr = frag->base.des_dst->seg_addr.lval;
frag->wr_desc.sr_desc.wr.rdma.rkey = frag->base.des_dst->seg_key.key32[0];
frag->sg_entry.addr = (unsigned long) frag->base.des_src->seg_addr.pval;
frag->sg_entry.length = frag->base.des_src->seg_len;
if(ibv_post_send(endpoint->lcl_qp[BTL_OPENIB_LP_QP],
&frag->wr_desc.sr_desc,
&bad_wr)){
rc = OMPI_ERROR;
} else {
rc = OMPI_SUCCESS;
}
if(mca_btl_openib_component.use_srq) {
mca_btl_openib_post_srr(openib_btl, 1, BTL_OPENIB_HP_QP);
mca_btl_openib_post_srr(openib_btl, 1, BTL_OPENIB_LP_QP);
} else {
btl_openib_endpoint_post_rr(endpoint, 1, BTL_OPENIB_HP_QP);
btl_openib_endpoint_post_rr(endpoint, 1, BTL_OPENIB_LP_QP);
}
}
return rc;
}
/*
* RDMA READ remote buffer to local buffer address.
*/
int mca_btl_openib_get( mca_btl_base_module_t* btl,
mca_btl_base_endpoint_t* endpoint,
mca_btl_base_descriptor_t* descriptor)
{
int rc;
struct ibv_send_wr* bad_wr;
mca_btl_openib_frag_t* frag = (mca_btl_openib_frag_t*) descriptor;
mca_btl_openib_module_t* openib_btl = (mca_btl_openib_module_t*) btl;
frag->endpoint = endpoint;
frag->wr_desc.sr_desc.opcode = IBV_WR_RDMA_READ;
/* check for a send wqe */
if (OPAL_THREAD_ADD32(&endpoint->sd_wqe[BTL_OPENIB_LP_QP],-1) < 0) {
OPAL_THREAD_ADD32(&endpoint->sd_wqe[BTL_OPENIB_LP_QP],1);
OPAL_THREAD_LOCK(&endpoint->endpoint_lock);
opal_list_append(&endpoint->pending_get_frags, (opal_list_item_t*)frag);
OPAL_THREAD_UNLOCK(&endpoint->endpoint_lock);
return OMPI_SUCCESS;
/* check for a get token */
} else if(OPAL_THREAD_ADD32(&endpoint->get_tokens,-1) < 0) {
OPAL_THREAD_ADD32(&endpoint->sd_wqe[BTL_OPENIB_LP_QP],1);
OPAL_THREAD_ADD32(&endpoint->get_tokens,1);
OPAL_THREAD_LOCK(&endpoint->endpoint_lock);
opal_list_append(&endpoint->pending_get_frags, (opal_list_item_t*)frag);
OPAL_THREAD_UNLOCK(&endpoint->endpoint_lock);
return OMPI_SUCCESS;
} else {
frag->wr_desc.sr_desc.send_flags = IBV_SEND_SIGNALED;
frag->wr_desc.sr_desc.wr.rdma.remote_addr = frag->base.des_src->seg_addr.lval;
frag->wr_desc.sr_desc.wr.rdma.rkey = frag->base.des_src->seg_key.key32[0];
frag->sg_entry.addr = (unsigned long) frag->base.des_dst->seg_addr.pval;
frag->sg_entry.length = frag->base.des_dst->seg_len;
if(ibv_post_send(endpoint->lcl_qp[BTL_OPENIB_LP_QP],
&frag->wr_desc.sr_desc,
&bad_wr)){
BTL_ERROR(("error posting send request errno (%d) says %s", errno, strerror(errno)));
rc = ORTE_ERROR;
} else {
rc = ORTE_SUCCESS;
}
if(mca_btl_openib_component.use_srq) {
mca_btl_openib_post_srr(openib_btl, 1, BTL_OPENIB_HP_QP);
mca_btl_openib_post_srr(openib_btl, 1, BTL_OPENIB_LP_QP);
} else {
btl_openib_endpoint_post_rr(endpoint, 1, BTL_OPENIB_HP_QP);
btl_openib_endpoint_post_rr(endpoint, 1, BTL_OPENIB_LP_QP);
}
}
return rc;
}
/*
* create both the high and low priority completion queues
* and the shared receive queue (if requested)
*/
int mca_btl_openib_create_cq_srq(mca_btl_openib_module_t *openib_btl)
{
/* Allocate Protection Domain */
openib_btl->poll_cq = false;
if (mca_btl_openib_component.use_srq) {
struct ibv_srq_init_attr attr;
attr.attr.max_wr = mca_btl_openib_component.srq_rd_max;
attr.attr.max_sge = mca_btl_openib_component.ib_sg_list_size;
openib_btl->srd_posted[BTL_OPENIB_HP_QP] = 0;
openib_btl->srd_posted[BTL_OPENIB_LP_QP] = 0;
openib_btl->srq[BTL_OPENIB_HP_QP] =
ibv_create_srq(openib_btl->hca->ib_pd, &attr);
if (NULL == openib_btl->srq[BTL_OPENIB_HP_QP]) {
show_init_error(__FILE__, __LINE__, "ibv_create_srq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
openib_btl->srq[BTL_OPENIB_LP_QP] =
ibv_create_srq(openib_btl->hca->ib_pd, &attr);
if (NULL == openib_btl->srq[BTL_OPENIB_LP_QP]) {
show_init_error(__FILE__, __LINE__, "ibv_create_srq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
} else {
openib_btl->srq[BTL_OPENIB_HP_QP] = NULL;
openib_btl->srq[BTL_OPENIB_LP_QP] = NULL;
}
/* Create the low and high priority queue pairs */
#if OMPI_ENABLE_PROGRESS_THREADS == 1
#if OMPI_MCA_BTL_OPENIB_IBV_CREATE_CQ_ARGS == 3
openib_btl->ib_cq[BTL_OPENIB_LP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, openib_btl->hca->ib_channel);
#else
openib_btl->ib_cq[BTL_OPENIB_LP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, openib_btl, openib_btl->hca->ib_channel, 0);
#endif
#else /* OMPI_ENABLE_PROGRESS_THREADS DISABLED */
#if OMPI_MCA_BTL_OPENIB_IBV_CREATE_CQ_ARGS == 3
openib_btl->ib_cq[BTL_OPENIB_LP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, NULL);
#else
openib_btl->ib_cq[BTL_OPENIB_LP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, NULL, NULL, 0);
#endif
#endif /* OMPI_ENABLE_PROGRESS_THREADS */
if (NULL == openib_btl->ib_cq[BTL_OPENIB_LP_QP]) {
show_init_error(__FILE__, __LINE__, "ibv_create_cq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
#if OMPI_ENABLE_PROGRESS_THREADS == 1
if(ibv_req_notify_cq(openib_btl->ib_cq[BTL_OPENIB_LP_QP], 0)) {
show_init_error(__FILE__, __LINE__, "ibv_req_notify_cq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
#if OMPI_MCA_BTL_OPENIB_IBV_CREATE_CQ_ARGS == 3
openib_btl->ib_cq[BTL_OPENIB_HP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, openib_btl->hca->ib_channel);
#else
openib_btl->ib_cq[BTL_OPENIB_HP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, openib_btl, openib_btl->hca->ib_channel, 0);
#endif
#else /* OMPI_ENABLE_PROGRESS_THREADS DISABLED */
#if OMPI_MCA_BTL_OPENIB_IBV_CREATE_CQ_ARGS == 3
openib_btl->ib_cq[BTL_OPENIB_HP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, NULL);
#else
openib_btl->ib_cq[BTL_OPENIB_HP_QP] =
ibv_create_cq(openib_btl->hca->ib_dev_context,
mca_btl_openib_component.ib_cq_size, NULL, NULL, 0);
#endif
#endif /* OMPI_ENABLE_PROGRESS_THREADS */
if(NULL == openib_btl->ib_cq[BTL_OPENIB_HP_QP]) {
show_init_error(__FILE__, __LINE__, "ibv_create_cq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
#if OMPI_ENABLE_PROGRESS_THREADS == 1
if(ibv_req_notify_cq(openib_btl->ib_cq[BTL_OPENIB_HP_QP], 0)) {
show_init_error(__FILE__, __LINE__, "ibv_req_notify_cq",
ibv_get_device_name(openib_btl->hca->ib_dev));
return OMPI_ERROR;
}
OPAL_THREAD_LOCK(&openib_btl->hca->hca_lock);
if (!openib_btl->hca->progress){
int rc;
openib_btl->hca->progress = true;
if(OPAL_SUCCESS != (rc = opal_thread_start(&openib_btl->hca->thread))) {
BTL_ERROR(("Unable to create progress thread, retval=%d", rc));
return rc;
}
}
OPAL_THREAD_UNLOCK(&openib_btl->hca->hca_lock);
#endif
return OMPI_SUCCESS;
}