433f8a7694
receive queues are shared among all PMLs, they are declared in the base PML, and the selected PML is in charge of initializing and releasing them. The CM PML is slightly different compared with OB1 or DR. Internally it use 2 different types of requests: light and heavy. However, now with this patch both types of requests are stored in the same queue, and cast appropriately on the allocation macro. This means we might use less memory than we allocate, but in exchange we got full support for most of the parallel debuggers. Another thing with this patch, is that now for all PML (CM included) the basic PML requests start with the same fields, and they are declared in the same order in the request structure. Moreover, the fields have been moved in such a way that only one volatile/atomic will exist per line of cache (hopefully). This commit was SVN r15346.
401 строка
20 KiB
C
401 строка
20 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) 2004-2005 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
|
|
*/
|
|
#ifndef OMPI_PML_OB1_RECV_REQUEST_H
|
|
#define OMPI_PML_OB1_RECV_REQUEST_H
|
|
|
|
#include "pml_ob1.h"
|
|
#include "pml_ob1_rdma.h"
|
|
#include "pml_ob1_rdmafrag.h"
|
|
#include "ompi/proc/proc.h"
|
|
#include "ompi/mca/pml/ob1/pml_ob1_comm.h"
|
|
#include "ompi/mca/mpool/base/base.h"
|
|
#include "ompi/mca/pml/base/pml_base_recvreq.h"
|
|
#include "ompi/datatype/datatype.h"
|
|
|
|
#if defined(c_plusplus) || defined(__cplusplus)
|
|
extern "C" {
|
|
#endif
|
|
|
|
|
|
struct mca_pml_ob1_recv_request_t {
|
|
mca_pml_base_recv_request_t req_recv;
|
|
ompi_ptr_t req_send;
|
|
#if OMPI_HAVE_THREAD_SUPPORT
|
|
volatile int32_t req_lock;
|
|
#else
|
|
int32_t req_lock;
|
|
#endif
|
|
size_t req_pipeline_depth;
|
|
size_t req_bytes_received;
|
|
size_t req_bytes_delivered;
|
|
size_t req_rdma_offset;
|
|
size_t req_send_offset;
|
|
uint32_t req_rdma_cnt;
|
|
uint32_t req_rdma_idx;
|
|
bool req_pending;
|
|
bool req_ack_sent; /**< whether ack was sent to the sender */
|
|
mca_pml_ob1_com_btl_t req_rdma[1];
|
|
};
|
|
typedef struct mca_pml_ob1_recv_request_t mca_pml_ob1_recv_request_t;
|
|
|
|
OBJ_CLASS_DECLARATION(mca_pml_ob1_recv_request_t);
|
|
|
|
|
|
/**
|
|
* Allocate a recv request from the modules free list.
|
|
*
|
|
* @param rc (OUT) OMPI_SUCCESS or error status on failure.
|
|
* @return Receive request.
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_ALLOC(recvreq, rc) \
|
|
do { \
|
|
ompi_free_list_item_t* item; \
|
|
rc = OMPI_SUCCESS; \
|
|
OMPI_FREE_LIST_GET(&mca_pml_base_recv_requests, item, rc); \
|
|
recvreq = (mca_pml_ob1_recv_request_t*)item; \
|
|
} while(0)
|
|
|
|
|
|
/**
|
|
* Initialize a receive request with call parameters.
|
|
*
|
|
* @param request (IN) Receive request.
|
|
* @param addr (IN) User buffer.
|
|
* @param count (IN) Number of elements of indicated datatype.
|
|
* @param datatype (IN) User defined datatype.
|
|
* @param src (IN) Source rank w/in the communicator.
|
|
* @param tag (IN) User defined tag.
|
|
* @param comm (IN) Communicator.
|
|
* @param persistent (IN) Is this a ersistent request.
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_INIT( request, \
|
|
addr, \
|
|
count, \
|
|
datatype, \
|
|
src, \
|
|
tag, \
|
|
comm, \
|
|
persistent) \
|
|
do { \
|
|
MCA_PML_BASE_RECV_REQUEST_INIT( &(request)->req_recv, \
|
|
addr, \
|
|
count, \
|
|
datatype, \
|
|
src, \
|
|
tag, \
|
|
comm, \
|
|
persistent); \
|
|
} while(0)
|
|
|
|
/**
|
|
* Mark the request as completed at MPI level for internal purposes.
|
|
*
|
|
* @param recvreq (IN) Receive request.
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_MPI_COMPLETE( recvreq ) \
|
|
do { \
|
|
PERUSE_TRACE_COMM_EVENT( PERUSE_COMM_REQ_COMPLETE, \
|
|
&(recvreq->req_recv.req_base), PERUSE_RECV ); \
|
|
MCA_PML_BASE_REQUEST_MPI_COMPLETE( &(recvreq->req_recv.req_base.req_ompi) ); \
|
|
} while (0)
|
|
|
|
/**
|
|
* Return a recv request to the modules free list.
|
|
*
|
|
* @param recvreq (IN) Receive request.
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_PML_COMPLETE(recvreq) \
|
|
do { \
|
|
size_t r; \
|
|
\
|
|
assert( false == recvreq->req_recv.req_base.req_pml_complete ); \
|
|
\
|
|
if((recvreq)->req_recv.req_bytes_packed > 0) { \
|
|
PERUSE_TRACE_COMM_EVENT( PERUSE_COMM_REQ_XFER_END, \
|
|
&(recvreq->req_recv.req_base), PERUSE_RECV ); \
|
|
} \
|
|
\
|
|
for( r = 0; r < recvreq->req_rdma_cnt; r++ ) { \
|
|
mca_mpool_base_registration_t* btl_reg = recvreq->req_rdma[r].btl_reg; \
|
|
if( NULL != btl_reg && btl_reg->mpool != NULL) { \
|
|
btl_reg->mpool->mpool_deregister( btl_reg->mpool, btl_reg ); \
|
|
} \
|
|
} \
|
|
recvreq->req_rdma_cnt = 0; \
|
|
\
|
|
OPAL_THREAD_LOCK(&ompi_request_lock); \
|
|
\
|
|
if( true == recvreq->req_recv.req_base.req_free_called ) { \
|
|
MCA_PML_OB1_RECV_REQUEST_RETURN( recvreq ); \
|
|
} else { \
|
|
/* initialize request status */ \
|
|
recvreq->req_recv.req_base.req_pml_complete = true; \
|
|
recvreq->req_recv.req_base.req_ompi.req_status._count = \
|
|
(int)recvreq->req_bytes_received; \
|
|
if (recvreq->req_bytes_received > recvreq->req_bytes_delivered) { \
|
|
recvreq->req_recv.req_base.req_ompi.req_status._count = \
|
|
(int)recvreq->req_bytes_delivered; \
|
|
recvreq->req_recv.req_base.req_ompi.req_status.MPI_ERROR = \
|
|
MPI_ERR_TRUNCATE; \
|
|
} \
|
|
MCA_PML_OB1_RECV_REQUEST_MPI_COMPLETE( recvreq ); \
|
|
} \
|
|
OPAL_THREAD_UNLOCK(&ompi_request_lock); \
|
|
} while(0)
|
|
|
|
/*
|
|
* Free the PML receive request
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_RETURN(recvreq) \
|
|
{ \
|
|
MCA_PML_BASE_RECV_REQUEST_FINI(&(recvreq)->req_recv); \
|
|
OMPI_FREE_LIST_RETURN( &mca_pml_base_recv_requests, \
|
|
(ompi_free_list_item_t*)(recvreq)); \
|
|
}
|
|
|
|
/**
|
|
* Attempt to match the request against the unexpected fragment list
|
|
* for all source ranks w/in the communicator.
|
|
*
|
|
* @param request (IN) Request to match.
|
|
*/
|
|
void mca_pml_ob1_recv_request_match_wild(mca_pml_ob1_recv_request_t* request);
|
|
|
|
/**
|
|
* Attempt to match the request against the unexpected fragment list
|
|
* for a specific source rank.
|
|
*
|
|
* @param request (IN) Request to match.
|
|
*/
|
|
void mca_pml_ob1_recv_request_match_specific(mca_pml_ob1_recv_request_t* request);
|
|
|
|
/**
|
|
* Initialize diagnostic code for tracing rdma protocol timing
|
|
*/
|
|
|
|
/**
|
|
* Start an initialized request.
|
|
*
|
|
* @param request Receive request.
|
|
* @return OMPI_SUCESS or error status on failure.
|
|
*/
|
|
#define MCA_PML_OB1_RECV_REQUEST_START(request) \
|
|
do { \
|
|
/* init/re-init the request */ \
|
|
(request)->req_lock = 0; \
|
|
(request)->req_pipeline_depth = 0; \
|
|
(request)->req_bytes_received = 0; \
|
|
(request)->req_bytes_delivered = 0; \
|
|
/* What about req_rdma_cnt ? */ \
|
|
(request)->req_rdma_idx = 0; \
|
|
(request)->req_pending = false; \
|
|
(request)->req_ack_sent = false; \
|
|
\
|
|
MCA_PML_BASE_RECV_START( &(request)->req_recv.req_base ); \
|
|
\
|
|
/* attempt to match posted recv */ \
|
|
if((request)->req_recv.req_base.req_peer == OMPI_ANY_SOURCE) { \
|
|
mca_pml_ob1_recv_request_match_wild(request); \
|
|
} else { \
|
|
(request)->req_recv.req_base.req_proc = \
|
|
(request)->req_recv.req_base.req_comm->c_pml_comm->procs \
|
|
[(request)->req_recv.req_base.req_peer].ompi_proc; \
|
|
if( (0 != (request)->req_recv.req_base.req_datatype->size) && \
|
|
(0 != (request)->req_recv.req_base.req_count) ) { \
|
|
ompi_convertor_copy_and_prepare_for_recv( \
|
|
(request)->req_recv.req_base.req_proc->proc_convertor, \
|
|
(request)->req_recv.req_base.req_datatype, \
|
|
(request)->req_recv.req_base.req_count, \
|
|
(request)->req_recv.req_base.req_addr, \
|
|
0, \
|
|
&(request)->req_recv.req_base.req_convertor ); \
|
|
ompi_convertor_get_unpacked_size( &(request)->req_recv.req_base.req_convertor, \
|
|
&(request)->req_bytes_delivered ); \
|
|
} \
|
|
mca_pml_ob1_recv_request_match_specific(request); \
|
|
} \
|
|
} while (0)
|
|
|
|
|
|
/**
|
|
*
|
|
*/
|
|
|
|
#define MCA_PML_OB1_RECV_REQUEST_MATCHED( request, hdr ) \
|
|
do { \
|
|
(request)->req_recv.req_base.req_ompi.req_status.MPI_SOURCE = (hdr)->hdr_src; \
|
|
(request)->req_recv.req_base.req_ompi.req_status.MPI_TAG = (hdr)->hdr_tag; \
|
|
\
|
|
PERUSE_TRACE_COMM_EVENT( PERUSE_COMM_MSG_MATCH_POSTED_REQ, \
|
|
&((request)->req_recv.req_base), PERUSE_RECV ); \
|
|
\
|
|
if((request)->req_recv.req_bytes_packed > 0) { \
|
|
if( MPI_ANY_SOURCE == (request)->req_recv.req_base.req_peer ) { \
|
|
ompi_convertor_copy_and_prepare_for_recv( \
|
|
(request)->req_recv.req_base.req_proc->proc_convertor, \
|
|
(request)->req_recv.req_base.req_datatype, \
|
|
(request)->req_recv.req_base.req_count, \
|
|
(request)->req_recv.req_base.req_addr, \
|
|
0, \
|
|
&(request)->req_recv.req_base.req_convertor ); \
|
|
ompi_convertor_get_unpacked_size( &(request)->req_recv.req_base.req_convertor, \
|
|
&(request)->req_bytes_delivered ); \
|
|
} \
|
|
PERUSE_TRACE_COMM_EVENT (PERUSE_COMM_REQ_XFER_BEGIN, \
|
|
&((request)->req_recv.req_base), PERUSE_RECV); \
|
|
} \
|
|
} while (0)
|
|
|
|
|
|
/**
|
|
*
|
|
*/
|
|
|
|
#define MCA_PML_OB1_RECV_REQUEST_UNPACK( \
|
|
request, \
|
|
segments, \
|
|
num_segments, \
|
|
seg_offset, \
|
|
data_offset, \
|
|
bytes_received, \
|
|
bytes_delivered) \
|
|
do { \
|
|
bytes_delivered = 0; \
|
|
if(request->req_recv.req_bytes_packed > 0) { \
|
|
struct iovec iov[MCA_BTL_DES_MAX_SEGMENTS]; \
|
|
uint32_t iov_count = 0; \
|
|
size_t max_data = bytes_received; \
|
|
size_t n, offset = seg_offset; \
|
|
mca_btl_base_segment_t* segment = segments; \
|
|
\
|
|
for( n = 0; n < num_segments; n++, segment++ ) { \
|
|
if(offset >= segment->seg_len) { \
|
|
offset -= segment->seg_len; \
|
|
} else { \
|
|
iov[iov_count].iov_len = segment->seg_len - offset; \
|
|
iov[iov_count].iov_base = (IOVBASE_TYPE*)((unsigned char*)segment->seg_addr.pval + offset); \
|
|
iov_count++; \
|
|
} \
|
|
} \
|
|
PERUSE_TRACE_COMM_OMPI_EVENT (PERUSE_COMM_REQ_XFER_CONTINUE, \
|
|
&(recvreq->req_recv.req_base), max_data, \
|
|
PERUSE_RECV); \
|
|
ompi_convertor_set_position( &(request->req_recv.req_base.req_convertor), \
|
|
&data_offset ); \
|
|
ompi_convertor_unpack( &(request)->req_recv.req_base.req_convertor, \
|
|
iov, \
|
|
&iov_count, \
|
|
&max_data ); \
|
|
bytes_delivered = max_data; \
|
|
} \
|
|
} while (0)
|
|
|
|
/**
|
|
*
|
|
*/
|
|
|
|
void mca_pml_ob1_recv_request_progress(
|
|
mca_pml_ob1_recv_request_t* req,
|
|
struct mca_btl_base_module_t* btl,
|
|
mca_btl_base_segment_t* segments,
|
|
size_t num_segments);
|
|
|
|
/**
|
|
*
|
|
*/
|
|
|
|
void mca_pml_ob1_recv_request_matched_probe(
|
|
mca_pml_ob1_recv_request_t* req,
|
|
struct mca_btl_base_module_t* btl,
|
|
mca_btl_base_segment_t* segments,
|
|
size_t num_segments);
|
|
|
|
/**
|
|
*
|
|
*/
|
|
|
|
int mca_pml_ob1_recv_request_schedule_exclusive(
|
|
mca_pml_ob1_recv_request_t* req, mca_bml_base_btl_t* start_bml_btl);
|
|
|
|
static inline void mca_pml_ob1_recv_request_schedule(
|
|
mca_pml_ob1_recv_request_t* req,
|
|
mca_bml_base_btl_t* start_bml_btl)
|
|
{
|
|
if(OPAL_THREAD_ADD32(&req->req_lock,1) == 1)
|
|
mca_pml_ob1_recv_request_schedule_exclusive(req, start_bml_btl);
|
|
}
|
|
|
|
#define MCA_PML_OB1_ADD_ACK_TO_PENDING(P, S, D, O) \
|
|
do { \
|
|
mca_pml_ob1_pckt_pending_t *_pckt; \
|
|
int _rc; \
|
|
\
|
|
MCA_PML_OB1_PCKT_PENDING_ALLOC(_pckt,_rc); \
|
|
_pckt->hdr.hdr_common.hdr_type = MCA_PML_OB1_HDR_TYPE_ACK; \
|
|
_pckt->hdr.hdr_ack.hdr_src_req.lval = (S); \
|
|
_pckt->hdr.hdr_ack.hdr_dst_req.pval = (D); \
|
|
_pckt->hdr.hdr_ack.hdr_send_offset = (O); \
|
|
_pckt->proc = (P); \
|
|
_pckt->bml_btl = NULL; \
|
|
OPAL_THREAD_LOCK(&mca_pml_ob1.lock); \
|
|
opal_list_append(&mca_pml_ob1.pckt_pending, \
|
|
(opal_list_item_t*)_pckt); \
|
|
OPAL_THREAD_UNLOCK(&mca_pml_ob1.lock); \
|
|
} while(0)
|
|
|
|
int mca_pml_ob1_recv_request_ack_send_btl(ompi_proc_t* proc,
|
|
mca_bml_base_btl_t* bml_btl, uint64_t hdr_src_req, void *hdr_dst_req,
|
|
uint64_t hdr_rdma_offset);
|
|
|
|
static inline int mca_pml_ob1_recv_request_ack_send(ompi_proc_t* proc,
|
|
uint64_t hdr_src_req, void *hdr_dst_req, uint64_t hdr_send_offset)
|
|
{
|
|
size_t i;
|
|
mca_bml_base_btl_t* bml_btl;
|
|
mca_bml_base_endpoint_t* endpoint =
|
|
(mca_bml_base_endpoint_t*)proc->proc_bml;
|
|
|
|
for(i = 0; i < mca_bml_base_btl_array_get_size(&endpoint->btl_eager); i++) {
|
|
bml_btl = mca_bml_base_btl_array_get_next(&endpoint->btl_eager);
|
|
if(mca_pml_ob1_recv_request_ack_send_btl(proc, bml_btl, hdr_src_req,
|
|
hdr_dst_req, hdr_send_offset) == OMPI_SUCCESS)
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
MCA_PML_OB1_ADD_ACK_TO_PENDING(proc, hdr_src_req, hdr_dst_req,
|
|
hdr_send_offset);
|
|
|
|
return OMPI_ERR_OUT_OF_RESOURCE;
|
|
}
|
|
|
|
int mca_pml_ob1_recv_request_get_frag(mca_pml_ob1_rdma_frag_t* frag);
|
|
|
|
/* This function tries to continue recvreq that stuck due to resource
|
|
* unavailability. Recvreq is added to recv_pending list if scheduling of put
|
|
* operation cannot be accomplished for some reason. */
|
|
void mca_pml_ob1_recv_request_process_pending(void);
|
|
|
|
#if defined(c_plusplus) || defined(__cplusplus)
|
|
}
|
|
#endif
|
|
#endif
|
|
|