代码搜索:MPICH2
找到约 36 项符合「MPICH2」的源代码
代码结果 36
www.eeworm.com/read/169697/5412375
readme
----------------------------------------------------------------------
Mellanox has contributed substantially to the code for the ib channel.
To our original implementation based on the Infiniband se
www.eeworm.com/read/169697/5412923
c mpich2_fileutil.c
/* -*- Mode: C; c-basic-offset:4 ; -*- */
/*
* (C) 2001 by Argonne National Laboratory.
* See COPYRIGHT in top-level directory.
*/
#include "mpioimpl.h"
#include "adio_extern.h"
#ifdef MPI
www.eeworm.com/read/169697/5414334
bat sm2ps.bat
perl ..\..\..\..\doc\mpich2\sm2dot %1.tex | dot -Tps > %1.ps
www.eeworm.com/read/341021/3260987
bat sm2ps.bat
perl ..\..\..\..\doc\mpich2\sm2dot %1.tex | dot -Tps > %1.ps
www.eeworm.com/read/262717/11392789
c mat_inv.c
/*************************************************
* Parallel Computation
* for Matrix Inversion
* Using MPICH2 Library
* by Wei WANG, ISEE, ZJU
* Date: July 19, 2006 Su
www.eeworm.com/read/127438/6005097
tex opaqoverview.tex
\subsection{MPI Opaque Objects}
MPI Opaque objects such as \texttt{MPI_Comm} or \texttt{MPI_Datatype} are specified by
integers (in the MPICH2 implementation); the MPI standard calls these
han
www.eeworm.com/read/169697/5412094
txt design.txt
Goals
* Define semantics of all routines so that others can write to the interface
* Primary design is for MPICH2, but should also be useable by SMPD, etc.
* Allow implementation of passive target
www.eeworm.com/read/341021/3258386
txt design.txt
Goals
* Define semantics of all routines so that others can write to the interface
* Primary design is for MPICH2, but should also be useable by SMPD, etc.
* Allow implementation of passive target
www.eeworm.com/read/341021/3259062
sm makefile.sm
# All of these routines are internal
mpi_sources =
# tvtest builds a main program that uses the routines in dll_mpich2 to
# access the internal structure of an MPICH2 program. This is only a parti
www.eeworm.com/read/341021/3261748
testlist
noalias 2
rerr 2
# MPICH2 currently fails the bcastlength test because it needs to do more
# checking on the receive status, so for the 1.0.5 release, we are
# commenting out this test.
#bcastlength