2016-11-23 02:03:20 +03:00
|
|
|
/* -*- Mode: C; c-basic-offset:4 ; -*- */
|
|
|
|
/*
|
|
|
|
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
|
|
|
|
* University Research and Technology
|
|
|
|
* Corporation. All rights reserved.
|
|
|
|
* Copyright (c) 2004-2007 The University of Tennessee and The University
|
|
|
|
* of Tennessee Research Foundation. All rights
|
|
|
|
* reserved.
|
|
|
|
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
|
|
|
|
* University of Stuttgart. All rights reserved.
|
|
|
|
* Copyright (c) 2004-2005 The Regents of the University of California.
|
|
|
|
* All rights reserved.
|
|
|
|
* Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
|
2017-06-01 22:28:23 +03:00
|
|
|
* Copyright (c) 2009-2017 Cisco Systems, Inc. All rights reserved
|
2017-12-01 09:50:29 +03:00
|
|
|
* Copyright (c) 2015-2017 Research Organization for Information Science
|
2016-11-23 02:03:20 +03:00
|
|
|
* and Technology (RIST). All rights reserved.
|
|
|
|
* Copyright (c) 2016 University of Houston. All rights reserved.
|
fixes for Dave's get/set info code
The expected sequence of events for processing info during object creation
is that if there's an incoming info arg, it is opal_info_dup()ed into the obj
at obj->s_info first. Then interested components register callbacks for
keys they want to know about using opal_infosubscribe_infosubscribe().
Inside info_subscribe_subscribe() the specified callback() is called with
whatever matching k/v is in the object's info, or with the default. The
return string from the callback goes into the new k/v stored in info, and
the input k/v is saved as __IN_<key>/<val>. It's saved the same way
whether the input came from info or whether it was a default. A null return
from the callback indicates an ignored key/val, and no k/v is stored for
it, but an __IN_<key>/<val> is still kept so we still have access to the
original.
At MPI_*_set_info() time, opal_infosubscribe_change_info() is used. That
function calls the registered callbacks for each item in the provided info.
If the callback returns non-null, the info is updated with that k/v, or if
the callback returns null, that key is deleted from info. An __IN_<key>/<val>
is saved either way, and overwrites any previously saved value.
When MPI_*_get_info() is called, opal_info_dup_mpistandard() is used, which
allows relatively easy changes in interpretation of the standard, by looking
at both the <key>/<val> and __IN_<key>/<val> in info. Right now it does
1. includes system extras, eg k/v defaults not expliclty set by the user
2. omits ignored keys
3. shows input values, not callback modifications, eg not the internal values
Currently the callbacks are doing things like
return some_condition ? "true" : "false"
that is, returning static strings that are not to be freed. If the return
strings start becoming more dynamic in the future I don't see how unallocated
strings could support that, so I'd propose a change for the future that
the callback()s registered with info_subscribe_subscribe() do a strdup on
their return, and we change the callers of callback() to free the strings
it returns (there are only two callers).
Rough outline of the smaller changes spread over the less central files:
comm.c
initialize comm->super.s_info to NULL
copy into comm->super.s_info in comm creation calls that provide info
OBJ_RELEASE comm->super.s_info at free time
comm_init.c
initialize comm->super.s_info to NULL
file.c
copy into file->super.s_info if file creation provides info
OBJ_RELEASE file->super.s_info at free time
win.c
copy into win->super.s_info if win creation provides info
OBJ_RELEASE win->super.s_info at free time
comm_get_info.c
file_get_info.c
win_get_info.c
change_info() if there's no info attached (shouldn't happen if callbacks
are registered)
copy the info for the user
The other category of change is generally addressing compiler warnings where
ompi_info_t and opal_info_t were being used a little too interchangably. An
ompi_info_t* contains an opal_info_t*, at &(ompi_info->super)
Also this commit updates the copyrights.
Signed-off-by: Mark Allen <markalle@us.ibm.com>
2017-01-31 04:29:50 +03:00
|
|
|
* Copyright (c) 2016-2017 IBM Corporation. All rights reserved.
|
2016-11-23 02:03:20 +03:00
|
|
|
* $COPYRIGHT$
|
|
|
|
*
|
|
|
|
* Additional copyrights may follow
|
|
|
|
*
|
|
|
|
* $HEADER$
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef OMPI_FILE_H
|
|
|
|
#define OMPI_FILE_H
|
|
|
|
|
|
|
|
#include "ompi_config.h"
|
|
|
|
#include "mpi.h"
|
|
|
|
#include "opal/class/opal_list.h"
|
|
|
|
#include "ompi/errhandler/errhandler.h"
|
|
|
|
#include "opal/threads/mutex.h"
|
Major structural changes to data types: .super infosubscriber
ompi_communicator_t, ompi_win_t, ompi_file_t all have a super class of type opal_infosubscriber_t instead of a base/super type of opal_object_t (in previous code comm used c_base, but file used super). It may be a bit bold to say that being a subscriber of MPI_Info is the foundational piece that ties these three things together, but if you object, then I would prefer to turn infosubscriber into a more general name that encompasses other common features rather than create a different super class. The key here is that we want to be able to pass comm, win and file objects as if they were opal_infosubscriber_t, so that one routine can heandle all 3 types of objects being passed to it.
MPI_INFO_NULL is still an ompi_predefined_info_t type since an MPI_Info is part of ompi but the internal details of the underlying information concept is part of opal.
An ompi_info_t type still exists for exposure to the user, but it is simply a wrapper for the opal object.
Routines such as ompi_info_dup, etc have all been moved to opal_info_dup and related to the opal directory.
Fortran to C translation tables are only used for MPI_Info that is exposed to the application and are therefore part of the ompi_info_t and not the opal_info_t
The data structure changes are primarily in the following files:
communicator/communicator.h
ompi/info/info.h
ompi/win/win.h
ompi/file/file.h
The following new files were created:
opal/util/info.h
opal/util/info.c
opal/util/info_subscriber.h
opal/util/info_subscriber.c
This infosubscriber concept is that communicators, files and windows can have subscribers that subscribe to any changes in the info associated with the comm/file/window. When xxx_set_info is called, the new info is presented to each subscriber who can modify the info in any way they want. The new value is presented to the next subscriber and so on until all subscribers have had a chance to modify the value. Therefore, the order of subscribers can make a difference but we hope that there is generally only one subscriber that cares or modifies any given key/value pair. The final info is then stored and returned by a call to xxx_get_info.
The new model can be seen in the following files:
ompi/mpi/c/comm_get_info.c
ompi/mpi/c/comm_set_info.c
ompi/mpi/c/file_get_info.c
ompi/mpi/c/file_set_info.c
ompi/mpi/c/win_get_info.c
ompi/mpi/c/win_set_info.c
The current subscribers where changed as follows:
mca/io/ompio/io_ompio_file_open.c
mca/io/ompio/io_ompio_module.c
mca/osc/rmda/osc_rdma_component.c (This one actually subscribes to "no_locks")
mca/osc/sm/osc_sm_component.c (This one actually subscribes to "blocking_fence" and "alloc_shared_contig")
Signed-off-by: Mark Allen <markalle@us.ibm.com>
Conflicts:
AUTHORS
ompi/communicator/comm.c
ompi/debuggers/ompi_mpihandles_dll.c
ompi/file/file.c
ompi/file/file.h
ompi/info/info.c
ompi/mca/io/ompio/io_ompio.h
ompi/mca/io/ompio/io_ompio_file_open.c
ompi/mca/io/ompio/io_ompio_file_set_view.c
ompi/mca/osc/pt2pt/osc_pt2pt.h
ompi/mca/sharedfp/addproc/sharedfp_addproc.h
ompi/mca/sharedfp/addproc/sharedfp_addproc_file_open.c
ompi/mca/topo/treematch/topo_treematch_dist_graph_create.c
ompi/mpi/c/lookup_name.c
ompi/mpi/c/publish_name.c
ompi/mpi/c/unpublish_name.c
opal/mca/mpool/base/mpool_base_alloc.c
opal/util/Makefile.am
2016-01-22 20:02:01 +03:00
|
|
|
#include "opal/util/info_subscriber.h"
|
2016-11-23 02:03:20 +03:00
|
|
|
#include "ompi/mca/io/io.h"
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Flags
|
|
|
|
*/
|
|
|
|
#define OMPI_FILE_ISCLOSED 0x00000001
|
|
|
|
#define OMPI_FILE_HIDDEN 0x00000002
|
|
|
|
|
|
|
|
BEGIN_C_DECLS
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Back-end structure for MPI_File
|
|
|
|
*/
|
|
|
|
struct ompi_file_t {
|
|
|
|
/** Base of OBJ_* interface */
|
Major structural changes to data types: .super infosubscriber
ompi_communicator_t, ompi_win_t, ompi_file_t all have a super class of type opal_infosubscriber_t instead of a base/super type of opal_object_t (in previous code comm used c_base, but file used super). It may be a bit bold to say that being a subscriber of MPI_Info is the foundational piece that ties these three things together, but if you object, then I would prefer to turn infosubscriber into a more general name that encompasses other common features rather than create a different super class. The key here is that we want to be able to pass comm, win and file objects as if they were opal_infosubscriber_t, so that one routine can heandle all 3 types of objects being passed to it.
MPI_INFO_NULL is still an ompi_predefined_info_t type since an MPI_Info is part of ompi but the internal details of the underlying information concept is part of opal.
An ompi_info_t type still exists for exposure to the user, but it is simply a wrapper for the opal object.
Routines such as ompi_info_dup, etc have all been moved to opal_info_dup and related to the opal directory.
Fortran to C translation tables are only used for MPI_Info that is exposed to the application and are therefore part of the ompi_info_t and not the opal_info_t
The data structure changes are primarily in the following files:
communicator/communicator.h
ompi/info/info.h
ompi/win/win.h
ompi/file/file.h
The following new files were created:
opal/util/info.h
opal/util/info.c
opal/util/info_subscriber.h
opal/util/info_subscriber.c
This infosubscriber concept is that communicators, files and windows can have subscribers that subscribe to any changes in the info associated with the comm/file/window. When xxx_set_info is called, the new info is presented to each subscriber who can modify the info in any way they want. The new value is presented to the next subscriber and so on until all subscribers have had a chance to modify the value. Therefore, the order of subscribers can make a difference but we hope that there is generally only one subscriber that cares or modifies any given key/value pair. The final info is then stored and returned by a call to xxx_get_info.
The new model can be seen in the following files:
ompi/mpi/c/comm_get_info.c
ompi/mpi/c/comm_set_info.c
ompi/mpi/c/file_get_info.c
ompi/mpi/c/file_set_info.c
ompi/mpi/c/win_get_info.c
ompi/mpi/c/win_set_info.c
The current subscribers where changed as follows:
mca/io/ompio/io_ompio_file_open.c
mca/io/ompio/io_ompio_module.c
mca/osc/rmda/osc_rdma_component.c (This one actually subscribes to "no_locks")
mca/osc/sm/osc_sm_component.c (This one actually subscribes to "blocking_fence" and "alloc_shared_contig")
Signed-off-by: Mark Allen <markalle@us.ibm.com>
Conflicts:
AUTHORS
ompi/communicator/comm.c
ompi/debuggers/ompi_mpihandles_dll.c
ompi/file/file.c
ompi/file/file.h
ompi/info/info.c
ompi/mca/io/ompio/io_ompio.h
ompi/mca/io/ompio/io_ompio_file_open.c
ompi/mca/io/ompio/io_ompio_file_set_view.c
ompi/mca/osc/pt2pt/osc_pt2pt.h
ompi/mca/sharedfp/addproc/sharedfp_addproc.h
ompi/mca/sharedfp/addproc/sharedfp_addproc_file_open.c
ompi/mca/topo/treematch/topo_treematch_dist_graph_create.c
ompi/mpi/c/lookup_name.c
ompi/mpi/c/publish_name.c
ompi/mpi/c/unpublish_name.c
opal/mca/mpool/base/mpool_base_alloc.c
opal/util/Makefile.am
2016-01-22 20:02:01 +03:00
|
|
|
opal_infosubscriber_t super;
|
2016-11-23 02:03:20 +03:00
|
|
|
|
|
|
|
/** Communicator that this file was created with */
|
|
|
|
struct ompi_communicator_t *f_comm;
|
|
|
|
|
|
|
|
/** Filename that this file was created with */
|
|
|
|
char *f_filename;
|
|
|
|
|
|
|
|
/** Amode that this file was created with */
|
|
|
|
int f_amode;
|
|
|
|
|
|
|
|
/** Bit flags */
|
|
|
|
int32_t f_flags;
|
|
|
|
|
|
|
|
/** Index in Fortran <-> C translation array */
|
|
|
|
int f_f_to_c_index;
|
|
|
|
|
|
|
|
/** Error handler. This field does not have the "f_" prefix so
|
|
|
|
that the OMPI_ERRHDL_* macros can find it, regardless of
|
|
|
|
whether it's a comm, window, or file. */
|
|
|
|
struct ompi_errhandler_t *error_handler;
|
|
|
|
|
|
|
|
/** Type of the error handler. This field does not have the "f_"
|
|
|
|
prefix for the same reason as the field error_handler. */
|
|
|
|
ompi_errhandler_type_t errhandler_type;
|
|
|
|
|
|
|
|
/** Indicate what version of the IO component we're using (this
|
|
|
|
indicates what member to look at in the union, below) */
|
|
|
|
mca_io_base_version_t f_io_version;
|
|
|
|
|
|
|
|
/** Mutex to be used to protect access to the selected component
|
|
|
|
on a per file-handle basis */
|
2017-12-01 09:50:29 +03:00
|
|
|
opal_mutex_t f_lock;
|
2016-11-23 02:03:20 +03:00
|
|
|
|
|
|
|
/** The selected component (note that this is a union) -- we need
|
|
|
|
this to add and remove the component from the list of
|
|
|
|
components currently in use by the io framework for
|
|
|
|
progression porpoises. */
|
|
|
|
mca_io_base_components_t f_io_selected_component;
|
|
|
|
|
|
|
|
/** The selected module (note that this is a union) */
|
|
|
|
mca_io_base_modules_t f_io_selected_module;
|
|
|
|
|
|
|
|
/** Allow the selected module to cache data on the file */
|
|
|
|
struct mca_io_base_file_t *f_io_selected_data;
|
|
|
|
};
|
|
|
|
/**
|
|
|
|
* Convenience typedef
|
|
|
|
*/
|
|
|
|
typedef struct ompi_file_t ompi_file_t;
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Padded struct to maintain back compatibiltiy.
|
|
|
|
* See ompi/communicator/communicator.h comments with struct ompi_communicator_t
|
|
|
|
* for full explanation why we chose the following padding construct for predefines.
|
|
|
|
*/
|
2017-06-01 22:28:23 +03:00
|
|
|
#define PREDEFINED_FILE_PAD 1536
|
2016-11-23 02:03:20 +03:00
|
|
|
|
|
|
|
struct ompi_predefined_file_t {
|
|
|
|
struct ompi_file_t file;
|
|
|
|
char padding[PREDEFINED_FILE_PAD - sizeof(ompi_file_t)];
|
|
|
|
};
|
|
|
|
typedef struct ompi_predefined_file_t ompi_predefined_file_t;
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Back-end instances for MPI_FILE_NULL (_addr flavor is for F03 bindings)
|
|
|
|
*/
|
|
|
|
OMPI_DECLSPEC extern ompi_predefined_file_t ompi_mpi_file_null;
|
|
|
|
OMPI_DECLSPEC extern ompi_predefined_file_t *ompi_mpi_file_null_addr;
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Fortran to C conversion table
|
|
|
|
*/
|
|
|
|
extern opal_pointer_array_t ompi_file_f_to_c_table;
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Initialize MPI_File handling.
|
|
|
|
*
|
|
|
|
* @retval OMPI_SUCCESS Always.
|
|
|
|
*
|
|
|
|
* Invoked during ompi_mpi_init().
|
|
|
|
*/
|
|
|
|
int ompi_file_init(void);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Back-end to MPI_FILE_OPEN: create a file handle, select an io
|
|
|
|
* component to use, and have that componet open the file.
|
|
|
|
*
|
|
|
|
* @param comm Communicator
|
|
|
|
* @param filename String filename
|
|
|
|
* @param amode Mode flags
|
|
|
|
* @param info Info
|
|
|
|
* @param fh Output file handle
|
|
|
|
*
|
|
|
|
* @retval OMPI_SUCCESS Upon success
|
|
|
|
* @retval OMPI_ERR* Upon error
|
|
|
|
*
|
|
|
|
* Create a file handle and select an io module to be paired with
|
|
|
|
* it. There is a corresponding ompi_file_close() function; it
|
|
|
|
* mainly calls OBJ_RELEASE() but also does some other error
|
|
|
|
* handling as well.
|
|
|
|
*/
|
|
|
|
int ompi_file_open(struct ompi_communicator_t *comm, const char *filename,
|
Major structural changes to data types: .super infosubscriber
ompi_communicator_t, ompi_win_t, ompi_file_t all have a super class of type opal_infosubscriber_t instead of a base/super type of opal_object_t (in previous code comm used c_base, but file used super). It may be a bit bold to say that being a subscriber of MPI_Info is the foundational piece that ties these three things together, but if you object, then I would prefer to turn infosubscriber into a more general name that encompasses other common features rather than create a different super class. The key here is that we want to be able to pass comm, win and file objects as if they were opal_infosubscriber_t, so that one routine can heandle all 3 types of objects being passed to it.
MPI_INFO_NULL is still an ompi_predefined_info_t type since an MPI_Info is part of ompi but the internal details of the underlying information concept is part of opal.
An ompi_info_t type still exists for exposure to the user, but it is simply a wrapper for the opal object.
Routines such as ompi_info_dup, etc have all been moved to opal_info_dup and related to the opal directory.
Fortran to C translation tables are only used for MPI_Info that is exposed to the application and are therefore part of the ompi_info_t and not the opal_info_t
The data structure changes are primarily in the following files:
communicator/communicator.h
ompi/info/info.h
ompi/win/win.h
ompi/file/file.h
The following new files were created:
opal/util/info.h
opal/util/info.c
opal/util/info_subscriber.h
opal/util/info_subscriber.c
This infosubscriber concept is that communicators, files and windows can have subscribers that subscribe to any changes in the info associated with the comm/file/window. When xxx_set_info is called, the new info is presented to each subscriber who can modify the info in any way they want. The new value is presented to the next subscriber and so on until all subscribers have had a chance to modify the value. Therefore, the order of subscribers can make a difference but we hope that there is generally only one subscriber that cares or modifies any given key/value pair. The final info is then stored and returned by a call to xxx_get_info.
The new model can be seen in the following files:
ompi/mpi/c/comm_get_info.c
ompi/mpi/c/comm_set_info.c
ompi/mpi/c/file_get_info.c
ompi/mpi/c/file_set_info.c
ompi/mpi/c/win_get_info.c
ompi/mpi/c/win_set_info.c
The current subscribers where changed as follows:
mca/io/ompio/io_ompio_file_open.c
mca/io/ompio/io_ompio_module.c
mca/osc/rmda/osc_rdma_component.c (This one actually subscribes to "no_locks")
mca/osc/sm/osc_sm_component.c (This one actually subscribes to "blocking_fence" and "alloc_shared_contig")
Signed-off-by: Mark Allen <markalle@us.ibm.com>
Conflicts:
AUTHORS
ompi/communicator/comm.c
ompi/debuggers/ompi_mpihandles_dll.c
ompi/file/file.c
ompi/file/file.h
ompi/info/info.c
ompi/mca/io/ompio/io_ompio.h
ompi/mca/io/ompio/io_ompio_file_open.c
ompi/mca/io/ompio/io_ompio_file_set_view.c
ompi/mca/osc/pt2pt/osc_pt2pt.h
ompi/mca/sharedfp/addproc/sharedfp_addproc.h
ompi/mca/sharedfp/addproc/sharedfp_addproc_file_open.c
ompi/mca/topo/treematch/topo_treematch_dist_graph_create.c
ompi/mpi/c/lookup_name.c
ompi/mpi/c/publish_name.c
ompi/mpi/c/unpublish_name.c
opal/mca/mpool/base/mpool_base_alloc.c
opal/util/Makefile.am
2016-01-22 20:02:01 +03:00
|
|
|
int amode, struct opal_info_t *info,
|
2016-11-23 02:03:20 +03:00
|
|
|
ompi_file_t **fh);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Atomicly set a name on a file handle.
|
|
|
|
*
|
|
|
|
* @param file MPI_File handle to set the name on
|
|
|
|
* @param name NULL-terminated string to use
|
|
|
|
*
|
|
|
|
* @returns OMPI_SUCCESS Always.
|
|
|
|
*
|
|
|
|
* At most (MPI_MAX_OBJECT_NAME-1) characters will be copied over to
|
|
|
|
* the file name's name. This function is performed atomically -- a
|
|
|
|
* lock is used to ensure that there are not multiple writers to the
|
|
|
|
* name to ensure that we don't end up with an erroneous name (e.g.,
|
|
|
|
* a name without a \0 at the end). After invoking this function,
|
|
|
|
* ompi_file_is_name_set() will return true.
|
|
|
|
*/
|
|
|
|
int ompi_file_set_name(ompi_file_t *file, char *name);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Back-end to MPI_FILE_CLOSE: destroy an ompi_file_t handle and
|
|
|
|
* close the file.
|
|
|
|
*
|
|
|
|
* @param file Pointer to ompi_file_t
|
|
|
|
*
|
|
|
|
* @returns OMPI_SUCCESS Always.
|
|
|
|
*
|
|
|
|
* This is the preferred mechanism for freeing an ompi_file_t.
|
|
|
|
* Although the main action that it performs is OBJ_RELEASE(), it
|
|
|
|
* also does some additional handling for error checking, etc.
|
|
|
|
*/
|
|
|
|
int ompi_file_close(ompi_file_t **file);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Tear down MPI_File handling.
|
|
|
|
*
|
|
|
|
* @retval OMPI_SUCCESS Always.
|
|
|
|
*
|
|
|
|
* Invoked during ompi_mpi_finalize().
|
|
|
|
*/
|
|
|
|
int ompi_file_finalize(void);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Check to see if an MPI_File handle is valid.
|
|
|
|
*
|
|
|
|
* @param file The MPI file handle
|
|
|
|
*
|
|
|
|
* @retval true If the file handle is not valid
|
|
|
|
* @retval false If the file handle is valid
|
|
|
|
*
|
|
|
|
* This is a convenience function, mainly for error checking in
|
|
|
|
* top-level MPI API functions.
|
|
|
|
*/
|
|
|
|
static inline bool ompi_file_invalid(ompi_file_t *file)
|
|
|
|
{
|
|
|
|
return (NULL == file ||
|
|
|
|
&ompi_mpi_file_null.file == file ||
|
|
|
|
0 != (file->f_flags & OMPI_FILE_ISCLOSED));
|
|
|
|
}
|
|
|
|
|
|
|
|
END_C_DECLS
|
|
|
|
|
|
|
|
#endif /* OMPI_FILE_H */
|