6c5532072a
OMPI and a language agnostic part in OPAL. The convertor is completely moved into OPAL. This offers several benefits as described in RFC http://www.open-mpi.org/community/lists/devel/2009/07/6387.php namely: - Fewer basic types (int* and float* types, boolean and wchar - Fixing naming scheme to ompi-nomenclature. - Usability outside of the ompi-layer. - Due to the fixed nature of simple opal types, their information is completely known at compile time and therefore constified - With fewer datatypes (22), the actual sizes of bit-field types may be reduced from 64 to 32 bits, allowing reorganizing the opal_datatype structure, eliminating holes and keeping data required in convertor (upon send/recv) in one cacheline... This has implications to the convertor-datastructure and other parts of the code. - Several performance tests have been run, the netpipe latency does not change with this patch on Linux/x86-64 on the smoky cluster. - Extensive tests have been done to verify correctness (no new regressions) using: 1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and ompi-ddt: a. running both trunk and ompi-ddt resulted in no differences (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run correctly). b. with --enable-memchecker and running under valgrind (one buglet when run with static found in test-suite, commited) 2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt: all passed (except for the dynamic/ tests failed!! as trunk/MTT) 3. compilation and usage of HDF5 tests on Jaguar using PGI and PathScale compilers. 4. compilation and usage on Scicortex. - Please note, that for the heterogeneous case, (-m32 compiled binaries/ompi), neither ompi-trunk, nor ompi-ddt branch would successfully launch. This commit was SVN r21641.
143 строки
5.4 KiB
C
143 строки
5.4 KiB
C
/*
|
|
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
|
|
* University Research and Technology
|
|
* Corporation. All rights reserved.
|
|
* Copyright (c) 2004-2007 The University of Tennessee and The University
|
|
* of Tennessee Research Foundation. All rights
|
|
* reserved.
|
|
* Copyright (c) 2004-2008 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) 2006 Cisco Systems, Inc. All rights reserved.
|
|
* $COPYRIGHT$
|
|
*
|
|
* Additional copyrights may follow
|
|
*
|
|
* $HEADER$
|
|
*/
|
|
|
|
#include "ompi_config.h"
|
|
|
|
#include "pml_csum.h"
|
|
#include "pml_csum_recvreq.h"
|
|
#include "pml_csum_sendreq.h"
|
|
#include "ompi/memchecker.h"
|
|
|
|
|
|
int mca_pml_csum_start(size_t count, ompi_request_t** requests)
|
|
{
|
|
int rc;
|
|
size_t i;
|
|
bool reuse_old_request = true;
|
|
|
|
for(i=0; i<count; i++) {
|
|
mca_pml_base_request_t *pml_request = (mca_pml_base_request_t*)requests[i];
|
|
if(NULL == pml_request) {
|
|
continue;
|
|
}
|
|
if (OMPI_REQUEST_PML != requests[i]->req_type) {
|
|
continue;
|
|
}
|
|
|
|
/* If the persistent request is currently active - obtain the
|
|
* request lock and verify the status is incomplete. if the
|
|
* pml layer has not completed the request - mark the request
|
|
* as free called - so that it will be freed when the request
|
|
* completes - and create a new request.
|
|
*/
|
|
|
|
reuse_old_request = true;
|
|
switch(pml_request->req_ompi.req_state) {
|
|
case OMPI_REQUEST_INACTIVE:
|
|
if(pml_request->req_pml_complete == true)
|
|
break;
|
|
/* otherwise fall through */
|
|
case OMPI_REQUEST_ACTIVE: {
|
|
|
|
ompi_request_t *request;
|
|
OPAL_THREAD_LOCK(&ompi_request_lock);
|
|
if (pml_request->req_pml_complete == false) {
|
|
/* free request after it completes */
|
|
pml_request->req_free_called = true;
|
|
} else {
|
|
/* can reuse the existing request */
|
|
OPAL_THREAD_UNLOCK(&ompi_request_lock);
|
|
break;
|
|
}
|
|
|
|
reuse_old_request = false;
|
|
/* allocate a new request */
|
|
switch(pml_request->req_type) {
|
|
case MCA_PML_REQUEST_SEND: {
|
|
mca_pml_base_send_mode_t sendmode =
|
|
((mca_pml_base_send_request_t*)pml_request)->req_send_mode;
|
|
rc = mca_pml_csum_isend_init(
|
|
pml_request->req_addr,
|
|
pml_request->req_count,
|
|
pml_request->req_datatype,
|
|
pml_request->req_peer,
|
|
pml_request->req_tag,
|
|
sendmode,
|
|
pml_request->req_comm,
|
|
&request);
|
|
break;
|
|
}
|
|
case MCA_PML_REQUEST_RECV:
|
|
rc = mca_pml_csum_irecv_init(
|
|
pml_request->req_addr,
|
|
pml_request->req_count,
|
|
pml_request->req_datatype,
|
|
pml_request->req_peer,
|
|
pml_request->req_tag,
|
|
pml_request->req_comm,
|
|
&request);
|
|
break;
|
|
default:
|
|
rc = OMPI_ERR_REQUEST;
|
|
break;
|
|
}
|
|
OPAL_THREAD_UNLOCK(&ompi_request_lock);
|
|
if(OMPI_SUCCESS != rc)
|
|
return rc;
|
|
pml_request = (mca_pml_base_request_t*)request;
|
|
requests[i] = request;
|
|
break;
|
|
}
|
|
default:
|
|
return OMPI_ERR_REQUEST;
|
|
}
|
|
|
|
/* start the request */
|
|
switch(pml_request->req_type) {
|
|
case MCA_PML_REQUEST_SEND:
|
|
{
|
|
mca_pml_csum_send_request_t* sendreq = (mca_pml_csum_send_request_t*)pml_request;
|
|
if( reuse_old_request && (sendreq->req_send.req_bytes_packed != 0) ) {
|
|
size_t offset = 0;
|
|
/**
|
|
* Reset the convertor in case we're dealing with the original
|
|
* request, which when completed do not reset the convertor.
|
|
*/
|
|
opal_convertor_set_position( &sendreq->req_send.req_base.req_convertor,
|
|
&offset );
|
|
}
|
|
MCA_PML_CSUM_SEND_REQUEST_START(sendreq, rc);
|
|
if(rc != OMPI_SUCCESS)
|
|
return rc;
|
|
break;
|
|
}
|
|
case MCA_PML_REQUEST_RECV:
|
|
{
|
|
mca_pml_csum_recv_request_t* recvreq = (mca_pml_csum_recv_request_t*)pml_request;
|
|
MCA_PML_CSUM_RECV_REQUEST_START(recvreq);
|
|
break;
|
|
}
|
|
default:
|
|
return OMPI_ERR_REQUEST;
|
|
}
|
|
}
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|