Upgrade MPIs on linux
This commit is contained in:
Bassem Girgis
File diff suppressed because it is too large
Load Diff
Binary file not shown.
Binary file not shown.
Binary file not shown.
23
linx64/mpi/mpich/share/doc/mpich/www1/hydra_nameserver.html
Normal file
23
linx64/mpi/mpich/share/doc/mpich/www1/hydra_nameserver.html
Normal file
@@ -0,0 +1,23 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>hydra_nameserver</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="hydra_nameserver">hydra_nameserver</H1>
|
||||
program to support MPI<tt>s name publishing features with hydra
|
||||
<H2>Description</H2>
|
||||
Hydra is the default name service for MPI\</tt>s name publishing features
|
||||
(MPI_PUBLISH_NAME/MPI_LOOKUP_NAME). For programs launched by separate
|
||||
hydra instances, a separate nameserver mechanism need be launched in
|
||||
order for the name publishing to work across multiple process managers.
|
||||
<P>
|
||||
This can be achieved with hydra_nameserver. For example:
|
||||
<PRE>
|
||||
shell@myhost1$ hydra_nameserver &
|
||||
shell@myhost1$ mpiexec -hosts myhost1,myhost2 -n 4 -nameserver myhost1 ./a.out
|
||||
</PRE>
|
||||
|
||||
<P><B>Location:</B>/var/tmp/qbvJ6gPnF7/mpich-4.3.0/src/pm/hydra/mansrc/hydra_nameserver.txt<P>
|
||||
</BODY></HTML>
|
||||
16
linx64/mpi/mpich/share/doc/mpich/www1/hydra_persist.html
Normal file
16
linx64/mpi/mpich/share/doc/mpich/www1/hydra_persist.html
Normal file
@@ -0,0 +1,16 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>hydra_persist</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="hydra_persist">hydra_persist</H1>
|
||||
Internal executable used by Hydra
|
||||
<H2>Description</H2>
|
||||
This executable is designed to not be run directly by users, but to
|
||||
support the Hydra process manager. As such, no documentation will be
|
||||
provided here.
|
||||
<P>
|
||||
<P><B>Location:</B>/var/tmp/qbvJ6gPnF7/mpich-4.3.0/src/pm/hydra/mansrc/hydra_persist.txt<P>
|
||||
</BODY></HTML>
|
||||
16
linx64/mpi/mpich/share/doc/mpich/www1/hydra_pmi_proxy.html
Normal file
16
linx64/mpi/mpich/share/doc/mpich/www1/hydra_pmi_proxy.html
Normal file
@@ -0,0 +1,16 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>hydra_pmi_proxy</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="hydra_pmi_proxy">hydra_pmi_proxy</H1>
|
||||
Internal executable used by Hydra
|
||||
<H2>Description</H2>
|
||||
This executable is designed to not be run directly by users, but to
|
||||
support the Hydra process manager. As such, no documentation will be
|
||||
provided here.
|
||||
<P>
|
||||
<P><B>Location:</B>/var/tmp/qbvJ6gPnF7/mpich-4.3.0/src/pm/hydra/mansrc/hydra_pmi_proxy.txt<P>
|
||||
</BODY></HTML>
|
||||
@@ -1,17 +1,21 @@
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<TITLE>Manpages for MPICH</TITLE>
|
||||
<!-- This file generated by createhtmlindex on -->
|
||||
<!-- This file generated by createhtmlindex on 2/3/2025 -->
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1>Manpages for MPICH</H1>
|
||||
<TABLE>
|
||||
<TR><TD><A HREF="mpicc.html">mpicc</A></TD>
|
||||
<TD><A HREF="mpiexec.html">mpiexec</A></TD>
|
||||
<TR><TD><A HREF="hydra_nameserver.html">hydra_nameserver</A></TD>
|
||||
<TD><A HREF="mpicc.html">mpicc</A></TD>
|
||||
<TD><A HREF="mpifort.html">mpifort</A></TD>
|
||||
</TR>
|
||||
<TR><TD><A HREF="mpicxx.html">mpicxx</A></TD>
|
||||
<TD><A HREF="mpif77.html">mpif77</A></TD>
|
||||
<TR><TD><A HREF="hydra_persist.html">hydra_persist</A></TD>
|
||||
<TD><A HREF="mpicxx.html">mpicxx</A></TD>
|
||||
<TD></TD>
|
||||
</TR>
|
||||
<TR><TD><A HREF="hydra_pmi_proxy.html">hydra_pmi_proxy</A></TD>
|
||||
<TD><A HREF="mpiexec.html">mpiexec</A></TD>
|
||||
<TD></TD>
|
||||
</TR>
|
||||
</TABLE>
|
||||
|
||||
@@ -5,7 +5,7 @@
|
||||
<TITLE>mpicc</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="mpicc"><H1>mpicc</H1></A>
|
||||
<H1 id="mpicc">mpicc</H1>
|
||||
Compiles and links MPI programs written in C
|
||||
<H2>Description</H2>
|
||||
This command can be used to compile and link MPI programs written in
|
||||
@@ -17,67 +17,69 @@ as it provides the necessary libraries.
|
||||
<P>
|
||||
<H2>Command line arguments</H2>
|
||||
<DL>
|
||||
<DT><B>-show </B><DD>Show the commands that would be used without
|
||||
<DT><B>-show </B> <DD> Show the commands that would be used without
|
||||
running them
|
||||
|
||||
<DT><B>-help </B><DD>Give short help
|
||||
<DT><B>-help </B> <DD> Give short help
|
||||
|
||||
<DT><B>-cc=name </B><DD>Use compiler <TT>name</TT> instead of the default choice. Use
|
||||
<DT><B>-cc=name </B> <DD> Use compiler <tt>name</tt> instead of the default choice. Use
|
||||
this only if the compiler is compatible with the MPICH
|
||||
library (see below)
|
||||
|
||||
<DT><B>-config=name </B><DD>Load a configuration file for a particular compiler.
|
||||
This allows a single <TT>mpicc</TT> command to be used with
|
||||
<DT><B>-config=name </B> <DD> Load a configuration file for a particular compiler.
|
||||
This allows a single <tt>mpicc</tt> command to be used with
|
||||
multiple compilers.
|
||||
|
||||
<DT><B>-compile_info </B><DD>Show the steps for compiling a program. This option
|
||||
<DT><B>-compile_info </B> <DD> Show the steps for compiling a program. This option
|
||||
can be used to see what options and include paths are
|
||||
used by mpicc.
|
||||
|
||||
<DT><B>-link_info </B><DD>Show the steps for linking a program. This option
|
||||
<DT><B>-link_info </B> <DD> Show the steps for linking a program. This option
|
||||
can be used to see what options and libraries are used by
|
||||
mpicc.
|
||||
|
||||
<DT><B>-profile=name </B><DD>Use the MPI profiling given by name. See below for
|
||||
<DT><B>-profile=name </B> <DD> Use the MPI profiling given by name. See below for
|
||||
details
|
||||
|
||||
<DT><B>-echo </B><DD>Show exactly what this program is doing.
|
||||
<DT><B>-echo </B> <DD> Show exactly what this program is doing.
|
||||
This option should normally not be used.
|
||||
|
||||
<DT><B>-static</B><DD>mpi - Use a statically compile MPI library, but shared libraries
|
||||
<DT><B>-static</B> <DD> mpi - Use a statically compile MPI library, but shared libraries
|
||||
for all of the other dependencies.
|
||||
|
||||
<DT><B>others </B><DD>are passed to the compiler or linker. For example, <TT>\-c
|
||||
</TT>causes files to be compiled, <TT>\-g</TT> selects compilation with
|
||||
debugging on most systems, and <TT>\-o name</TT> causes linking
|
||||
with the output executable given the name <TT>name</TT>.
|
||||
<DT><B>-mpi</B> <DD> abi - Compile and link using the standard MPI ABI.
|
||||
|
||||
<DT><B>others </B> <DD> are passed to the compiler or linker. For example, <tt>\-c
|
||||
</tt>causes files to be compiled, <tt>\-g</tt> selects compilation with
|
||||
debugging on most systems, and <tt>\-o name</tt> causes linking
|
||||
with the output executable given the name <tt>name</tt>.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Environment Variables</H2>
|
||||
The environment variable <TT>MPICH_CC</TT> may be used
|
||||
The environment variable <tt>MPICH_CC</tt> may be used
|
||||
to select different C compiler and linker. Note that since
|
||||
MPICH is built with a particular C and Fortran compiler, changing the
|
||||
compilers used can cause problems. Use this only if you could intermix
|
||||
code compiled with the different compilers.
|
||||
<P>
|
||||
The environment variable <TT>MPICC_PROFILE</TT> specifies a profile library
|
||||
and has the same effect as if <TT>\-profile=$MPICC_PROFILE</TT> were used as
|
||||
an argument to <TT>mpicc</TT>. See the discussion of <TT>\-profile</TT> below for more
|
||||
The environment variable <tt>MPICC_PROFILE</tt> specifies a profile library
|
||||
and has the same effect as if <tt>\-profile=$MPICC_PROFILE</tt> were used as
|
||||
an argument to <tt>mpicc</tt>. See the discussion of <tt>\-profile</tt> below for more
|
||||
details.
|
||||
<P>
|
||||
<H2>Compatible Compilers</H2>
|
||||
The MPI library may be used with any compiler that uses the same
|
||||
lengths for basic data objects (such as <TT>long double</TT>) and that
|
||||
lengths for basic data objects (such as <tt>long double</tt>) and that
|
||||
uses compatible run-time libraries. On many systems, the various
|
||||
compilers are compatible and may be used interchangably. There are
|
||||
exceptions; if you use the <TT>MPICH_CC</TT> environment variable or the
|
||||
<TT>\-cc=name</TT> command-line argument to override the choice of compiler
|
||||
compilers are compatible and may be used interchangeably. There are
|
||||
exceptions; if you use the <tt>MPICH_CC</tt> environment variable or the
|
||||
<tt>\-cc=name</tt> command-line argument to override the choice of compiler
|
||||
and encounter problems, try reconfiguring MPICH with the new compiler
|
||||
and installing MPICH in a separate location. See the installation manual
|
||||
for more details.
|
||||
<P>
|
||||
<H2>Examples</H2>
|
||||
To compile a single file <TT>foo.c</TT>, use
|
||||
To compile a single file <tt>foo.c</tt>, use
|
||||
<PRE>
|
||||
mpicc -c foo.c
|
||||
</PRE>
|
||||
@@ -96,29 +98,29 @@ Combining compilation and linking in a single command
|
||||
is a convenient way to build simple programs.
|
||||
<P>
|
||||
<H2>Selecting a Profiling Library</H2>
|
||||
The <TT>\-profile=name</TT> argument allows you to specify an MPI profiling
|
||||
library to be used. <TT>name</TT> can have two forms:
|
||||
The <tt>\-profile=name</tt> argument allows you to specify an MPI profiling
|
||||
library to be used. <tt>name</tt> can have two forms:
|
||||
<P>
|
||||
<BR>A library in the same directory as the MPI library
|
||||
<BR>The name of a profile configuration file
|
||||
<BR>
|
||||
<br>A library in the same directory as the MPI library
|
||||
<br>The name of a profile configuration file
|
||||
<br>
|
||||
<P>
|
||||
If <TT>name</TT> is a library, then this library is included before the MPI
|
||||
If <tt>name</tt> is a library, then this library is included before the MPI
|
||||
library. This allows the simple use of libraries that make use of the
|
||||
MPI profiling interface and that are installed in the same directory as
|
||||
the MPI library.
|
||||
<P>
|
||||
If <TT>name.conf</TT> is the name of a file in the sysconfdir directory, then this
|
||||
If <tt>name.conf</tt> is the name of a file in the sysconfdir directory, then this
|
||||
is read and may define the following variables:
|
||||
<DL>
|
||||
<DT><B>PROFILE_PRELIB </B><DD>Libraries (and paths) to include before the MPI library
|
||||
<DT><B>PROFILE_PRELIB </B> <DD> Libraries (and paths) to include before the MPI library
|
||||
|
||||
<DT><B>PROFILE_POSTLIB </B><DD>Libraries to include after the MPI library
|
||||
<DT><B>PROFILE_POSTLIB </B> <DD> Libraries to include after the MPI library
|
||||
|
||||
<DT><B>PROFILE_INCPATHS </B><DD>C preprocessor arguments for any include files
|
||||
For example, to add <TT>/usr/local/myprof/include</TT> to the include path and
|
||||
the library <TT>libmyprof.a</TT> in <TT>/usr/local/myprof/lib</TT> to the link step,
|
||||
you could create the file <TT>myprof.conf</TT> with the lines
|
||||
<DT><B>PROFILE_INCPATHS </B> <DD> C preprocessor arguments for any include files
|
||||
For example, to add <tt>/usr/local/myprof/include</tt> to the include path and
|
||||
the library <tt>libmyprof.a</tt> in <tt>/usr/local/myprof/lib</tt> to the link step,
|
||||
you could create the file <tt>myprof.conf</tt> with the lines
|
||||
</DL>
|
||||
<P>
|
||||
<PRE>
|
||||
@@ -128,10 +130,10 @@ you could create the file <TT>myprof.conf</TT> with the lines
|
||||
|
||||
and place it in the sysconfdir directory (this directory is set at
|
||||
configure time when MPICH is built). Then using the command-line
|
||||
argument <TT>\-profile=myprof</TT> will cause these
|
||||
argument <tt>\-profile=myprof</tt> will cause these
|
||||
definitions to be added to the relevant compile commands.
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
mpicxx, mpifort, mpiexec
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,7 +5,7 @@
|
||||
<TITLE>mpicxx</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="mpicxx"><H1>mpicxx</H1></A>
|
||||
<H1 id="mpicxx">mpicxx</H1>
|
||||
Compiles and links MPI programs written in C++
|
||||
<H2>Description</H2>
|
||||
This command can be used to compile and link MPI programs written in
|
||||
@@ -17,67 +17,69 @@ as it provides the necessary libraries.
|
||||
<P>
|
||||
<H2>Command line arguments</H2>
|
||||
<DL>
|
||||
<DT><B>-show </B><DD>Show the commands that would be used without
|
||||
<DT><B>-show </B> <DD> Show the commands that would be used without
|
||||
running them
|
||||
|
||||
<DT><B>-help </B><DD>Give short help
|
||||
<DT><B>-help </B> <DD> Give short help
|
||||
|
||||
<DT><B>-cxx=name </B><DD>Use compiler <TT>name</TT> instead of the default choice. Use
|
||||
<DT><B>-cxx=name </B> <DD> Use compiler <tt>name</tt> instead of the default choice. Use
|
||||
this only if the compiler is compatible with the MPICH
|
||||
library (see below)
|
||||
|
||||
<DT><B>-config=name </B><DD>Load a configuration file for a particular compiler.
|
||||
This allows a single <TT>mpicxx</TT> command to be used with
|
||||
<DT><B>-config=name </B> <DD> Load a configuration file for a particular compiler.
|
||||
This allows a single <tt>mpicxx</tt> command to be used with
|
||||
multiple compilers.
|
||||
|
||||
<DT><B>-compile_info </B><DD>Show the steps for compiling a program. This option
|
||||
<DT><B>-compile_info </B> <DD> Show the steps for compiling a program. This option
|
||||
can be used to see what options and include paths are
|
||||
used by mpicxx.
|
||||
|
||||
<DT><B>-link_info </B><DD>Show the steps for linking a program. This option
|
||||
<DT><B>-link_info </B> <DD> Show the steps for linking a program. This option
|
||||
can be used to see what options and libraries are used by
|
||||
mpicxx.
|
||||
|
||||
<DT><B>-profile=name </B><DD>Use the MPI profiling given by name. See below for
|
||||
<DT><B>-profile=name </B> <DD> Use the MPI profiling given by name. See below for
|
||||
details
|
||||
|
||||
<DT><B>-echo </B><DD>Show exactly what this program is doing.
|
||||
<DT><B>-echo </B> <DD> Show exactly what this program is doing.
|
||||
This option should normally not be used.
|
||||
|
||||
<DT><B>-static</B><DD>mpi - Use a statically compile MPI library, but shared libraries
|
||||
<DT><B>-static</B> <DD> mpi - Use a statically compile MPI library, but shared libraries
|
||||
for all of the other dependencies.
|
||||
|
||||
<DT><B>others </B><DD>are passed to the compiler or linker. For example, <TT>\-c
|
||||
</TT>causes files to be compiled, <TT>\-g</TT> selects compilation with
|
||||
debugging on most systems, and <TT>\-o name</TT> causes linking
|
||||
with the output executable given the name <TT>name</TT>.
|
||||
<DT><B>-mpi</B> <DD> abi - Compile and link using the standard MPI ABI.
|
||||
|
||||
<DT><B>others </B> <DD> are passed to the compiler or linker. For example, <tt>\-c
|
||||
</tt>causes files to be compiled, <tt>\-g</tt> selects compilation with
|
||||
debugging on most systems, and <tt>\-o name</tt> causes linking
|
||||
with the output executable given the name <tt>name</tt>.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Environment Variables</H2>
|
||||
The environment variables <TT>MPICH_CXX</TT> may be used
|
||||
The environment variables <tt>MPICH_CXX</tt> may be used
|
||||
to select different C++ compiler and linker. Note that since
|
||||
MPICH is built with a particular C and Fortran compiler, changing the
|
||||
compilers used can cause problems. Use this only if you could intermix
|
||||
code compiled with the different compilers.
|
||||
<P>
|
||||
The environment variable <TT>MPICC_PROFILE</TT> specifies a profile library
|
||||
and has the same effect as if <TT>\-profile=$MPICC_PROFILE</TT> were used as
|
||||
an argument to <TT>mpicc</TT>. See the discussion of <TT>\-profile</TT> below for more
|
||||
The environment variable <tt>MPICC_PROFILE</tt> specifies a profile library
|
||||
and has the same effect as if <tt>\-profile=$MPICC_PROFILE</tt> were used as
|
||||
an argument to <tt>mpicc</tt>. See the discussion of <tt>\-profile</tt> below for more
|
||||
details.
|
||||
<P>
|
||||
<H2>Compatible Compilers</H2>
|
||||
The MPI library may be used with any compiler that uses the same
|
||||
lengths for basic data objects (such as <TT>long double</TT>) and that
|
||||
lengths for basic data objects (such as <tt>long double</tt>) and that
|
||||
uses compatible run-time libraries. On many systems, the various
|
||||
compilers are compatible and may be used interchangably. There are
|
||||
exceptions; if you use the <TT>MPICH_CXX</TT> environment variable or the
|
||||
<TT>\-cxx=name</TT> command-line argument to override the choice of compiler
|
||||
compilers are compatible and may be used interchangeably. There are
|
||||
exceptions; if you use the <tt>MPICH_CXX</tt> environment variable or the
|
||||
<tt>\-cxx=name</tt> command-line argument to override the choice of compiler
|
||||
and encounter problems, try reconfiguring MPICH with the new compiler,
|
||||
and installing MPICH in a separate location. See the installation manual
|
||||
for more details.
|
||||
<P>
|
||||
<H2>Examples</H2>
|
||||
To compile a single file <TT>foo.c</TT>, use
|
||||
To compile a single file <tt>foo.c</tt>, use
|
||||
<PRE>
|
||||
mpicxx -c foo.cxx
|
||||
</PRE>
|
||||
@@ -96,29 +98,29 @@ Combining compilation and linking in a single command
|
||||
is a convenient way to build simple programs.
|
||||
<P>
|
||||
<H2>Selecting a Profiling Library</H2>
|
||||
The <TT>\-profile=name</TT> argument allows you to specify an MPI profiling
|
||||
library to be used. <TT>name</TT> can have two forms:
|
||||
The <tt>\-profile=name</tt> argument allows you to specify an MPI profiling
|
||||
library to be used. <tt>name</tt> can have two forms:
|
||||
<P>
|
||||
<BR>A library in the same directory as the MPI library
|
||||
<BR>The name of a profile configuration file
|
||||
<BR>
|
||||
<br>A library in the same directory as the MPI library
|
||||
<br>The name of a profile configuration file
|
||||
<br>
|
||||
<P>
|
||||
If <TT>name</TT> is a library, then this library is included before the MPI
|
||||
If <tt>name</tt> is a library, then this library is included before the MPI
|
||||
library. This allows the simple use of libraries that make use of the
|
||||
MPI profiling interface and that are installed in the same directory as
|
||||
the MPI library.
|
||||
<P>
|
||||
If <TT>name.conf</TT> is the name of a file in the sysconfdir directory, then this
|
||||
If <tt>name.conf</tt> is the name of a file in the sysconfdir directory, then this
|
||||
is read and may define the following variables:
|
||||
<DL>
|
||||
<DT><B>PROFILE_PRELIB </B><DD>Libraries (and paths) to include before the MPI library
|
||||
<DT><B>PROFILE_PRELIB </B> <DD> Libraries (and paths) to include before the MPI library
|
||||
|
||||
<DT><B>PROFILE_POSTLIB </B><DD>Libraries to include after the MPI library
|
||||
<DT><B>PROFILE_POSTLIB </B> <DD> Libraries to include after the MPI library
|
||||
|
||||
<DT><B>PROFILE_INCPATHS </B><DD>C preprocessor arguments for any include files
|
||||
For example, to add <TT>/usr/local/myprof/include</TT> to the include path and
|
||||
the library <TT>libmyprof.a</TT> in <TT>/usr/local/myprof/lib</TT> to the link step,
|
||||
you could create the file <TT>myprof.conf</TT> with the lines
|
||||
<DT><B>PROFILE_INCPATHS </B> <DD> C preprocessor arguments for any include files
|
||||
For example, to add <tt>/usr/local/myprof/include</tt> to the include path and
|
||||
the library <tt>libmyprof.a</tt> in <tt>/usr/local/myprof/lib</tt> to the link step,
|
||||
you could create the file <tt>myprof.conf</tt> with the lines
|
||||
</DL>
|
||||
<P>
|
||||
<PRE>
|
||||
@@ -128,10 +130,10 @@ you could create the file <TT>myprof.conf</TT> with the lines
|
||||
|
||||
and place it in the sysconfdir directory (this directory is set at
|
||||
configure time when MPICH is built). Then using the command-line
|
||||
argument <TT>\-profile=myprof</TT> will cause these
|
||||
argument <tt>\-profile=myprof</tt> will cause these
|
||||
definitions to be added to the relevant compile commands.
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
mpicc, mpifort, mpiexec
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,7 +5,7 @@
|
||||
<TITLE>mpiexec</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="mpiexec"><H1>mpiexec</H1></A>
|
||||
<H1 id="mpiexec">mpiexec</H1>
|
||||
Run an MPI program
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
@@ -14,52 +14,52 @@ Run an MPI program
|
||||
mpiexec args executable pgmargs [ : args executable pgmargs ... ]
|
||||
</PRE>
|
||||
|
||||
where <TT>args</TT> are command line arguments for <TT>mpiexec</TT> (see below),
|
||||
<TT>executable</TT> is the name of an executable MPI program, and <TT>pgmargs
|
||||
</TT>are command line arguments for the executable. Multiple executables
|
||||
where <tt>args</tt> are command line arguments for <tt>mpiexec</tt> (see below),
|
||||
<tt>executable</tt> is the name of an executable MPI program, and <tt>pgmargs
|
||||
</tt>are command line arguments for the executable. Multiple executables
|
||||
can be specified by using the colon notation (for MPMD - Multiple Program
|
||||
Multiple Data applications). For example, the following command will run
|
||||
the MPI program <TT>a.out</TT> on 4 processes:
|
||||
the MPI program <tt>a.out</tt> on 4 processes:
|
||||
<PRE>
|
||||
mpiexec -n 4 a.out
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
The MPI standard specifies the following arguments and their meanings:
|
||||
The MPI standard suggests the following arguments and their meanings:
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>-n <np> </B><DD>Specify the number of processes to use
|
||||
<DT><B>-n <np> </B> <DD> Specify the number of processes to use
|
||||
|
||||
<DT><B>-host <hostname> </B><DD>Name of host on which to run processes
|
||||
<DT><B>-host <hostname> </B> <DD> Name of host on which to run processes
|
||||
|
||||
<DT><B>-arch <architecture name> </B><DD>Pick hosts with this architecture type
|
||||
<DT><B>-wdir <working directory> </B> <DD> cd to this one <em>before</em> running executable
|
||||
|
||||
<DT><B>-wdir <working directory> </B><DD>cd to this one <EM>before</EM> running executable
|
||||
<DT><B>-configfile <name> </B> <DD> file containing specifications of host/program,
|
||||
one per line, with # as a comment indicator, e.g., the usual
|
||||
mpiexec input. The lines excluding comments are concatenated with newlines
|
||||
replaced with spaces, and are then parsed as command line options.
|
||||
</DL>
|
||||
<P>
|
||||
The following options are reserved by the MPI standard, but they are not
|
||||
supported by MPICH (hydra):
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>-arch <architecture name> </B> <DD> Pick hosts with this architecture type
|
||||
|
||||
<DT><B>-path <pathlist> </B><DD>use this to find the executable
|
||||
<DT><B>-path <pathlist> </B> <DD> use this to find the executable
|
||||
|
||||
<DT><B>-soft <triplets> </B><DD>comma separated triplets that specify requested numbers
|
||||
<DT><B>-soft <triplets> </B> <DD> comma separated triplets that specify requested numbers
|
||||
of processes (see the MPI-2 specification for more details)
|
||||
|
||||
<DT><B>-file <name> </B><DD>implementation-defined specification file
|
||||
|
||||
<DT><B>-configfile <name> </B><DD>file containing specifications of host/program,
|
||||
one per line, with # as a comment indicator, e.g., the usual
|
||||
mpiexec input, but with ":" replaced with a newline. That is,
|
||||
the configfile contains lines with -soft, -n etc.
|
||||
<DT><B>-file <name> </B> <DD> implementation-defined specification file
|
||||
</DL>
|
||||
<P>
|
||||
Additional arguments that are specific to the MPICH implementation
|
||||
are discussed below.
|
||||
<P>
|
||||
Note that not all of these parameters are meaningful for all
|
||||
systems. For example, the <TT>gforker</TT> version of <TT>mpiexec</TT> creates all
|
||||
processes on the same system on which it is running; in that case, the
|
||||
<TT>\-arch</TT> and <TT>\-host</TT> options are ignored.
|
||||
<P>
|
||||
The colon character (<TT>:</TT>) may be used to separate different executables
|
||||
The colon character (<tt>:</tt>) may be used to separate different executables
|
||||
for MPMD (multiple program multiple data) programming. For example,
|
||||
to run the program <TT>ocean</TT> on 4 processes and <TT>air</TT> on 8 processes, use:
|
||||
to run the program <tt>ocean</tt> on 4 processes and <tt>air</tt> on 8 processes, use:
|
||||
<P>
|
||||
<PRE>
|
||||
mpiexec -n 4 ocean : -n 8 air
|
||||
@@ -68,85 +68,81 @@ to run the program <TT>ocean</TT> on 4 processes and <TT>air</TT> on 8 processes
|
||||
<P>
|
||||
<P>
|
||||
<H2>MPICH-Specific Arguments</H2>
|
||||
<P>
|
||||
MPICH implements a few process managers. Here we list a general set of
|
||||
arguments that many of the implementations support. More detailed options
|
||||
that are supported by a particular implementation, e.g. hydra, are
|
||||
available via:
|
||||
<PRE>
|
||||
mpiexec -h
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
Many of the implementations of process managers in MPICH support the
|
||||
following arguments to <TT>mpiexec</TT>:
|
||||
following arguments to <tt>mpiexec</tt>:
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>-np <num> </B><DD>A synonym for the standard <TT>\-n</TT> argument
|
||||
<DT><B>-np <num> </B> <DD> A synonym for the standard <tt>\-n</tt> argument
|
||||
|
||||
<DT><B>-env <name> <value> </B><DD>Set the environment variable <TT><name></TT> to <TT><value></TT> for
|
||||
the processes being run by <TT>mpiexec
|
||||
</TT>
|
||||
<DT><B>-envnone </B><DD>Pass no environment variables (other than ones specified with
|
||||
other <TT>\-env</TT> or <TT>\-genv</TT> arguments) to the processes being run by <TT>mpiexec</TT>.
|
||||
<DT><B>-env <name> <value> </B> <DD> Set the environment variable <tt><name></tt> to <tt><value></tt> for
|
||||
the processes being run by <tt>mpiexec
|
||||
</tt>
|
||||
<DT><B>-envnone </B> <DD> Pass no environment variables (other than ones specified with
|
||||
other <tt>\-env</tt> or <tt>\-genv</tt> arguments) to the processes being run by <tt>mpiexec</tt>.
|
||||
By default, all environment
|
||||
variables are provided to each MPI process (rationale: principle of
|
||||
least surprise for the user)
|
||||
|
||||
<DT><B>-envlist <list> </B><DD>Pass the listed environment variables (names separated
|
||||
<DT><B>-envlist <list> </B> <DD> Pass the listed environment variables (names separated
|
||||
by commas), with their current values, to the processes being run by
|
||||
<TT>mpiexec</TT>.
|
||||
<tt>mpiexec</tt>.
|
||||
|
||||
<DT><B>-genv <name> <value> </B><DD>The <TT>\-genv</TT> options have the same meaning as their
|
||||
corresponding <TT>\-env</TT> version, except they apply to all executables, not just
|
||||
<DT><B>-genv <name> <value> </B> <DD> The <tt>\-genv</tt> options have the same meaning as their
|
||||
corresponding <tt>\-env</tt> version, except they apply to all executables, not just
|
||||
the current executable (in the case that the colon syntax is used to specify
|
||||
multiple execuables).
|
||||
multiple executables).
|
||||
|
||||
<DT><B>-genvnone </B><DD>Like <TT>\-envnone</TT>, but for all executables
|
||||
<DT><B>-genvnone </B> <DD> Like <tt>\-envnone</tt>, but for all executables
|
||||
|
||||
<DT><B>-genvlist <list> </B><DD>Like <TT>\-envlist</TT>, but for all executables
|
||||
<DT><B>-genvlist <list> </B> <DD> Like <tt>\-envlist</tt>, but for all executables
|
||||
|
||||
<DT><B>-usize <n> </B><DD>Specify the value returned for the value of the attribute
|
||||
<TT>MPI_UNIVERSE_SIZE</TT>.
|
||||
<DT><B>-usize <n> </B> <DD> Specify the value returned for the value of the attribute
|
||||
<tt>MPI_UNIVERSE_SIZE</tt>.
|
||||
|
||||
<DT><B>-l </B><DD>Label standard out and standard error (<TT>stdout</TT> and <TT>stderr</TT>) with
|
||||
<DT><B>-l </B> <DD> Label standard out and standard error (<tt>stdout</tt> and <tt>stderr</tt>) with
|
||||
the rank of the process
|
||||
|
||||
<DT><B>-maxtime <n> </B><DD>Set a timelimit of <TT><n></TT> seconds.
|
||||
<DT><B>-maxtime <n> </B> <DD> Set a timelimit of <tt><n></tt> seconds.
|
||||
|
||||
<DT><B>-exitinfo </B><DD>Provide more information on the reason each process exited if
|
||||
<DT><B>-exitinfo </B> <DD> Provide more information on the reason each process exited if
|
||||
there is an abnormal exit
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Environment variables for mpiexec</H2>
|
||||
The following environment variables are understood by some versions of
|
||||
<TT>mpiexec</TT>. The command line arguments have priority over these; that is,
|
||||
<tt>mpiexec</tt>. The command line arguments have priority over these; that is,
|
||||
if both the environment variable and command line argument are used, the
|
||||
value specified by the command line argument is used.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPIEXEC_TIMEOUT </B><DD>Maximum running time in seconds. <TT>mpiexec</TT> will
|
||||
<DT><B>MPIEXEC_TIMEOUT </B> <DD> Maximum running time in seconds. <tt>mpiexec</tt> will
|
||||
terminate MPI programs that take longer than the value specified by
|
||||
<TT>MPIEXEC_TIMEOUT</TT>.
|
||||
<tt>MPIEXEC_TIMEOUT</tt>.
|
||||
|
||||
<DT><B>MPIEXEC_UNIVERSE_SIZE </B><DD>Set the universe size
|
||||
<DT><B>MPIEXEC_UNIVERSE_SIZE </B> <DD> Set the universe size
|
||||
|
||||
<DT><B>MPIEXEC_PORT_RANGE </B><DD>Set the range of ports that <TT>mpiexec</TT> will use
|
||||
<DT><B>MPIEXEC_PORT_RANGE </B> <DD> Set the range of ports that <tt>mpiexec</tt> will use
|
||||
in communicating with the processes that it starts. The format of
|
||||
this is <TT><low>:<high></TT>. For example, to specify any port between
|
||||
10000 and 10100, use <TT>10000:10100</TT>.
|
||||
this is <tt><low>:<high></tt>. For example, to specify any port between
|
||||
10000 and 10100, use <tt>10000:10100</tt>.
|
||||
|
||||
<DT><B>MPICH_PORT_RANGE </B><DD>Has the same meaning as <TT>MPIEXEC_PORT_RANGE</TT> and
|
||||
is used if <TT>MPIEXEC_PORT_RANGE</TT> is not set.
|
||||
<DT><B>MPICH_PORT_RANGE </B> <DD> Has the same meaning as <tt>MPIEXEC_PORT_RANGE</tt> and
|
||||
is used if <tt>MPIEXEC_PORT_RANGE</tt> is not set.
|
||||
|
||||
<DT><B>MPIEXEC_PREFIX_DEFAULT </B><DD>If this environment variable is set, output
|
||||
to standard output is prefixed by the rank in <TT>MPI_COMM_WORLD</TT> of the
|
||||
process and output to standard error is prefixed by the rank and the
|
||||
text <TT>(err)</TT>; both are followed by an angle bracket (<TT>></TT>). If
|
||||
this variable is not set, there is no prefix.
|
||||
|
||||
<DT><B>MPIEXEC_PREFIX_STDOUT </B><DD>Set the prefix used for lines sent to standard
|
||||
output. A <TT>%d</TT> is replaced with the rank in <TT>MPI_COMM_WORLD</TT>; a <TT>%w</TT> is
|
||||
replaced with an indication of which <TT>MPI_COMM_WORLD</TT> in MPI jobs that
|
||||
involve multiple <TT>MPI_COMM_WORLD</TT>s (e.g., ones that use <TT>MPI_Comm_spawn</TT> or
|
||||
<TT>MPI_Comm_connect</TT>).
|
||||
|
||||
<DT><B>MPIEXEC_PREFIX_STDERR </B><DD>Like <TT>MPIEXEC_PREFIX_STDOUT</TT>, but for standard error.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Status</H2>
|
||||
<TT>mpiexec</TT> returns the maximum of the exit status values of all of the
|
||||
processes created by <TT>mpiexec</TT>.
|
||||
<tt>mpiexec</tt> returns the maximum of the exit status values of all of the
|
||||
processes created by <tt>mpiexec</tt>.
|
||||
<P>
|
||||
<P><B>Location:</B>/var/tmp/qbvJ6gPnF7/mpich-4.3.0/src/pm/hydra/mansrc/mpiexec.txt<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -1,132 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>mpif77</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="mpif77"><H1>mpif77</H1></A>
|
||||
Compiles and links MPI programs written in Fortran 77
|
||||
<H2>Description</H2>
|
||||
This command can be used to compile and link MPI programs written in
|
||||
Fortran. It provides the options and any special libraries that are
|
||||
needed to compile and link MPI programs.
|
||||
<P>
|
||||
It is important to use this command, particularly when linking programs,
|
||||
as it provides the necessary libraries.
|
||||
<P>
|
||||
<H2>Command line arguments</H2>
|
||||
<DL>
|
||||
<DT><B>-show </B><DD>Show the commands that would be used without
|
||||
running them
|
||||
|
||||
<DT><B>-help </B><DD>Give short help
|
||||
|
||||
<DT><B>-f77=name </B><DD>Use compiler <TT>name</TT> instead of the default choice. Use
|
||||
this only if the compiler is compatible with the MPICH
|
||||
library (see below)
|
||||
|
||||
<DT><B>-config=name </B><DD>Load a configuration file for a particular compiler.
|
||||
This allows a single <TT>mpif77</TT> command to be used with
|
||||
multiple compilers.
|
||||
|
||||
<DT><B>-compile_info </B><DD>Show the steps for compiling a program. This option
|
||||
can be used to see what options and include paths are
|
||||
used by mpif77.
|
||||
|
||||
<DT><B>-link_info </B><DD>Show the steps for linking a program. This option
|
||||
can be used to see what options and libraries are used by
|
||||
mpif77.
|
||||
|
||||
<DT><B>-profile=name </B><DD>Use the MPI profiling given by name. See below for
|
||||
details
|
||||
|
||||
<DT><B>-echo </B><DD>Show exactly what this program is doing.
|
||||
This option should normally not be used.
|
||||
|
||||
<DT><B>-static</B><DD>mpi - Use a statically compile MPI library, but shared libraries
|
||||
for all of the other dependencies.
|
||||
|
||||
<DT><B>others </B><DD>are passed to the compiler or linker. For example, <TT>\-c
|
||||
</TT>causes files to be compiled, <TT>\-g</TT> selects compilation with
|
||||
debugging on most systems, and <TT>\-o name</TT> causes linking
|
||||
with the output executable given the name <TT>name</TT>.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Environment Variables</H2>
|
||||
The environment variables <TT>MPICH_F77</TT> may be used
|
||||
to select different Fortran compiler and linker. Note that since
|
||||
MPICH is built with a particular C and Fortran compiler, change the
|
||||
compilers used can cause problems. Use this only if you could intermix
|
||||
code compiled with the different compilers.
|
||||
<P>
|
||||
<H2>Compatible Compilers</H2>
|
||||
The MPI library may be used with any compiler that uses the same
|
||||
lengths for basic data objects (such as <TT>long double</TT>) and that
|
||||
uses compatible run-time libraries. On many systems, the various
|
||||
compilers are compatible and may be used interchangably. There are
|
||||
exceptions; if you use the <TT>MPICH_F77</TT> environment variable or the
|
||||
<TT>\-f77=name</TT> command-line argument to override the choice of compiler
|
||||
and encounter problems, try reconfiguring MPICH with the new compiler
|
||||
and installing MPICH in a separate location. See the installation manual
|
||||
for more details.
|
||||
<P>
|
||||
<H2>Examples</H2>
|
||||
To compile a single file <TT>foo.f</TT>, use
|
||||
<PRE>
|
||||
mpif77 -c foo.f
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
To link the output and make an executable, use
|
||||
<PRE>
|
||||
mpif77 -o foo foo.o
|
||||
</PRE>
|
||||
|
||||
Combining compilation and linking in a single command
|
||||
<PRE>
|
||||
mpif77 -o foo foo.f
|
||||
</PRE>
|
||||
|
||||
is a convenient way to build simple programs.
|
||||
<P>
|
||||
<H2>Selecting a Profiling Library</H2>
|
||||
The <TT>\-profile=name</TT> argument allows you to specify an MPI profiling
|
||||
library to be used. <TT>name</TT> can have two forms:
|
||||
<P>
|
||||
<BR>A library in the same directory as the MPI library
|
||||
<BR>The name of a profile configuration file
|
||||
<BR>
|
||||
<P>
|
||||
If <TT>name</TT> is a library, then this library is included before the MPI
|
||||
library. This allows the simple use of libraries that make use of the
|
||||
MPI profiling interface and that are installed in the same directory as
|
||||
the MPI library.
|
||||
<P>
|
||||
If <TT>name.conf</TT> is the name of a file in the sysconfdir directory, then this
|
||||
is read and may define the following variables:
|
||||
<DL>
|
||||
<DT><B>PROFILE_PRELIB </B><DD>Libraries (and paths) to include before the MPI library
|
||||
|
||||
<DT><B>PROFILE_POSTLIB </B><DD>Libraries to include after the MPI library
|
||||
|
||||
<DT><B>PROFILE_INCPATHS </B><DD>C preprocessor arguments for any include files
|
||||
For example, to add <TT>/usr/local/myprof/include</TT> to the include path and
|
||||
the library <TT>libmyprof.a</TT> in <TT>/usr/local/myprof/lib</TT> to the link step,
|
||||
you could create the file <TT>myprof.conf</TT> with the lines
|
||||
</DL>
|
||||
<P>
|
||||
<PRE>
|
||||
PROFILE_PRELIB="-L/usr/local/myprof/lib -lmyprof"
|
||||
PROFILE_INCPATHS="-I/usr/local/myprof/include"
|
||||
</PRE>
|
||||
|
||||
and place it in the sysconfdir directory (this directory is set at
|
||||
configure time when MPICH is built). Then using the command-line
|
||||
argument <TT>\-profile=myprof</TT> will cause these
|
||||
definitions to be added to the relevant compile commands.
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
mpicc, mpicxx, mpifort, mpiexec
|
||||
<BR>
|
||||
</BODY></HTML>
|
||||
@@ -5,7 +5,7 @@
|
||||
<TITLE>mpifort</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="mpifort"><H1>mpifort</H1></A>
|
||||
<H1 id="mpifort">mpifort</H1>
|
||||
Compiles and links MPI programs written in Fortran 90
|
||||
<H2>Description</H2>
|
||||
This command can be used to compile and link MPI programs written in
|
||||
@@ -17,44 +17,44 @@ as it provides the necessary libraries.
|
||||
<P>
|
||||
<H2>Command line arguments</H2>
|
||||
<DL>
|
||||
<DT><B>-show </B><DD>Show the commands that would be used without
|
||||
<DT><B>-show </B> <DD> Show the commands that would be used without
|
||||
running them
|
||||
|
||||
<DT><B>-help </B><DD>Give short help
|
||||
<DT><B>-help </B> <DD> Give short help
|
||||
|
||||
<DT><B>-fc=name </B><DD>Use compiler <TT>name</TT> instead of the default choice. Use
|
||||
<DT><B>-fc=name </B> <DD> Use compiler <tt>name</tt> instead of the default choice. Use
|
||||
this only if the compiler is compatible with the MPICH
|
||||
library (see below)
|
||||
|
||||
<DT><B>-config=name </B><DD>Load a configuration file for a particular compiler.
|
||||
This allows a single <TT>mpifort</TT> command to be used with
|
||||
<DT><B>-config=name </B> <DD> Load a configuration file for a particular compiler.
|
||||
This allows a single <tt>mpifort</tt> command to be used with
|
||||
multiple compilers.
|
||||
|
||||
<DT><B>-compile_info </B><DD>Show the steps for compiling a program. This option
|
||||
<DT><B>-compile_info </B> <DD> Show the steps for compiling a program. This option
|
||||
can be used to see what options and include paths are
|
||||
used by mpifort.
|
||||
|
||||
<DT><B>-link_info </B><DD>Show the steps for linking a program. This option
|
||||
<DT><B>-link_info </B> <DD> Show the steps for linking a program. This option
|
||||
can be used to see what options and libraries are used by
|
||||
mpifort.
|
||||
|
||||
<DT><B>-profile=name </B><DD>Use the MPI profiling given by name. See below for
|
||||
<DT><B>-profile=name </B> <DD> Use the MPI profiling given by name. See below for
|
||||
details
|
||||
|
||||
<DT><B>-echo </B><DD>Show exactly what this program is doing.
|
||||
<DT><B>-echo </B> <DD> Show exactly what this program is doing.
|
||||
This option should normally not be used.
|
||||
|
||||
<DT><B>-static</B><DD>mpi - Use a statically compile MPI library, but shared libraries
|
||||
<DT><B>-static</B> <DD> mpi - Use a statically compile MPI library, but shared libraries
|
||||
for all of the other dependencies.
|
||||
|
||||
<DT><B>others </B><DD>are passed to the compiler or linker. For example, <TT>\-c
|
||||
</TT>causes files to be compiled, <TT>\-g</TT> selects compilation with
|
||||
debugging on most systems, and <TT>\-o name</TT> causes linking
|
||||
with the output executable given the name <TT>name</TT>.
|
||||
<DT><B>others </B> <DD> are passed to the compiler or linker. For example, <tt>\-c
|
||||
</tt>causes files to be compiled, <tt>\-g</tt> selects compilation with
|
||||
debugging on most systems, and <tt>\-o name</tt> causes linking
|
||||
with the output executable given the name <tt>name</tt>.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Environment Variables</H2>
|
||||
The environment variables <TT>MPICH_FC</TT> may be used
|
||||
The environment variables <tt>MPICH_FC</tt> may be used
|
||||
to select different Fortran compiler and linker. Note that since
|
||||
MPICH is built with a particular C and Fortran compiler, change the
|
||||
compilers used can cause problems. Use this only if you could intermix
|
||||
@@ -62,17 +62,17 @@ code compiled with the different compilers.
|
||||
<P>
|
||||
<H2>Compatible Compilers</H2>
|
||||
The MPI library may be used with any compiler that uses the same
|
||||
lengths for basic data objects (such as <TT>long double</TT>) and that
|
||||
lengths for basic data objects (such as <tt>long double</tt>) and that
|
||||
uses compatible run-time libraries. On many systems, the various
|
||||
compilers are compatible and may be used interchangably. There are
|
||||
exceptions; if you use the <TT>MPICH_FC</TT> environment variable or the
|
||||
<TT>\-fc=name</TT> command-line argument to override the choice of compiler
|
||||
compilers are compatible and may be used interchangeably. There are
|
||||
exceptions; if you use the <tt>MPICH_FC</tt> environment variable or the
|
||||
<tt>\-fc=name</tt> command-line argument to override the choice of compiler
|
||||
and encounter problems, try reconfiguring MPICH with the new compiler
|
||||
and installing MPICH in a separate location. See the installation manual
|
||||
for more details.
|
||||
<P>
|
||||
<H2>Examples</H2>
|
||||
To compile a single file <TT>foo.f</TT>, use
|
||||
To compile a single file <tt>foo.f</tt>, use
|
||||
<PRE>
|
||||
mpifort -c foo.f
|
||||
</PRE>
|
||||
@@ -91,29 +91,29 @@ Combining compilation and linking in a single command
|
||||
is a convenient way to build simple programs.
|
||||
<P>
|
||||
<H2>Selecting a Profiling Library</H2>
|
||||
The <TT>\-profile=name</TT> argument allows you to specify an MPI profiling
|
||||
library to be used. <TT>name</TT> can have two forms:
|
||||
The <tt>\-profile=name</tt> argument allows you to specify an MPI profiling
|
||||
library to be used. <tt>name</tt> can have two forms:
|
||||
<P>
|
||||
<BR>A library in the same directory as the MPI library
|
||||
<BR>The name of a profile configuration file
|
||||
<BR>
|
||||
<br>A library in the same directory as the MPI library
|
||||
<br>The name of a profile configuration file
|
||||
<br>
|
||||
<P>
|
||||
If <TT>name</TT> is a library, then this library is included before the MPI
|
||||
If <tt>name</tt> is a library, then this library is included before the MPI
|
||||
library. This allows the simple use of libraries that make use of the
|
||||
MPI profiling interface and that are installed in the same directory as
|
||||
the MPI library.
|
||||
<P>
|
||||
If <TT>name.conf</TT> is the name of a file in the sysconfdir directory, then this
|
||||
If <tt>name.conf</tt> is the name of a file in the sysconfdir directory, then this
|
||||
is read and may define the following variables:
|
||||
<DL>
|
||||
<DT><B>PROFILE_PRELIB </B><DD>Libraries (and paths) to include before the MPI library
|
||||
<DT><B>PROFILE_PRELIB </B> <DD> Libraries (and paths) to include before the MPI library
|
||||
|
||||
<DT><B>PROFILE_POSTLIB </B><DD>Libraries to include after the MPI library
|
||||
<DT><B>PROFILE_POSTLIB </B> <DD> Libraries to include after the MPI library
|
||||
|
||||
<DT><B>PROFILE_INCPATHS </B><DD>C preprocessor arguments for any include files
|
||||
For example, to add <TT>/usr/local/myprof/include</TT> to the include path and
|
||||
the library <TT>libmyprof.a</TT> in <TT>/usr/local/myprof/lib</TT> to the link step,
|
||||
you could create the file <TT>myprof.conf</TT> with the lines
|
||||
<DT><B>PROFILE_INCPATHS </B> <DD> C preprocessor arguments for any include files
|
||||
For example, to add <tt>/usr/local/myprof/include</tt> to the include path and
|
||||
the library <tt>libmyprof.a</tt> in <tt>/usr/local/myprof/lib</tt> to the link step,
|
||||
you could create the file <tt>myprof.conf</tt> with the lines
|
||||
</DL>
|
||||
<P>
|
||||
<PRE>
|
||||
@@ -123,10 +123,10 @@ you could create the file <TT>myprof.conf</TT> with the lines
|
||||
|
||||
and place it in the sysconfdir directory (this directory is set at
|
||||
configure time when MPICH is built). Then using the command-line
|
||||
argument <TT>\-profile=myprof</TT> will cause these
|
||||
argument <tt>\-profile=myprof</tt> will cause these
|
||||
definitions to be added to the relevant compile commands.
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
mpicc, mpicxx, mpifort, mpiexec
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,18 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_commit</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_commit"><H1>MPIR_Type_commit</H1></A>
|
||||
nput Parameters: . datatype_p - pointer to MPI datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIR_Type_commit(MPI_Datatype * datatype_p)
|
||||
</PRE>
|
||||
<H2>Output Parameters</H2>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
0 on success, -1 on failure.
|
||||
</BODY></HTML>
|
||||
@@ -1,27 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_contiguous</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_contiguous"><H1>MPIR_Type_contiguous</H1></A>
|
||||
create a contiguous datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIR_Type_contiguous(int count, MPI_Datatype oldtype, MPI_Datatype * newtype)
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B><DD>number of elements in the contiguous block
|
||||
|
||||
<DT><B>oldtype </B><DD>type (using handle) of datatype on which vector is based
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newtype </B> <DD> handle of new contiguous datatype
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
MPI_SUCCESS on success, MPI error code on failure.
|
||||
</BODY></HTML>
|
||||
@@ -1,29 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_dup</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_dup"><H1>MPIR_Type_dup</H1></A>
|
||||
create a copy of a datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
#undef FUNCNAME
|
||||
#define FUNCNAME MPIR_Type_dup
|
||||
#undef FCNAME
|
||||
#define FCNAME MPL_QUOTE(FUNCNAME)
|
||||
int MPIR_Type_dup(MPI_Datatype oldtype, MPI_Datatype * newtype)
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>oldtype </B> <DD> handle of original datatype
|
||||
</DL>
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newtype </B> <DD> handle of newly created copy of datatype
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
0 on success, MPI error code on failure.
|
||||
</BODY></HTML>
|
||||
@@ -1,39 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_get_contents</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_get_contents"><H1>MPIR_Type_get_contents</H1></A>
|
||||
get content information from datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIR_Type_get_contents(MPI_Datatype datatype,
|
||||
int max_integers,
|
||||
int max_addresses,
|
||||
int max_datatypes,
|
||||
int array_of_integers[],
|
||||
MPI_Aint array_of_addresses[], MPI_Datatype array_of_datatypes[])
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>datatype </B><DD>MPI datatype
|
||||
|
||||
<DT><B>max_integers </B><DD>size of array_of_integers
|
||||
|
||||
<DT><B>max_addresses </B><DD>size of array_of_addresses
|
||||
|
||||
<DT><B>max_datatypes </B><DD>size of array_of_datatypes
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>array_of_integers </B><DD>integers used in creating type
|
||||
|
||||
<DT><B>array_of_addresses </B><DD>MPI_Aints used in creating type
|
||||
|
||||
<DT><B>array_of_datatypes </B><DD>MPI_Datatypes used in creating type
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -1,39 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_indexed</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_indexed"><H1>MPIR_Type_indexed</H1></A>
|
||||
create an indexed datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIR_Type_indexed(int count,
|
||||
const int *blocklength_array,
|
||||
const void *displacement_array,
|
||||
int dispinbytes, MPI_Datatype oldtype, MPI_Datatype * newtype)
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B><DD>number of blocks in type
|
||||
|
||||
<DT><B>blocklength_array </B><DD>number of elements in each block
|
||||
|
||||
<DT><B>displacement_array </B><DD>offsets of blocks from start of type (see next
|
||||
parameter for units)
|
||||
|
||||
<DT><B>dispinbytes </B><DD>if nonzero, then displacements are in bytes (the
|
||||
displacement_array is an array of ints), otherwise they in terms of
|
||||
extent of oldtype (the displacement_array is an array of MPI_Aints)
|
||||
|
||||
<DT><B>oldtype </B><DD>type (using handle) of datatype on which new type is based
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newtype </B> <DD> handle of new indexed datatype
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
0 on success, -1 on failure.
|
||||
</BODY></HTML>
|
||||
@@ -1,38 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_struct</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_struct"><H1>MPIR_Type_struct</H1></A>
|
||||
create a struct datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
#undef FUNCNAME
|
||||
#define FUNCNAME MPIR_Type_struct
|
||||
#undef FCNAME
|
||||
#define FCNAME MPL_QUOTE(FUNCNAME)
|
||||
int MPIR_Type_struct(int count,
|
||||
const int *blocklength_array,
|
||||
const MPI_Aint * displacement_array,
|
||||
const MPI_Datatype * oldtype_array, MPI_Datatype * newtype)
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B><DD>number of blocks in vector
|
||||
|
||||
<DT><B>blocklength_array </B><DD>number of elements in each block
|
||||
|
||||
<DT><B>displacement_array </B><DD>offsets of blocks from start of type in bytes
|
||||
|
||||
<DT><B>oldtype_array </B><DD>types (using handle) of datatypes on which vector is based
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newtype </B> <DD> handle of new struct datatype
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
MPI_SUCCESS on success, MPI errno on failure.
|
||||
</BODY></HTML>
|
||||
@@ -1,38 +0,0 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIR_Type_vector</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIR_Type_vector"><H1>MPIR_Type_vector</H1></A>
|
||||
create a vector datatype
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIR_Type_vector(int count,
|
||||
int blocklength,
|
||||
MPI_Aint stride,
|
||||
int strideinbytes, MPI_Datatype oldtype, MPI_Datatype * newtype)
|
||||
</PRE>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B><DD>number of blocks in vector
|
||||
|
||||
<DT><B>blocklength </B><DD>number of elements in each block
|
||||
|
||||
<DT><B>stride </B><DD>distance from beginning of one block to the next (see next
|
||||
parameter for units)
|
||||
|
||||
<DT><B>strideinbytes </B><DD>if nonzero, then stride is in bytes, otherwise stride
|
||||
is in terms of extent of oldtype
|
||||
|
||||
<DT><B>oldtype </B><DD>type (using handle) of datatype on which vector is based
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newtype </B> <DD> handle of new vector datatype
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return Value</H2>
|
||||
0 on success, MPI error code on failure.
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,110 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Allreduce_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Allreduce_enqueue">MPIX_Allreduce_enqueue</H1>
|
||||
Enqueue an allreduce operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Allreduce_enqueue(const void *sendbuf, void *recvbuf, int count,
|
||||
MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Allreduce_enqueue_c(const void *sendbuf, void *recvbuf,
|
||||
MPI_Count count, MPI_Datatype datatype, MPI_Op op,
|
||||
MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> data type of elements of send buffer (handle)
|
||||
|
||||
<DT><B>op </B> <DD> operation (handle)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> starting address of receive buffer (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OP </B> <DD> Invalid operation. MPI operations (objects of type <tt>MPI_Op</tt>)
|
||||
must either be one of the predefined operations (e.g., <tt>MPI_SUM</tt>) or
|
||||
created with <tt>MPI_Op_create</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Allreduce_enqueue.html">
|
||||
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Async_get_state</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Async_get_state">MPIX_Async_get_state</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
void * MPIX_Async_get_state(MPIX_Async_thing async_thing)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>async_thing </B> <DD> opaque pointer for async thing (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
76
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Async_spawn.html
Normal file
76
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Async_spawn.html
Normal file
@@ -0,0 +1,76 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Async_spawn</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Async_spawn">MPIX_Async_spawn</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Async_spawn(MPIX_Async_thing async_thing,
|
||||
MPIX_Async_poll_function *poll_fn, void *extra_state,
|
||||
MPIX_Stream stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>async_thing </B> <DD> opaque pointer for async thing (None)
|
||||
|
||||
<DT><B>poll_fn </B> <DD> user defined poll function for progressing async things (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state for callback function (None)
|
||||
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
73
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Async_start.html
Normal file
73
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Async_start.html
Normal file
@@ -0,0 +1,73 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Async_start</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Async_start">MPIX_Async_start</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Async_start(MPIX_Async_poll_function *poll_fn, void *extra_state,
|
||||
MPIX_Stream stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>poll_fn </B> <DD> user defined poll function for progressing async things (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state for callback function (None)
|
||||
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -5,57 +5,82 @@
|
||||
<TITLE>MPIX_Comm_agree</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIX_Comm_agree"><H1>MPIX_Comm_agree</H1></A>
|
||||
<H1 id="MPIX_Comm_agree">MPIX_Comm_agree</H1>
|
||||
Performs agreement operation on comm
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_agree(MPI_Comm comm, int *flag)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>comm </B> <DD> communicator (handle)
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newcomm </B> <DD> new communicator (handle)
|
||||
<DL>
|
||||
<DT><B>flag </B> <DD> new communicator (logical)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The purpose of this collective communication is to agree on the integer value flag and on the group of failed processes in comm.
|
||||
<P>
|
||||
On completion, all non-failed MPI processes have agreed to set the output integer value of flag to the result of a bitwise AND operation over the contributed input values of flag. If comm is an intercommunicator, the value of flagis a bitwise AND operation over the values contributed by the remote group.
|
||||
<P>
|
||||
When an MPI process fails before contributing to the operation, the flag is computed ignoring its contribution, and MPIX_Comm_agree raises an error of class MPI_ERR_PROC_FAILED. However, if all MPI processes have acknowledged this failure priorto the call to MPIX_Comm_agree, using MPIX_Comm_failure_ack, the error related to this failure is not raised. When an error of class MPI_ERR_PROC_FAILED is raised, it is consistently raised at allMPI processes, in both the local and remote groups (if applicable).
|
||||
<P>
|
||||
After MPIX_Comm_agree raised an error of class MPI_ERR_PROC_FAILED, a subse-quent call to MPIX_Comm_failure_ack on comm acknowledges the failure of every MPI process that didn<tt>t contribute to the computation offlag.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as </tt>malloc
|
||||
<tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for </tt>MPI_WTIME<tt> and </tt>MPI_WTICK<tt>) have
|
||||
an additional argument </tt>ierr<tt> at the end of the argument list. </tt>ierr
|
||||
<tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
</tt>call<tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., </tt>MPI_Datatype<tt>, </tt>MPI_Comm<tt>) are of type </tt>INTEGER
|
||||
<tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except </tt>MPI_Wtime<tt> and </tt>MPI_Wtick<tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with </tt>MPI_Comm_set_errhandler<tt> (for communicators),
|
||||
</tt>MPI_File_set_errhandler<tt> (for files), and </tt>MPI_Win_set_errhandler<tt> (for
|
||||
RMA windows). The MPI-1 routine </tt>MPI_Errhandler_set<tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
</tt>MPI_ERRORS_RETURN<tt> may be used to cause error values to be returned.
|
||||
Note that MPI does </tt><em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., MPI_ERR_RANK<tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in </tt>MPI_Comm_rank<tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use </tt>MPI_Error_string<tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1,80 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Comm_create_errhandler_x</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Comm_create_errhandler_x">MPIX_Comm_create_errhandler_x</H1>
|
||||
Creates an MPI-style communicator errorhandler with an extra_state
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_create_errhandler_x(MPIX_Comm_errhandler_function_x *comm_errhandler_fn_x,
|
||||
MPIX_Destructor_function *destructor_fn,
|
||||
void *extra_state,
|
||||
MPI_Errhandler *errhandler)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm_errhandler_fn_x </B> <DD> user defined error handling procedure with extra_state (function)
|
||||
|
||||
<DT><B>destructor_fn </B> <DD> callback when op is freed (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>errhandler </B> <DD> MPI error handler (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -5,57 +5,67 @@
|
||||
<TITLE>MPIX_Comm_failure_ack</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIX_Comm_failure_ack"><H1>MPIX_Comm_failure_ack</H1></A>
|
||||
<H1 id="MPIX_Comm_failure_ack">MPIX_Comm_failure_ack</H1>
|
||||
Acknowledge the current group of failed processes
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_failure_ack(MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>comm </B> <DD> Communicator (handle)
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
Because MPI specifies that null objects (e.g., <TT>MPI_COMM_NULL</TT>) are invalid
|
||||
as input to MPI routines unless otherwise specified, using <TT>MPI_COMM_NULL
|
||||
</TT>as input to this routine is an error.
|
||||
This local operation gives the users a way to acknowledge all locally notified failures on comm. After the call, unmatched MPI_ANY_SOURCE receive operations that would haveraised an error of class MPI_ERR_PROC_FAILED_PENDING due to MPI process failure proceed without further raising errors due to those acknowledged failures. Also after this call, MPIX_Comm_agree will not raise an error of class MPI_ERR_PROC_FAILED due to those acknowledged failures.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,61 +5,76 @@
|
||||
<TITLE>MPIX_Comm_failure_get_acked</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIX_Comm_failure_get_acked"><H1>MPIX_Comm_failure_get_acked</H1></A>
|
||||
<H1 id="MPIX_Comm_failure_get_acked">MPIX_Comm_failure_get_acked</H1>
|
||||
Get the group of acknowledged failures.
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIX_Comm_failure_get_acked(MPI_Comm comm, MPI_Group * failedgrp)
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_failure_get_acked(MPI_Comm comm, MPI_Group *failedgrp)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>comm </B> <DD> Communicator (handle)
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>failed_group </B> <DD> Group (handle)
|
||||
<DL>
|
||||
<DT><B>failedgrp </B> <DD> group of failed processes (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
Because MPI specifies that null objects (e.g., <TT>MPI_COMM_NULL</TT>) are invalid
|
||||
as input to MPI routines unless otherwise specified, using <TT>MPI_COMM_NULL
|
||||
</TT>as input to this routine is an error.
|
||||
This local operation returns the group failedgrp of processes, from the communicatorcomm, that have been locally acknowledged as failed by preceding calls to MPIX_Comm_failure_ack. The failedgrp can be empty, that is, equal to MPI_GROUP_EMPTY.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1,82 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Comm_get_failed</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Comm_get_failed">MPIX_Comm_get_failed</H1>
|
||||
This local operation returns the group of processes that are locally known to have failed.
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_get_failed(MPI_Comm comm, MPI_Group *failedgrp)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>failedgrp </B> <DD> group of failed processes (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The returned failedgrp can be empty, that is, equal to MPI_GROUP_EMPTY.
|
||||
<P>
|
||||
For any two groups obtained from calls to this routine at the same MPI process with the same comm, the smaller group is a prefix of the larger group, that is, the same failed process will have the same rank in the returned failedgrp.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,79 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Comm_get_stream</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Comm_get_stream">MPIX_Comm_get_stream</H1>
|
||||
Get the stream object that is attached to the communicator
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_get_stream(MPI_Comm comm, int idx, MPIX_Stream *stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>idx </B> <DD> idx (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -5,46 +5,20 @@
|
||||
<TITLE>MPIX_Comm_revoke</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIX_Comm_revoke"><H1>MPIX_Comm_revoke</H1></A>
|
||||
<H1 id="MPIX_Comm_revoke">MPIX_Comm_revoke</H1>
|
||||
Prevent a communicator from being used in the future
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_revoke(MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>comm </B> <DD> communicator to revoke
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator to revoke (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
Asynchronously notifies all MPI processes associated with the communicator <TT>comm</TT>.
|
||||
This will be manifest by returning the MPIX_ERR_REVOKED during a subsequent MPI
|
||||
call.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,57 +5,25 @@
|
||||
<TITLE>MPIX_Comm_shrink</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPIX_Comm_shrink"><H1>MPIX_Comm_shrink</H1></A>
|
||||
<H1 id="MPIX_Comm_shrink">MPIX_Comm_shrink</H1>
|
||||
Creates a new communitor from an existing communicator while excluding failed processes
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPIX_Comm_shrink(MPI_Comm comm, MPI_Comm * newcomm)
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_shrink(MPI_Comm comm, MPI_Comm *newcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>comm </B> <DD> communicator (handle)
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>newcomm </B> <DD> new communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>newcomm </B> <DD> new communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1,77 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Comm_test_threadcomm</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Comm_test_threadcomm">MPIX_Comm_test_threadcomm</H1>
|
||||
Tests to see if a comm is a threadcomm
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Comm_test_threadcomm(MPI_Comm comm, int *flag)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>flag </B> <DD> true if comm is an inter-communicator (logical)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,80 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_File_create_errhandler_x</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_File_create_errhandler_x">MPIX_File_create_errhandler_x</H1>
|
||||
Creates an MPI-style file errorhandler with an extra_state
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_File_create_errhandler_x(MPIX_File_errhandler_function_x *comm_errhandler_fn_x,
|
||||
MPIX_Destructor_function *destructor_fn,
|
||||
void *extra_state,
|
||||
MPI_Errhandler *errhandler)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm_errhandler_fn_x </B> <DD> user defined error handling procedure with extra_state (function)
|
||||
|
||||
<DT><B>destructor_fn </B> <DD> callback when op is freed (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>errhandler </B> <DD> MPI error handler (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,73 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_GPU_query_support</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_GPU_query_support">MPIX_GPU_query_support</H1>
|
||||
Returns whether the specific type of GPU is supported
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_GPU_query_support(int gpu_type, int *is_supported)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>gpu_type </B> <DD> MPIX_GPU_SUPPORT_CUDA, MPIX_GPU_SUPPORT_ZE, or MPIX_GPU_SUPPORT_HIP (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>is_supported </B> <DD> true if gpu of given type is supported, false otherwise. (logical)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,76 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Grequest_class_allocate</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Grequest_class_allocate">MPIX_Grequest_class_allocate</H1>
|
||||
Create and return a user-defined extended request based on a generalized request class
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Grequest_class_allocate(MPIX_Grequest_class greq_class,
|
||||
void *extra_state, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>greq_class </B> <DD> generalized request class (None)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> generalized request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,86 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Grequest_class_create</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Grequest_class_create">MPIX_Grequest_class_create</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Grequest_class_create(MPI_Grequest_query_function *query_fn,
|
||||
MPI_Grequest_free_function *free_fn,
|
||||
MPI_Grequest_cancel_function *cancel_fn,
|
||||
MPIX_Grequest_poll_function *poll_fn,
|
||||
MPIX_Grequest_wait_function *wait_fn,
|
||||
MPIX_Grequest_class *greq_class)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>query_fn </B> <DD> callback function invoked when request status is queried (function)
|
||||
|
||||
<DT><B>free_fn </B> <DD> callback function invoked when request is freed (function)
|
||||
|
||||
<DT><B>cancel_fn </B> <DD> callback function invoked when request is cancelled (function)
|
||||
|
||||
<DT><B>poll_fn </B> <DD> callback function invoked when request completion is tested (function)
|
||||
|
||||
<DT><B>wait_fn </B> <DD> callback function invoked when request is waited on (function)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>greq_class </B> <DD> generalized request class (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,88 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Grequest_start</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Grequest_start">MPIX_Grequest_start</H1>
|
||||
Create and return a user-defined extended request
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Grequest_start(MPI_Grequest_query_function *query_fn,
|
||||
MPI_Grequest_free_function *free_fn,
|
||||
MPI_Grequest_cancel_function *cancel_fn,
|
||||
MPIX_Grequest_poll_function *poll_fn,
|
||||
MPIX_Grequest_wait_function *wait_fn, void *extra_state,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>query_fn </B> <DD> callback function invoked when request status is queried (function)
|
||||
|
||||
<DT><B>free_fn </B> <DD> callback function invoked when request is freed (function)
|
||||
|
||||
<DT><B>cancel_fn </B> <DD> callback function invoked when request is cancelled (function)
|
||||
|
||||
<DT><B>poll_fn </B> <DD> callback function invoked when request completion is tested (function)
|
||||
|
||||
<DT><B>wait_fn </B> <DD> callback function invoked when request is waited on (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> generalized request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
78
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Info_set_hex.html
Normal file
78
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Info_set_hex.html
Normal file
@@ -0,0 +1,78 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Info_set_hex</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Info_set_hex">MPIX_Info_set_hex</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Info_set_hex(MPI_Info info, const char *key, const void *value,
|
||||
int value_size)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>info </B> <DD> info object (handle)
|
||||
|
||||
<DT><B>key </B> <DD> key (string)
|
||||
|
||||
<DT><B>value </B> <DD> value (choice)
|
||||
|
||||
<DT><B>value_size </B> <DD> size of value (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Irecv_enqueue.html
Normal file
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Irecv_enqueue.html
Normal file
@@ -0,0 +1,118 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Irecv_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Irecv_enqueue">MPIX_Irecv_enqueue</H1>
|
||||
Enqueue a nonblocking receive operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Irecv_enqueue(void *buf, int count, MPI_Datatype datatype, int source,
|
||||
int tag, MPI_Comm comm, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Irecv_enqueue_c(void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int source, int tag, MPI_Comm comm,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B> <DD> number of elements in receive buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each receive buffer element (handle)
|
||||
|
||||
<DT><B>source </B> <DD> rank of source or MPI_ANY_SOURCE (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag or MPI_ANY_TAG (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of receive buffer (choice)
|
||||
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Irecv_enqueue.html">
|
||||
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Isend_enqueue.html
Normal file
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Isend_enqueue.html
Normal file
@@ -0,0 +1,118 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Isend_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Isend_enqueue">MPIX_Isend_enqueue</H1>
|
||||
Enqueue a nonblocking send operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Isend_enqueue(const void *buf, int count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Isend_enqueue_c(const void *buf, MPI_Count count,
|
||||
MPI_Datatype datatype, int dest, int tag,
|
||||
MPI_Comm comm, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each send buffer element (handle)
|
||||
|
||||
<DT><B>dest </B> <DD> rank of destination (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Isend_enqueue.html">
|
||||
94
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Op_create_x.html
Normal file
94
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Op_create_x.html
Normal file
@@ -0,0 +1,94 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Op_create_x</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Op_create_x">MPIX_Op_create_x</H1>
|
||||
Creates a user-defined reduction op with an extra_state
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Op_create_x(MPIX_User_function_x *user_fn_x,
|
||||
MPIX_Destructor_function *destructor_fn, int commute,
|
||||
void *extra_state, MPI_Op *op)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>user_fn_x </B> <DD> user defined function with extra_state (function)
|
||||
|
||||
<DT><B>destructor_fn </B> <DD> callback when op is freed (function)
|
||||
|
||||
<DT><B>commute </B> <DD> true if commutative; false otherwise. (logical)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>op </B> <DD> operation (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes on MPIX_User_function_x</H2>
|
||||
The calling list for the user function type is
|
||||
<PRE>
|
||||
typedef void (MPIX_User_function_x) (void * a, void * b,
|
||||
MPI_Count len, MPI_Datatype datatype,
|
||||
void *extra_state);
|
||||
</PRE>
|
||||
|
||||
There are a few differences from MPI_User_function used in MPI_Op_create.
|
||||
1. It passes len and datatype as scalar input, rather than as addresses.
|
||||
2. The len is of MPI_Count, rather than int.
|
||||
3. It accepts a void *extra_state which is passed from users during MPIX_Op_create_x.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,36 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Query_cuda_support</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Query_cuda_support">MPIX_Query_cuda_support</H1>
|
||||
Returns whether CUDA is supported
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Query_cuda_support(void)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,36 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Query_hip_support</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Query_hip_support">MPIX_Query_hip_support</H1>
|
||||
Returns whether HIP (AMD GPU) is supported
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Query_hip_support(void)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,36 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Query_ze_support</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Query_ze_support">MPIX_Query_ze_support</H1>
|
||||
Returns whether ZE (Intel GPU) is supported
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Query_ze_support(void)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
117
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Recv_enqueue.html
Normal file
117
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Recv_enqueue.html
Normal file
@@ -0,0 +1,117 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Recv_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Recv_enqueue">MPIX_Recv_enqueue</H1>
|
||||
Enqueue a receive operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Recv_enqueue(void *buf, int count, MPI_Datatype datatype, int source,
|
||||
int tag, MPI_Comm comm, MPI_Status *status)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Recv_enqueue_c(void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int source, int tag, MPI_Comm comm, MPI_Status *status)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B> <DD> number of elements in receive buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each receive buffer element (handle)
|
||||
|
||||
<DT><B>source </B> <DD> rank of source or MPI_ANY_SOURCE (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag or MPI_ANY_TAG (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of receive buffer (choice)
|
||||
|
||||
<DT><B>status </B> <DD> status object (Status)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Recv_enqueue.html">
|
||||
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Request_is_complete</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Request_is_complete">MPIX_Request_is_complete</H1>
|
||||
[short description]
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Request_is_complete(MPI_Request request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
108
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Send_enqueue.html
Normal file
108
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Send_enqueue.html
Normal file
@@ -0,0 +1,108 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Send_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Send_enqueue">MPIX_Send_enqueue</H1>
|
||||
Enqueue a send operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Send_enqueue(const void *buf, int count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Send_enqueue_c(const void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each send buffer element (handle)
|
||||
|
||||
<DT><B>dest </B> <DD> rank of destination (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Send_enqueue.html">
|
||||
@@ -0,0 +1,80 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Session_create_errhandler_x</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Session_create_errhandler_x">MPIX_Session_create_errhandler_x</H1>
|
||||
Creates an MPI-style session errorhandler with an extra_state
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Session_create_errhandler_x(MPIX_Session_errhandler_function_x *comm_errhandler_fn_x,
|
||||
MPIX_Destructor_function *destructor_fn,
|
||||
void *extra_state,
|
||||
MPI_Errhandler *errhandler)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm_errhandler_fn_x </B> <DD> user defined error handling procedure with extra_state (function)
|
||||
|
||||
<DT><B>destructor_fn </B> <DD> callback when op is freed (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>errhandler </B> <DD> MPI error handler (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Start_progress_thread</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Start_progress_thread">MPIX_Start_progress_thread</H1>
|
||||
Start a progress thread that will poll progress on the given stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Start_progress_thread(MPIX_Stream stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stop_progress_thread</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stop_progress_thread">MPIX_Stop_progress_thread</H1>
|
||||
Stop the progress thread that polls progress on the given stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stop_progress_thread(MPIX_Stream stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,80 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_comm_create</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_comm_create">MPIX_Stream_comm_create</H1>
|
||||
Create a new communicator with local stream attached
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_comm_create(MPI_Comm comm, MPIX_Stream stream,
|
||||
MPI_Comm *newcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>newcomm </B> <DD> new stream-associated communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,87 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_comm_create_multiplex</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_comm_create_multiplex">MPIX_Stream_comm_create_multiplex</H1>
|
||||
Create a new communicator with multiple local streams attached
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_comm_create_multiplex(MPI_Comm comm, int count,
|
||||
MPIX_Stream array_of_streams[],
|
||||
MPI_Comm *newcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>count </B> <DD> list length (non-negative integer)
|
||||
|
||||
<DT><B>array_of_streams </B> <DD> stream object array (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>newcomm </B> <DD> new stream-associated communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,76 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_create</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_create">MPIX_Stream_create</H1>
|
||||
Create a new stream object
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_create(MPI_Info info, MPIX_Stream *stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>info </B> <DD> info argument (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object created (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
64
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_free.html
Normal file
64
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_free.html
Normal file
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_free</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_free">MPIX_Stream_free</H1>
|
||||
Free a stream object
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_free(MPIX_Stream *stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input/Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_irecv.html
Normal file
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_irecv.html
Normal file
@@ -0,0 +1,124 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_irecv</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_irecv">MPIX_Stream_irecv</H1>
|
||||
Start a nonblocking receive from a specific source stream to a specific local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_irecv(void *buf, int count, MPI_Datatype datatype, int source,
|
||||
int tag, MPI_Comm comm, int source_stream_index,
|
||||
int dest_stream_index, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Stream_irecv_c(void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int source, int tag, MPI_Comm comm,
|
||||
int source_stream_index, int dest_stream_index,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B> <DD> number of elements in receive buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each receive buffer element (handle)
|
||||
|
||||
<DT><B>source </B> <DD> rank of source or MPI_ANY_SOURCE (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag or MPI_ANY_TAG (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>source_stream_index </B> <DD> source_stream_index (integer)
|
||||
|
||||
<DT><B>dest_stream_index </B> <DD> dest_stream_index (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of receive buffer (choice)
|
||||
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Stream_irecv.html">
|
||||
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_isend.html
Normal file
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_isend.html
Normal file
@@ -0,0 +1,124 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_isend</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_isend">MPIX_Stream_isend</H1>
|
||||
Start a nonblocking send from a specific local stream to a specific remote stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_isend(const void *buf, int count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm, int source_stream_index,
|
||||
int dest_stream_index, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Stream_isend_c(const void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm,
|
||||
int source_stream_index, int dest_stream_index,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each send buffer element (handle)
|
||||
|
||||
<DT><B>dest </B> <DD> rank of destination (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>source_stream_index </B> <DD> source_stream_index (integer)
|
||||
|
||||
<DT><B>dest_stream_index </B> <DD> dest_stream_index (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Stream_isend.html">
|
||||
@@ -0,0 +1,64 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_progress</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_progress">MPIX_Stream_progress</H1>
|
||||
Invoke progress on the given stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_progress(MPIX_Stream stream)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>stream </B> <DD> stream object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_recv.html
Normal file
124
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_recv.html
Normal file
@@ -0,0 +1,124 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_recv</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_recv">MPIX_Stream_recv</H1>
|
||||
Receive a message from a specific source stream to a specific local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_recv(void *buf, int count, MPI_Datatype datatype, int source,
|
||||
int tag, MPI_Comm comm, int source_stream_index,
|
||||
int dest_stream_index, MPI_Status *status)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Stream_recv_c(void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int source, int tag, MPI_Comm comm,
|
||||
int source_stream_index, int dest_stream_index,
|
||||
MPI_Status *status)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B> <DD> number of elements in receive buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each receive buffer element (handle)
|
||||
|
||||
<DT><B>source </B> <DD> rank of source or MPI_ANY_SOURCE (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag or MPI_ANY_TAG (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>source_stream_index </B> <DD> source_stream_index (integer)
|
||||
|
||||
<DT><B>dest_stream_index </B> <DD> dest_stream_index (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of receive buffer (choice)
|
||||
|
||||
<DT><B>status </B> <DD> status object (Status)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Stream_recv.html">
|
||||
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_send.html
Normal file
118
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Stream_send.html
Normal file
@@ -0,0 +1,118 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Stream_send</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Stream_send">MPIX_Stream_send</H1>
|
||||
Send message from a specific local stream to a specific destination stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Stream_send(const void *buf, int count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm, int source_stream_index,
|
||||
int dest_stream_index)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPIX_Stream_send_c(const void *buf, MPI_Count count, MPI_Datatype datatype,
|
||||
int dest, int tag, MPI_Comm comm,
|
||||
int source_stream_index, int dest_stream_index)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>buf </B> <DD> initial address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> datatype of each send buffer element (handle)
|
||||
|
||||
<DT><B>dest </B> <DD> rank of destination (integer)
|
||||
|
||||
<DT><B>tag </B> <DD> message tag (integer)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>source_stream_index </B> <DD> source_stream_index (integer)
|
||||
|
||||
<DT><B>dest_stream_index </B> <DD> dest_stream_index (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TAG </B> <DD> Invalid tag argument. Tags must be non-negative; tags
|
||||
in a receive (<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may
|
||||
also be <tt>MPI_ANY_TAG</tt>. The largest tag value is available through the
|
||||
the attribute <tt>MPI_TAG_UB</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPIX_Stream_send.html">
|
||||
@@ -0,0 +1,71 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Threadcomm_finish</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Threadcomm_finish">MPIX_Threadcomm_finish</H1>
|
||||
finish/deactiviate a thread communicator (inside thread parallel regions)
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Threadcomm_finish(MPI_Comm threadcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>threadcomm </B> <DD> thread communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPIX_Threadcomm_init, MPIX_Threadcomm_start, MPIX_Threadcomm_free
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,71 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Threadcomm_free</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Threadcomm_free">MPIX_Threadcomm_free</H1>
|
||||
free a thread communicator (outside thread parallel regions)
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Threadcomm_free(MPI_Comm *threadcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input/Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>threadcomm </B> <DD> thread communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPIX_Threadcomm_init, MPIX_Threadcomm_start, MPIX_Threadcomm_finish
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,83 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Threadcomm_init</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Threadcomm_init">MPIX_Threadcomm_init</H1>
|
||||
initialize a thread communicator (outside thread parallel regions)
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Threadcomm_init(MPI_Comm comm, int num_threads,
|
||||
MPI_Comm *newthreadcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>num_threads </B> <DD> number of participating threads (non-negative integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>newthreadcomm </B> <DD> new thread communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPIX_Threadcomm_start, MPIX_Threadcomm_finish, MPIX_Threadcomm_free
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,71 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Threadcomm_start</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Threadcomm_start">MPIX_Threadcomm_start</H1>
|
||||
start/activiate a thread communicator (inside thread parallel regions)
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Threadcomm_start(MPI_Comm threadcomm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>threadcomm </B> <DD> thread communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPIX_Threadcomm_init, MPIX_Threadcomm_finish, MPIX_Threadcomm_free
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
85
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Type_iov.html
Normal file
85
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Type_iov.html
Normal file
@@ -0,0 +1,85 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Type_iov</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Type_iov">MPIX_Type_iov</H1>
|
||||
Return contiguous segments starting from an iov offset
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Type_iov(MPI_Datatype datatype, MPI_Count iov_offset, MPIX_Iov *iov,
|
||||
MPI_Count max_iov_len, MPI_Count *actual_iov_len)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>datatype </B> <DD> datatype (handle)
|
||||
|
||||
<DT><B>iov_offset </B> <DD> number of contiguous segments to skip (integer)
|
||||
|
||||
<DT><B>max_iov_len </B> <DD> capacity for the segments array (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>iov </B> <DD> contiguous segments array for output (None)
|
||||
|
||||
<DT><B>actual_iov_len </B> <DD> actual number of segments in output (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
83
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Type_iov_len.html
Normal file
83
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Type_iov_len.html
Normal file
@@ -0,0 +1,83 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Type_iov_len</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Type_iov_len">MPIX_Type_iov_len</H1>
|
||||
Return the number of contiguous segments within max_iov_bytes
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Type_iov_len(MPI_Datatype datatype, MPI_Count max_iov_bytes,
|
||||
MPI_Count *iov_len, MPI_Count *actual_iov_bytes)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>datatype </B> <DD> datatype (handle)
|
||||
|
||||
<DT><B>max_iov_bytes </B> <DD> maximum number bytes in the iovs (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>iov_len </B> <DD> number of iovs within max_iov_bytes (integer)
|
||||
|
||||
<DT><B>actual_iov_bytes </B> <DD> actual number of bytes in the iovs (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
77
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Wait_enqueue.html
Normal file
77
linx64/mpi/mpich/share/doc/mpich/www3/MPIX_Wait_enqueue.html
Normal file
@@ -0,0 +1,77 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Wait_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Wait_enqueue">MPIX_Wait_enqueue</H1>
|
||||
Enqueue a wait operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Wait_enqueue(MPI_Request *request, MPI_Status *status)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input/Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>request </B> <DD> request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>status </B> <DD> status object (Status)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_REQUEST </B> <DD> Invalid <tt>MPI_Request</tt>. Either null or, in the case of a
|
||||
<tt>MPI_Start</tt> or <tt>MPI_Startall</tt>, not a persistent request.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,83 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Waitall_enqueue</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Waitall_enqueue">MPIX_Waitall_enqueue</H1>
|
||||
Enqueue a waitall operation to a GPU stream that is associated with the local stream
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Waitall_enqueue(int count, MPI_Request array_of_requests[],
|
||||
MPI_Status *array_of_statuses)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>count </B> <DD> lists length (non-negative integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Input/Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>array_of_requests </B> <DD> array of requests (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>array_of_statuses </B> <DD> array of status objects (Status)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_REQUEST </B> <DD> Invalid <tt>MPI_Request</tt>. Either null or, in the case of a
|
||||
<tt>MPI_Start</tt> or <tt>MPI_Startall</tt>, not a persistent request.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1,79 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPIX_Win_create_errhandler_x</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPIX_Win_create_errhandler_x">MPIX_Win_create_errhandler_x</H1>
|
||||
Creates an MPI-style window errorhandler with an extra_state
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPIX_Win_create_errhandler_x(MPIX_Win_errhandler_function_x *comm_errhandler_fn_x,
|
||||
MPIX_Destructor_function *destructor_fn,
|
||||
void *extra_state, MPI_Errhandler *errhandler)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm_errhandler_fn_x </B> <DD> user defined error handling procedure with extra_state (function)
|
||||
|
||||
<DT><B>destructor_fn </B> <DD> callback when op is freed (function)
|
||||
|
||||
<DT><B>extra_state </B> <DD> extra state (None)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>errhandler </B> <DD> MPI error handler (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -5,22 +5,26 @@
|
||||
<TITLE>MPI_Abort</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Abort"><H1>MPI_Abort</H1></A>
|
||||
<H1 id="MPI_Abort">MPI_Abort</H1>
|
||||
Terminates MPI execution environment
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Abort(MPI_Comm comm, int errorcode)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>comm </B><DD>communicator of tasks to abort
|
||||
<DT><B>comm </B> <DD> communicator of tasks to abort (handle)
|
||||
|
||||
<DT><B>errorcode </B><DD>error code to return to invoking environment
|
||||
<DT><B>errorcode </B> <DD> error code to return to invoking environment (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
Terminates all MPI processes associated with the communicator <TT>comm</TT>; in
|
||||
most systems (all to date), terminates <EM>all</EM> processes.
|
||||
Terminates all MPI processes associated with the communicator <tt>comm</tt>; in
|
||||
most systems (all to date), terminates <em>all</em> processes.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
@@ -31,38 +35,48 @@ not be used from within a signal handler.
|
||||
The MPI standard defined a thread-safe interface but this does not
|
||||
mean that all routines may be called without any thread locks. For
|
||||
example, two threads must not attempt to change the contents of the
|
||||
same <TT>MPI_Info</TT> object concurrently. The user is responsible in this
|
||||
same <tt>MPI_Info</tt> object concurrently. The user is responsible in this
|
||||
case for using some mechanism, such as thread locks, to ensure that
|
||||
only one thread at a time makes use of this routine.
|
||||
Because the <TT>MPI_Abort</TT> routine is intended to ensure that an MPI
|
||||
Because the <tt>MPI_Abort</tt> routine is intended to ensure that an MPI
|
||||
process (and possibly an entire job), it cannot wait for a thread to
|
||||
release a lock or other mechanism for atomic access.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,86 +5,105 @@
|
||||
<TITLE>MPI_Accumulate</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Accumulate"><H1>MPI_Accumulate</H1></A>
|
||||
<H1 id="MPI_Accumulate">MPI_Accumulate</H1>
|
||||
Accumulate data into the target process using remote memory access
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPI_Accumulate(const void *origin_addr, int origin_count, MPI_Datatype
|
||||
origin_datatype, int target_rank, MPI_Aint
|
||||
target_disp, int target_count, MPI_Datatype
|
||||
target_datatype, MPI_Op op, MPI_Win win)
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Accumulate(const void *origin_addr, int origin_count,
|
||||
MPI_Datatype origin_datatype, int target_rank,
|
||||
MPI_Aint target_disp, int target_count,
|
||||
MPI_Datatype target_datatype, MPI_Op op, MPI_Win win)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Accumulate_c(const void *origin_addr, MPI_Count origin_count,
|
||||
MPI_Datatype origin_datatype, int target_rank,
|
||||
MPI_Aint target_disp, MPI_Count target_count,
|
||||
MPI_Datatype target_datatype, MPI_Op op, MPI_Win win)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>origin_addr </B><DD>initial address of buffer (choice)
|
||||
<DT><B>origin_addr </B> <DD> initial address of buffer (choice)
|
||||
|
||||
<DT><B>origin_count </B><DD>number of entries in buffer (nonnegative integer)
|
||||
<DT><B>origin_count </B> <DD> number of entries in buffer (non-negative integer)
|
||||
|
||||
<DT><B>origin_datatype </B><DD>datatype of each buffer entry (handle)
|
||||
<DT><B>origin_datatype </B> <DD> datatype of each entry (handle)
|
||||
|
||||
<DT><B>target_rank </B><DD>rank of target (nonnegative integer)
|
||||
<DT><B>target_rank </B> <DD> rank of target (non-negative integer)
|
||||
|
||||
<DT><B>target_disp </B><DD>displacement from start of window to beginning of target
|
||||
buffer (nonnegative integer)
|
||||
<DT><B>target_disp </B> <DD> displacement from start of window to beginning of target buffer (non-negative integer)
|
||||
|
||||
<DT><B>target_count </B><DD>number of entries in target buffer (nonnegative integer)
|
||||
<DT><B>target_count </B> <DD> number of entries in target buffer (non-negative integer)
|
||||
|
||||
<DT><B>target_datatype </B><DD>datatype of each entry in target buffer (handle)
|
||||
<DT><B>target_datatype </B> <DD> datatype of each entry in target buffer (handle)
|
||||
|
||||
<DT><B>op </B><DD>predefined reduce operation (handle)
|
||||
<DT><B>op </B> <DD> reduce operation (handle)
|
||||
|
||||
<DT><B>win </B><DD>window object (handle)
|
||||
<DT><B>win </B> <DD> window object (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The basic components of both the origin and target datatype must be the same
|
||||
predefined datatype (e.g., all <TT>MPI_INT</TT> or all <TT>MPI_DOUBLE_PRECISION</TT>).
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <TT>MPI_ERR_RANK</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<TT>MPI_Recv</TT>, <TT>MPI_Irecv</TT>, <TT>MPI_Sendrecv</TT>, etc.) may also be <TT>MPI_ANY_SOURCE</TT>.
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_DISP </B> <DD>
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <TT>MPI_Type_commit</TT>) is used
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_RANK </B> <DD> Invalid source or destination rank. Ranks must be between
|
||||
zero and the size of the communicator minus one; ranks in a receive
|
||||
(<tt>MPI_Recv</tt>, <tt>MPI_Irecv</tt>, <tt>MPI_Sendrecv</tt>, etc.) may also be <tt>MPI_ANY_SOURCE</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_WIN </B> <DD> Invalid MPI window object
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_WIN </B> <DD> Invalid MPI window object
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPI_Raccumulate
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Accumulate.html">
|
||||
@@ -5,14 +5,19 @@
|
||||
<TITLE>MPI_Add_error_class</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Add_error_class"><H1>MPI_Add_error_class</H1></A>
|
||||
<H1 id="MPI_Add_error_class">MPI_Add_error_class</H1>
|
||||
Add an MPI error class to the known classes
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Add_error_class(int *errorclass)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>errorclass </B> <DD> New error class
|
||||
<DL>
|
||||
<DT><B>errorclass </B> <DD> value for the new error class (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
@@ -20,37 +25,44 @@ int MPI_Add_error_class(int *errorclass)
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <TT>MPI_Error_string</TT> to get more information
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,18 +5,24 @@
|
||||
<TITLE>MPI_Add_error_code</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Add_error_code"><H1>MPI_Add_error_code</H1></A>
|
||||
<H1 id="MPI_Add_error_code">MPI_Add_error_code</H1>
|
||||
Add an MPI error code to an MPI error class
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Add_error_code(int errorclass, int *errorcode)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>errorclass </B> <DD> Error class to add an error code.
|
||||
<DL>
|
||||
<DT><B>errorclass </B> <DD> error class (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>errorcode </B> <DD> New error code for this error class.
|
||||
<DL>
|
||||
<DT><B>errorcode </B> <DD> new error code to be associated with errorclass (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
@@ -24,37 +30,44 @@ int MPI_Add_error_code(int errorclass, int *errorcode)
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <TT>MPI_Error_string</TT> to get more information
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,29 +5,33 @@
|
||||
<TITLE>MPI_Add_error_string</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Add_error_string"><H1>MPI_Add_error_string</H1></A>
|
||||
<H1 id="MPI_Add_error_string">MPI_Add_error_string</H1>
|
||||
Associates an error string with an MPI error code or class
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Add_error_string(int errorcode, const char *string)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>errorcode </B><DD>error code or class (integer)
|
||||
<DT><B>errorcode </B> <DD> error code or class (integer)
|
||||
|
||||
<DT><B>string </B><DD>text corresponding to errorcode (string)
|
||||
<DT><B>string </B> <DD> text corresponding to errorcode (string)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The string must be no more than <TT>MPI_MAX_ERROR_STRING</TT> characters long.
|
||||
The string must be no more than <tt>MPI_MAX_ERROR_STRING</tt> characters long.
|
||||
The length of the string is as defined in the calling language.
|
||||
The length of the string does not include the null terminator in C or C++.
|
||||
Note that the string is <TT>const</TT> even though the MPI standard does not
|
||||
Note that the string is <tt>const</tt> even though the MPI standard does not
|
||||
specify it that way.
|
||||
<P>
|
||||
According to the MPI-2 standard, it is erroneous to call <TT>MPI_Add_error_string
|
||||
</TT>for an error code or class with a value less than or equal
|
||||
to <TT>MPI_ERR_LASTCODE</TT>. Thus, you cannot replace the predefined error messages
|
||||
According to the MPI-2 standard, it is erroneous to call <tt>MPI_Add_error_string
|
||||
</tt>for an error code or class with a value less than or equal
|
||||
to <tt>MPI_ERR_LASTCODE</tt>. Thus, you cannot replace the predefined error messages
|
||||
with this routine.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
@@ -35,34 +39,44 @@ with this routine.
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,68 +5,74 @@
|
||||
<TITLE>MPI_Address</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Address"><H1>MPI_Address</H1></A>
|
||||
<H1 id="MPI_Address">MPI_Address</H1>
|
||||
Gets the address of a location in memory
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPI_Address(void *location, MPI_Aint * address)
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Address(void *location, MPI_Aint *address)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL><DT><B>location </B> <DD> location in caller memory (choice)
|
||||
<DL>
|
||||
<DT><B>location </B> <DD> location in caller memory (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>address </B> <DD> address of location (address integer)
|
||||
<DL>
|
||||
<DT><B>address </B> <DD> address of location (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Note</H2>
|
||||
This routine is provided for both the Fortran and C programmers.
|
||||
On many systems, the address returned by this routine will be the same
|
||||
as produced by the C <TT>&</TT> operator, but this is not required in C and
|
||||
may not be true of systems with word- rather than byte-oriented
|
||||
instructions or systems with segmented address spaces.
|
||||
<H2>Removed Function</H2>
|
||||
The current MPI standard defines this routine as removed. The user should use
|
||||
the replacement routine.
|
||||
The replacement for this routine is <tt>MPI_Get_address</tt>.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is both thread- and interrupt-safe.
|
||||
This means that this routine may safely be used by multiple threads and
|
||||
from within a signal handler.
|
||||
<P>
|
||||
<H2>Deprecated Function</H2>
|
||||
The MPI-2 standard deprecated a number of routines because MPI-2 provides
|
||||
better versions. This routine is one of those that was deprecated. The
|
||||
routine may continue to be used, but new code should use the replacement
|
||||
routine.
|
||||
The replacement for this routine is <TT>MPI_Get_address</TT>.
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <TT>MPI_Error_string</TT> to get more information
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,22 +5,23 @@
|
||||
<TITLE>MPI_Aint_add</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Aint_add"><H1>MPI_Aint_add</H1></A>
|
||||
<H1 id="MPI_Aint_add">MPI_Aint_add</H1>
|
||||
Returns the sum of base and disp
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
MPI_Aint MPI_Aint_add(MPI_Aint base, MPI_Aint disp)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>base </B><DD>base address (integer)
|
||||
<DT><B>base </B> <DD> base address (integer)
|
||||
|
||||
<DT><B>disp </B><DD>displacement (integer)
|
||||
<DT><B>disp </B> <DD> displacement (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return value</H2>
|
||||
Sum of the base and disp argument
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
MPI_Aint_Add produces a new MPI_Aint value that is equivalent to the sum of the
|
||||
base and disp arguments, where base represents a base address returned by a call
|
||||
@@ -28,7 +29,7 @@ to MPI_GET_ADDRESS and disp represents a signed integer displacement. The result
|
||||
address is valid only at the process that generated base, and it must correspond
|
||||
to a location in the same object referenced by base. The addition is performed in
|
||||
a manner that results in the correct MPI_Aint representation of the output address,
|
||||
as if the process that originally produced base had called:
|
||||
as if the process that originally produced base had called
|
||||
<PRE>
|
||||
MPI_Get_address((char *) base + disp, &result)
|
||||
</PRE>
|
||||
@@ -36,5 +37,5 @@ as if the process that originally produced base had called:
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPI_Aint_diff
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,22 +5,23 @@
|
||||
<TITLE>MPI_Aint_diff</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Aint_diff"><H1>MPI_Aint_diff</H1></A>
|
||||
<H1 id="MPI_Aint_diff">MPI_Aint_diff</H1>
|
||||
Returns the difference between addr1 and addr2
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
MPI_Aint MPI_Aint_diff(MPI_Aint addr1, MPI_Aint addr2)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>addr1 </B><DD>minuend address (integer)
|
||||
<DT><B>addr1 </B> <DD> minuend address (integer)
|
||||
|
||||
<DT><B>addr2 </B><DD>subtrahend address (integer)
|
||||
<DT><B>addr2 </B> <DD> subtrahend address (integer)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Return value</H2>
|
||||
Difference between addr1 and addr2
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
MPI_Aint_diff produces a new MPI_Aint value that is equivalent to the difference
|
||||
between addr1 and addr2 arguments, where addr1 and addr2 represent addresses
|
||||
@@ -37,5 +38,5 @@ on the addresses initially passed to MPI_GET_ADDRESS.
|
||||
<P>
|
||||
<H2>See Also</H2>
|
||||
MPI_Aint_add
|
||||
<BR>
|
||||
<br>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,54 +5,66 @@
|
||||
<TITLE>MPI_Allgather</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Allgather"><H1>MPI_Allgather</H1></A>
|
||||
<H1 id="MPI_Allgather">MPI_Allgather</H1>
|
||||
Gathers data from all tasks and distribute the combined data to all tasks
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
int MPI_Allgather(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
|
||||
void *recvbuf, int recvcount, MPI_Datatype recvtype, MPI_Comm comm)
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allgather(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
|
||||
void *recvbuf, int recvcount, MPI_Datatype recvtype,
|
||||
MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allgather_c(const void *sendbuf, MPI_Count sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf, MPI_Count recvcount,
|
||||
MPI_Datatype recvtype, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B><DD>starting address of send buffer (choice)
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>sendcount </B><DD>number of elements in send buffer (integer)
|
||||
<DT><B>sendcount </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>sendtype </B><DD>data type of send buffer elements (handle)
|
||||
<DT><B>sendtype </B> <DD> data type of send buffer elements (handle)
|
||||
|
||||
<DT><B>recvcount </B><DD>number of elements received from any process (integer)
|
||||
<DT><B>recvcount </B> <DD> number of elements received from any process (non-negative integer)
|
||||
|
||||
<DT><B>recvtype </B><DD>data type of receive buffer elements (handle)
|
||||
<DT><B>recvtype </B> <DD> data type of receive buffer elements (handle)
|
||||
|
||||
<DT><B>comm </B><DD>communicator (handle)
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The MPI standard (1.0 and 1.1) says that
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
The jth block of data sent from each process is received by every process
|
||||
and placed in the jth block of the buffer <TT>recvbuf</TT>.
|
||||
<BR>
|
||||
and placed in the jth block of the buffer <tt>recvbuf</tt>.
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
This is misleading; a better description is
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
The block of data sent from the jth process is received by every
|
||||
process and placed in the jth block of the buffer <TT>recvbuf</TT>.
|
||||
<BR>
|
||||
process and placed in the jth block of the buffer <tt>recvbuf</tt>.
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
This text was suggested by Rajeev Thakur and has been adopted as a
|
||||
clarification by the MPI Forum.
|
||||
@@ -62,45 +74,57 @@ clarification by the MPI Forum.
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <TT>MPI_Type_commit</TT>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Allgather.html">
|
||||
117
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allgather_init.html
Normal file
117
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allgather_init.html
Normal file
@@ -0,0 +1,117 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPI_Allgather_init</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPI_Allgather_init">MPI_Allgather_init</H1>
|
||||
Create a persistent request for allgather.
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allgather_init(const void *sendbuf, int sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf, int recvcount,
|
||||
MPI_Datatype recvtype, MPI_Comm comm, MPI_Info info,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allgather_init_c(const void *sendbuf, MPI_Count sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf,
|
||||
MPI_Count recvcount, MPI_Datatype recvtype,
|
||||
MPI_Comm comm, MPI_Info info, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>sendcount </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>sendtype </B> <DD> data type of send buffer elements (handle)
|
||||
|
||||
<DT><B>recvcount </B> <DD> number of elements received from any process (non-negative integer)
|
||||
|
||||
<DT><B>recvtype </B> <DD> data type of receive buffer elements (handle)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>info </B> <DD> info argument (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Allgather_init.html">
|
||||
@@ -5,61 +5,70 @@
|
||||
<TITLE>MPI_Allgatherv</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Allgatherv"><H1>MPI_Allgatherv</H1></A>
|
||||
<H1 id="MPI_Allgatherv">MPI_Allgatherv</H1>
|
||||
Gathers data from all tasks and deliver the combined data to all tasks
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allgatherv(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
|
||||
void *recvbuf, const int *recvcounts, const int *displs,
|
||||
void *recvbuf, const int recvcounts[], const int displs[],
|
||||
MPI_Datatype recvtype, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allgatherv_c(const void *sendbuf, MPI_Count sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf,
|
||||
const MPI_Count recvcounts[], const MPI_Aint displs[],
|
||||
MPI_Datatype recvtype, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B><DD>starting address of send buffer (choice)
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>sendcount </B><DD>number of elements in send buffer (integer)
|
||||
<DT><B>sendcount </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>sendtype </B><DD>data type of send buffer elements (handle)
|
||||
<DT><B>sendtype </B> <DD> data type of send buffer elements (handle)
|
||||
|
||||
<DT><B>recvcounts </B><DD>integer array (of length group size)
|
||||
containing the number of elements that are to be received from each process
|
||||
<DT><B>recvcounts </B> <DD> non-negative integer array (of length group size) containing the number of elements that are received from each process (non-negative integer)
|
||||
|
||||
<DT><B>displs </B><DD>integer array (of length group size). Entry
|
||||
<TT>i</TT> specifies the displacement (relative to recvbuf) at
|
||||
which to place the incoming data from process <TT>i
|
||||
</TT>
|
||||
<DT><B>recvtype </B><DD>data type of receive buffer elements (handle)
|
||||
<DT><B>displs </B> <DD> integer array (of length group size). Entry i specifies the displacement (relative to recvbuf) at which to place the incoming data from process i (integer)
|
||||
|
||||
<DT><B>comm </B><DD>communicator (handle)
|
||||
<DT><B>recvtype </B> <DD> data type of receive buffer elements (handle)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
The MPI standard (1.0 and 1.1) says that
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
The jth block of data sent from
|
||||
each process is received by every process and placed in the jth block of the
|
||||
buffer <TT>recvbuf</TT>.
|
||||
<BR>
|
||||
buffer <tt>recvbuf</tt>.
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
This is misleading; a better description is
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
The block of data sent from the jth process is received by every
|
||||
process and placed in the jth block of the buffer <TT>recvbuf</TT>.
|
||||
<BR>
|
||||
process and placed in the jth block of the buffer <tt>recvbuf</tt>.
|
||||
<br>
|
||||
<P>
|
||||
<BR>
|
||||
<br>
|
||||
<P>
|
||||
This text was suggested by Rajeev Thakur, and has been adopted as a
|
||||
clarification to the MPI standard by the MPI-Forum.
|
||||
@@ -69,42 +78,57 @@ clarification to the MPI standard by the MPI-Forum.
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <TT>MPI_Type_commit</TT>) is used
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Allgatherv.html">
|
||||
121
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allgatherv_init.html
Normal file
121
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allgatherv_init.html
Normal file
@@ -0,0 +1,121 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPI_Allgatherv_init</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPI_Allgatherv_init">MPI_Allgatherv_init</H1>
|
||||
Create a persistent request for allgatherv.
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allgatherv_init(const void *sendbuf, int sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf,
|
||||
const int recvcounts[], const int displs[],
|
||||
MPI_Datatype recvtype, MPI_Comm comm, MPI_Info info,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allgatherv_init_c(const void *sendbuf, MPI_Count sendcount,
|
||||
MPI_Datatype sendtype, void *recvbuf,
|
||||
const MPI_Count recvcounts[], const MPI_Aint displs[],
|
||||
MPI_Datatype recvtype, MPI_Comm comm, MPI_Info info,
|
||||
MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>sendcount </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>sendtype </B> <DD> data type of send buffer elements (handle)
|
||||
|
||||
<DT><B>recvcounts </B> <DD> non-negative integer array (of length group size) containing the number of elements that are received from each process (non-negative integer)
|
||||
|
||||
<DT><B>displs </B> <DD> integer array (of length group size). Entry i specifies the displacement (relative to recvbuf) at which to place the incoming data from process i (integer)
|
||||
|
||||
<DT><B>recvtype </B> <DD> data type of receive buffer elements (handle)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>info </B> <DD> info argument (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> address of receive buffer (choice)
|
||||
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Allgatherv_init.html">
|
||||
@@ -5,21 +5,26 @@
|
||||
<TITLE>MPI_Alloc_mem</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Alloc_mem"><H1>MPI_Alloc_mem</H1></A>
|
||||
<H1 id="MPI_Alloc_mem">MPI_Alloc_mem</H1>
|
||||
Allocate memory for message passing and RMA
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Alloc_mem(MPI_Aint size, MPI_Info info, void *baseptr)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>size </B><DD>size of memory segment in bytes (nonnegative integer)
|
||||
<DT><B>size </B> <DD> size of memory segment in bytes (non-negative integer)
|
||||
|
||||
<DT><B>info </B><DD>info argument (handle)
|
||||
<DT><B>info </B> <DD> info argument (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>baseptr </B> <DD> pointer to beginning of memory segment allocated
|
||||
<DL>
|
||||
<DT><B>baseptr </B> <DD> pointer to beginning of memory segment allocated (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes</H2>
|
||||
@@ -27,9 +32,9 @@ Using this routine from Fortran requires that the Fortran compiler accept
|
||||
a common pointer extension. See Section 4.11 (Memory Allocation) in the
|
||||
MPI-2 standard for more information and examples.
|
||||
<P>
|
||||
Also note that while <TT>baseptr</TT> is a <TT>void *</TT> type, this is
|
||||
Also note that while <tt>baseptr</tt> is a <tt>void *</tt> type, this is
|
||||
simply to allow easy use of any pointer object for this parameter.
|
||||
In fact, this argument is really a <TT>void **</TT> type, that is, a
|
||||
In fact, this argument is really a <tt>void **</tt> type, that is, a
|
||||
pointer to a pointer.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
@@ -37,43 +42,52 @@ pointer to a pointer.
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <TT>MPI_ERR_RANK</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_NO_MEM </B> <DD> Insufficient memory available for allocation by
|
||||
<TT>MPI_Alloc_mem
|
||||
</TT>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_NO_MEM </B> <DD> Insufficient memory available for allocation by
|
||||
<tt>MPI_Alloc_mem
|
||||
</tt>
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -5,89 +5,112 @@
|
||||
<TITLE>MPI_Allreduce</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<A NAME="MPI_Allreduce"><H1>MPI_Allreduce</H1></A>
|
||||
<H1 id="MPI_Allreduce">MPI_Allreduce</H1>
|
||||
Combines values from all processes and distributes the result back to all processes
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allreduce(const void *sendbuf, void *recvbuf, int count,
|
||||
MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allreduce_c(const void *sendbuf, void *recvbuf, MPI_Count count,
|
||||
MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B><DD>starting address of send buffer (choice)
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B><DD>number of elements in send buffer (integer)
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B><DD>data type of elements of send buffer (handle)
|
||||
<DT><B>datatype </B> <DD> data type of elements of send buffer (handle)
|
||||
|
||||
<DT><B>op </B><DD>operation (handle)
|
||||
<DT><B>op </B> <DD> operation (handle)
|
||||
|
||||
<DT><B>comm </B><DD>communicator (handle)
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL><DT><B>recvbuf </B> <DD> starting address of receive buffer (choice)
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> starting address of receive buffer (choice)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Notes on collective operations</H2>
|
||||
<P>
|
||||
The reduction functions (<tt>MPI_Op</tt>) do not return an error value. As a result,
|
||||
if the functions detect an error, all they can do is either call <tt>MPI_Abort
|
||||
</tt>or silently skip the problem. Thus, if you change the error handler from
|
||||
<tt>MPI_ERRORS_ARE_FATAL</tt> to something else, for example, <tt>MPI_ERRORS_RETURN</tt>,
|
||||
then no error may be indicated.
|
||||
<P>
|
||||
The reason for this is the performance problems in ensuring that
|
||||
all collective routines return the same error value.
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <TT>malloc
|
||||
</TT>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <TT>MPI_WTIME</TT> and <TT>MPI_WTICK</TT>) have
|
||||
an additional argument <TT>ierr</TT> at the end of the argument list. <TT>ierr
|
||||
</TT>is an integer and has the same meaning as the return value of the routine
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<TT>call</TT> statement.
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <TT>MPI_Datatype</TT>, <TT>MPI_Comm</TT>) are of type <TT>INTEGER
|
||||
</TT>in Fortran.
|
||||
<P>
|
||||
<H2>Notes on collective operations</H2>
|
||||
<P>
|
||||
The reduction functions (<TT>MPI_Op</TT>) do not return an error value. As a result,
|
||||
if the functions detect an error, all they can do is either call <TT>MPI_Abort
|
||||
</TT>or silently skip the problem. Thus, if you change the error handler from
|
||||
<TT>MPI_ERRORS_ARE_FATAL</TT> to something else, for example, <TT>MPI_ERRORS_RETURN</TT>,
|
||||
then no error may be indicated.
|
||||
<P>
|
||||
The reason for this is the performance problems in ensuring that
|
||||
all collective routines return the same error value.
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <TT>MPI_Wtime</TT> and <TT>MPI_Wtick</TT>) return an error value;
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <TT>MPI_Comm_set_errhandler</TT> (for communicators),
|
||||
<TT>MPI_File_set_errhandler</TT> (for files), and <TT>MPI_Win_set_errhandler</TT> (for
|
||||
RMA windows). The MPI-1 routine <TT>MPI_Errhandler_set</TT> may be used but
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<TT>MPI_ERRORS_RETURN</TT> may be used to cause error values to be returned.
|
||||
Note that MPI does <EM>not</EM> guarentee that an MPI program can continue past
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL><DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <TT>MPI_Type_commit</TT>) is used
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OP </B> <DD> Invalid operation. MPI operations (objects of type <tt>MPI_Op</tt>)
|
||||
must either be one of the predefined operations (e.g., <tt>MPI_SUM</tt>) or
|
||||
created with <tt>MPI_Op_create</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_OP </B> <DD> Invalid operation. MPI operations (objects of type <TT>MPI_Op</TT>)
|
||||
must either be one of the predefined operations (e.g., <TT>MPI_SUM</TT>) or
|
||||
created with <TT>MPI_Op_create</TT>.
|
||||
</DL>
|
||||
<DL><DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <TT>MPI_Comm_rank</TT>).
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
<meta http-equiv="refresh" content="0; url=MPI_Allreduce.html">
|
||||
118
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allreduce_init.html
Normal file
118
linx64/mpi/mpich/share/doc/mpich/www3/MPI_Allreduce_init.html
Normal file
@@ -0,0 +1,118 @@
|
||||
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML3.2 EN">
|
||||
<HTML>
|
||||
<HEAD>
|
||||
<META NAME="GENERATOR" CONTENT="DOCTEXT">
|
||||
<TITLE>MPI_Allreduce_init</TITLE>
|
||||
</HEAD>
|
||||
<BODY BGCOLOR="FFFFFF">
|
||||
<H1 id="MPI_Allreduce_init">MPI_Allreduce_init</H1>
|
||||
Create a persistent request for allreduce
|
||||
<H2>Synopsis</H2>
|
||||
<PRE>
|
||||
</PRE>
|
||||
<PRE>
|
||||
int MPI_Allreduce_init(const void *sendbuf, void *recvbuf, int count,
|
||||
MPI_Datatype datatype, MPI_Op op, MPI_Comm comm,
|
||||
MPI_Info info, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<PRE>
|
||||
int MPI_Allreduce_init_c(const void *sendbuf, void *recvbuf, MPI_Count count,
|
||||
MPI_Datatype datatype, MPI_Op op, MPI_Comm comm,
|
||||
MPI_Info info, MPI_Request *request)
|
||||
</PRE>
|
||||
|
||||
<P>
|
||||
<H2>Input Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>sendbuf </B> <DD> starting address of send buffer (choice)
|
||||
|
||||
<DT><B>count </B> <DD> number of elements in send buffer (non-negative integer)
|
||||
|
||||
<DT><B>datatype </B> <DD> data type of elements of send buffer (handle)
|
||||
|
||||
<DT><B>op </B> <DD> operation (handle)
|
||||
|
||||
<DT><B>comm </B> <DD> communicator (handle)
|
||||
|
||||
<DT><B>info </B> <DD> info argument (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Output Parameters</H2>
|
||||
<DL>
|
||||
<DT><B>recvbuf </B> <DD> starting address of receive buffer (choice)
|
||||
|
||||
<DT><B>request </B> <DD> communication request (handle)
|
||||
</DL>
|
||||
<P>
|
||||
<H2>Thread and Interrupt Safety</H2>
|
||||
<P>
|
||||
This routine is thread-safe. This means that this routine may be
|
||||
safely used by multiple threads without the need for any user-provided
|
||||
thread locks. However, the routine is not interrupt safe. Typically,
|
||||
this is due to the use of memory allocation routines such as <tt>malloc
|
||||
</tt>or other non-MPICH runtime routines that are themselves not interrupt-safe.
|
||||
<P>
|
||||
<H2>Notes for Fortran</H2>
|
||||
All MPI routines in Fortran (except for <tt>MPI_WTIME</tt> and <tt>MPI_WTICK</tt>) have
|
||||
an additional argument <tt>ierr</tt> at the end of the argument list. <tt>ierr
|
||||
</tt>is an integer and has the same meaning as the return value of the routine
|
||||
in C. In Fortran, MPI routines are subroutines, and are invoked with the
|
||||
<tt>call</tt> statement.
|
||||
<P>
|
||||
All MPI objects (e.g., <tt>MPI_Datatype</tt>, <tt>MPI_Comm</tt>) are of type <tt>INTEGER
|
||||
</tt>in Fortran.
|
||||
<P>
|
||||
<H2>Errors</H2>
|
||||
<P>
|
||||
All MPI routines (except <tt>MPI_Wtime</tt> and <tt>MPI_Wtick</tt>) return an error value;
|
||||
C routines as the value of the function and Fortran routines in the last
|
||||
argument. Before the value is returned, the current MPI error handler is
|
||||
called. By default, this error handler aborts the MPI job. The error handler
|
||||
may be changed with <tt>MPI_Comm_set_errhandler</tt> (for communicators),
|
||||
<tt>MPI_File_set_errhandler</tt> (for files), and <tt>MPI_Win_set_errhandler</tt> (for
|
||||
RMA windows). The MPI-1 routine <tt>MPI_Errhandler_set</tt> may be used but
|
||||
its use is deprecated. The predefined error handler
|
||||
<tt>MPI_ERRORS_RETURN</tt> may be used to cause error values to be returned.
|
||||
Note that MPI does <em>not</em> guarantee that an MPI program can continue past
|
||||
an error; however, MPI implementations will attempt to continue whenever
|
||||
possible.
|
||||
<P>
|
||||
<DL>
|
||||
<DT><B>MPI_SUCCESS </B> <DD> No error; MPI routine completed successfully.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_ARG </B> <DD> Invalid argument. Some argument is invalid and is not
|
||||
identified by a specific error class (e.g., <tt>MPI_ERR_RANK</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_BUFFER </B> <DD> Invalid buffer pointer. Usually a null buffer where
|
||||
one is not valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COMM </B> <DD> Invalid communicator. A common error is to use a null
|
||||
communicator in a call (not even allowed in <tt>MPI_Comm_rank</tt>).
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_COUNT </B> <DD> Invalid count argument. Count arguments must be
|
||||
non-negative; a count of zero is often valid.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_INFO </B> <DD> Invalid Info
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OP </B> <DD> Invalid operation. MPI operations (objects of type <tt>MPI_Op</tt>)
|
||||
must either be one of the predefined operations (e.g., <tt>MPI_SUM</tt>) or
|
||||
created with <tt>MPI_Op_create</tt>.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_TYPE </B> <DD> Invalid datatype argument. Additionally, this error can
|
||||
occur if an uncommitted MPI_Datatype (see <tt>MPI_Type_commit</tt>) is used
|
||||
in a communication call.
|
||||
</DL>
|
||||
<DL>
|
||||
<DT><B>MPI_ERR_OTHER </B> <DD> Other error; use <tt>MPI_Error_string</tt> to get more information
|
||||
about this error code.
|
||||
</DL>
|
||||
<P>
|
||||
</BODY></HTML>
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user