MPI_IN_PLACE fix for reduce.
(actually a work around for an optimisation in the reduce for not saving ops on the first recv of each segment) Minor change in topo. This commit was SVN r7758.
Этот коммит содержится в:
Graham Fagg
родитель
48f2563b4c
Коммит
61b8218d76
@ -128,12 +128,7 @@ int mca_coll_tuned_reduce_intra_chain( void *sendbuf, void *recvbuf, int count,
|
||||
|
||||
/* ompi_coll_tuned_topo_dump_chain (chain, rank); */
|
||||
|
||||
/* ----------------------------------------------------------------- */
|
||||
/* MPI_IN_PLACE is not yet supported by ompi.. ?!.. */
|
||||
/* set the char * buffer pointers */
|
||||
|
||||
|
||||
#if 0
|
||||
if (sendbuf != MPI_IN_PLACE) {
|
||||
sendtmpbuf = (char*) sendbuf;
|
||||
}
|
||||
@ -143,26 +138,13 @@ int mca_coll_tuned_reduce_intra_chain( void *sendbuf, void *recvbuf, int count,
|
||||
accumbuf = (char *) recvbuf;
|
||||
|
||||
/* handle special case when size == 1 */
|
||||
if (size == 1) {
|
||||
if (1 == size ) {
|
||||
if (sendbuf != MPI_IN_PLACE) {
|
||||
ompi_ddt_copy_content_same_ddt( datatype, count, recvbuf, sendbuf );
|
||||
}
|
||||
return MPI_SUCCESS;
|
||||
}
|
||||
/* ----------------------------------------------------------------- */
|
||||
/* MPI_IN_PLACE is not yet supported by ompi.. ?!.. */
|
||||
|
||||
#else
|
||||
/* while MPI_IN_PLACE is not available */
|
||||
if (1 == size) {
|
||||
ompi_ddt_copy_content_same_ddt( datatype, count, recvbuf, sendbuf );
|
||||
return (OMPI_SUCCESS);
|
||||
}
|
||||
else {
|
||||
accumbuf = (char *) recvbuf;
|
||||
sendtmpbuf = (char *) sendbuf;
|
||||
}
|
||||
#endif
|
||||
/* ----------------------------------------------------------------- */
|
||||
|
||||
/* non-leaf nodes -
|
||||
@ -199,16 +181,29 @@ int mca_coll_tuned_reduce_intra_chain( void *sendbuf, void *recvbuf, int count,
|
||||
if (segindex < num_segments) {
|
||||
|
||||
if (0==i) { /* for the first step (1st child per segment) */
|
||||
/* we irecv directly into the accumulate buffer so that we */
|
||||
/* can reduce this with our sendbuf in one step */
|
||||
/* as op_reduce only has two buffer pointers, this avoids */
|
||||
/* we might be able to irecv directly into the accumulate buffer so that we */
|
||||
/* can reduce(op) this with our sendbuf in one step */
|
||||
/* as ompi_op_reduce only has two buffer pointers, this avoids */
|
||||
/* an extra memory copy GEF */
|
||||
ret = MCA_PML_CALL(irecv(accumbuf+segindex*realsegsize,recvcount,datatype,
|
||||
chain->chain_next[i],
|
||||
MCA_COLL_BASE_TAG_REDUCE,
|
||||
comm, &reqs[inbi]));
|
||||
if (ret != MPI_SUCCESS) { line = __LINE__; goto error_hndl; }
|
||||
}
|
||||
|
||||
/* BUT if we are root and are USING MPI_IN_PLACE this is wrong ek! */
|
||||
/* check for root might not be needed as it should be checked higher up */
|
||||
if ((MPI_IN_PLACE==sendbuf)&&(rank==root)) {
|
||||
ret = MCA_PML_CALL(irecv(inbuf[inbi],
|
||||
recvcount,datatype,
|
||||
chain->chain_next[i],
|
||||
MCA_COLL_BASE_TAG_REDUCE,
|
||||
comm, &reqs[inbi]));
|
||||
if (ret != MPI_SUCCESS) { line = __LINE__; goto error_hndl; }
|
||||
} else {
|
||||
ret = MCA_PML_CALL(irecv(accumbuf+segindex*realsegsize,
|
||||
recvcount,datatype,
|
||||
chain->chain_next[i],
|
||||
MCA_COLL_BASE_TAG_REDUCE,
|
||||
comm, &reqs[inbi]));
|
||||
if (ret != MPI_SUCCESS) { line = __LINE__; goto error_hndl; }
|
||||
}
|
||||
} /* if first segment */
|
||||
else { /* perform a irecv into the standard inbuf */
|
||||
ret = MCA_PML_CALL(irecv(inbuf[inbi],recvcount,datatype,
|
||||
chain->chain_next[i],
|
||||
@ -227,7 +222,13 @@ int mca_coll_tuned_reduce_intra_chain( void *sendbuf, void *recvbuf, int count,
|
||||
/* apply operation */
|
||||
if (1==i) {
|
||||
/* our first operation is to combine our own [sendbuf] data with the data we recvd from down stream */
|
||||
ompi_op_reduce(op, sendtmpbuf+segindex*realsegsize, accumbuf+segindex*realsegsize, recvcount, datatype );
|
||||
/* (but only if we are not root and not using MPI_IN_PLACE) */
|
||||
if ((MPI_IN_PLACE==sendbuf)&&(rank==root)) {
|
||||
ompi_op_reduce(op, inbuf[previnbi], accumbuf+segindex*realsegsize, recvcount, datatype );
|
||||
}
|
||||
else {
|
||||
ompi_op_reduce(op, sendtmpbuf+segindex*realsegsize, accumbuf+segindex*realsegsize, recvcount, datatype );
|
||||
}
|
||||
}
|
||||
else { /* not the first child, we can accumulate straight into accumbuf normally from the inbuf buffers */
|
||||
ompi_op_reduce(op, inbuf[previnbi], accumbuf+segindex*realsegsize, recvcount, datatype );
|
||||
@ -243,7 +244,13 @@ int mca_coll_tuned_reduce_intra_chain( void *sendbuf, void *recvbuf, int count,
|
||||
ompi_op_reduce(op, inbuf[previnbi], accumbuf+(segindex-1)*realsegsize, prevcount, datatype );
|
||||
}
|
||||
else { /* I have only one child, so I must combine my data (sendbuf) with the accumulated data in accumbuf */
|
||||
ompi_op_reduce(op, sendtmpbuf+(segindex-1)*realsegsize, accumbuf+(segindex-1)*realsegsize, prevcount, datatype );
|
||||
/* (but only if we are not root and not using MPI_IN_PLACE) */
|
||||
if ((MPI_IN_PLACE==sendbuf)&&(rank==root)) {
|
||||
ompi_op_reduce(op, inbuf[previnbi], accumbuf+(segindex-1)*realsegsize, prevcount, datatype );
|
||||
}
|
||||
else {
|
||||
ompi_op_reduce(op, sendtmpbuf+(segindex-1)*realsegsize, accumbuf+(segindex-1)*realsegsize, prevcount, datatype );
|
||||
}
|
||||
}
|
||||
|
||||
/* all reduced on available data this step (i) complete, pass to the next process unless your the root */
|
||||
|
||||
@ -33,7 +33,13 @@ static int pown( int fanout, int num )
|
||||
{
|
||||
int j, p = 1;
|
||||
if( num < 0 ) return 0;
|
||||
for( j = 0; j < num; j++ ) p*= fanout;
|
||||
if (1==num) return fanout;
|
||||
if (2==fanout) {
|
||||
return p<<num;
|
||||
}
|
||||
else {
|
||||
for( j = 0; j < num; j++ ) { p*= fanout; }
|
||||
}
|
||||
return p;
|
||||
}
|
||||
|
||||
|
||||
Загрузка…
Ссылка в новой задаче
Block a user