ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
828ae26d90
openmpi/ompi/file/file.c
Jeff Squyres e7ecd56bd2 This commit represents a bunch of work on a Mercurial side branch. As 
such, the commit message back to the master SVN repository is fairly
long.

= ORTE Job-Level Output Messages =

Add two new interfaces that should be used for all new code throughout
the ORTE and OMPI layers (we already make the search-and-replace on
the existing ORTE / OMPI layers):

 * orte_output(): (and corresponding friends ORTE_OUTPUT,
   orte_output_verbose, etc.)  This function sends the output directly
   to the HNP for processing as part of a job-specific output
   channel.  It supports all the same outputs as opal_output()
   (syslog, file, stdout, stderr), but for stdout/stderr, the output
   is sent to the HNP for processing and output.  More on this below.
 * orte_show_help(): This function is a drop-in-replacement for
   opal_show_help(), with two differences in functionality:
   1. the rendered text help message output is sent to the HNP for
      display (rather than outputting directly into the process' stderr
      stream)
   1. the HNP detects duplicate help messages and does not display them
      (so that you don't see the same error message N times, once from
      each of your N MPI processes); instead, it counts "new" instances
      of the help message and displays a message every ~5 seconds when
      there are new ones ("I got X new copies of the help message...")

opal_show_help and opal_output still exist, but they only output in
the current process.  The intent for the new orte_* functions is that
they can apply job-level intelligence to the output.  As such, we
recommend that all new ORTE and OMPI code use the new orte_*
functions, not thei opal_* functions.

=== New code ===

For ORTE and OMPI programmers, here's what you need to do differently
in new code:

 * Do not include opal/util/show_help.h or opal/util/output.h.
   Instead, include orte/util/output.h (this one header file has
   declarations for both the orte_output() series of functions and
   orte_show_help()).
 * Effectively s/opal_output/orte_output/gi throughout your code.
   Note that orte_output_open() takes a slightly different argument
   list (as a way to pass data to the filtering stream -- see below),
   so you if explicitly call opal_output_open(), you'll need to
   slightly adapt to the new signature of orte_output_open().
 * Literally s/opal_show_help/orte_show_help/.  The function signature
   is identical.

=== Notes ===

 * orte_output'ing to stream 0 will do similar to what
   opal_output'ing did, so leaving a hard-coded "0" as the first
   argument is safe.
 * For systems that do not use ORTE's RML or the HNP, the effect of
   orte_output_* and orte_show_help will be identical to their opal
   counterparts (the additional information passed to
   orte_output_open() will be lost!).  Indeed, the orte_* functions
   simply become trivial wrappers to their opal_* counterparts.  Note
   that we have not tested this; the code is simple but it is quite
   possible that we mucked something up.

= Filter Framework =

Messages sent view the new orte_* functions described above and
messages output via the IOF on the HNP will now optionally be passed
through a new "filter" framework before being output to
stdout/stderr.  The "filter" OPAL MCA framework is intended to allow
preprocessing to messages before they are sent to their final
destinations.  The first component that was written in the filter
framework was to create an XML stream, segregating all the messages
into different XML tags, etc.  This will allow 3rd party tools to read
the stdout/stderr from the HNP and be able to know exactly what each
text message is (e.g., a help message, another OMPI infrastructure
message, stdout from the user process, stderr from the user process,
etc.).

Filtering is not active by default.  Filter components must be
specifically requested, such as:

{{{
$ mpirun --mca filter xml ...
}}}

There can only be one filter component active.

= New MCA Parameters =

The new functionality described above introduces two new MCA
parameters:

 * '''orte_base_help_aggregate''': Defaults to 1 (true), meaning that
   help messages will be aggregated, as described above.  If set to 0,
   all help messages will be displayed, even if they are duplicates
   (i.e., the original behavior).
 * '''orte_base_show_output_recursions''': An MCA parameter to help
   debug one of the known issues, described below.  It is likely that
   this MCA parameter will disappear before v1.3 final.

= Known Issues =

 * The XML filter component is not complete.  The current output from
   this component is preliminary and not real XML.  A bit more work
   needs to be done to configure.m4 search for an appropriate XML
   library/link it in/use it at run time.
 * There are possible recursion loops in the orte_output() and
   orte_show_help() functions -- e.g., if RML send calls orte_output()
   or orte_show_help().  We have some ideas how to fix these, but
   figured that it was ok to commit before feature freeze with known
   issues.  The code currently contains sub-optimal workarounds so
   that this will not be a problem, but it would be good to actually
   solve the problem rather than have hackish workarounds before v1.3 final.

This commit was SVN r18434.
2008-05-13 20:00:55 +00:00

327 строки
8.5 KiB
C
Исходник Ответственный История

/* -*- 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.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include "ompi/file/file.h"
#include "opal/class/opal_list.h"
#include "ompi/runtime/params.h"
#include "ompi/mca/io/base/base.h"
#include "ompi/info/info.h"
#include "orte/util/output.h"
/*
* Table for Fortran <-> C file handle conversion
*/
opal_pointer_array_t ompi_file_f_to_c_table;
/*
* MPI_FILE_NULL
*/
ompi_file_t ompi_mpi_file_null;
/*
* Local functions
*/
static void file_constructor(ompi_file_t *obj);
static void file_destructor(ompi_file_t *obj);
/*
* Class instance for ompi_file_t
*/
OBJ_CLASS_INSTANCE(ompi_file_t,
opal_object_t,
file_constructor,
file_destructor);
/*
* Initialize file handling bookeeping
*/
int ompi_file_init(void)
{
/* Setup file array */
OBJ_CONSTRUCT(&ompi_file_f_to_c_table, opal_pointer_array_t);
if( OPAL_SUCCESS != opal_pointer_array_init(&ompi_file_f_to_c_table, 0,
OMPI_FORTRAN_HANDLE_MAX, 64) ) {
return OMPI_ERROR;
}
/* Setup MPI_FILE_NULL. Note that it will have the default error
handler of MPI_ERRORS_RETURN, per MPI-2:9.7 (p265). */
OBJ_CONSTRUCT(&ompi_mpi_file_null, ompi_file_t);
ompi_mpi_file_null.f_comm = &ompi_mpi_comm_null;
OBJ_RETAIN(ompi_mpi_file_null.f_comm);
ompi_mpi_file_null.f_f_to_c_index = 0;
opal_pointer_array_set_item(&ompi_file_f_to_c_table, 0,
&ompi_mpi_file_null);
/* All done */
return OMPI_SUCCESS;
}
/*
* Back end to MPI_FILE_OPEN
*/
int ompi_file_open(struct ompi_communicator_t *comm, char *filename,
int amode, struct ompi_info_t *info, ompi_file_t **fh)
{
int ret;
ompi_file_t *file;
file = OBJ_NEW(ompi_file_t);
if (NULL == file) {
return OMPI_ERR_OUT_OF_RESOURCE;
}
/* Save the params */
file->f_comm = comm;
OBJ_RETAIN(comm);
if (MPI_INFO_NULL != info) {
if(NULL == file->f_info) {
file->f_info = OBJ_NEW(ompi_info_t);
}
if (OMPI_SUCCESS != (ret = ompi_info_dup(info, &file->f_info))) {
OBJ_RELEASE(file);
return ret;
}
} else {
file->f_info = MPI_INFO_NULL;
OBJ_RETAIN(MPI_INFO_NULL);
}
file->f_amode = amode;
file->f_filename = strdup(filename);
if (NULL == file->f_filename) {
OBJ_RELEASE(file);
return OMPI_ERR_OUT_OF_RESOURCE;
}
/* Select a module and actually open the file */
if (OMPI_SUCCESS != (ret = mca_io_base_file_select(file, NULL))) {
OBJ_RELEASE(file);
return ret;
}
/* All done */
*fh = file;
return OMPI_SUCCESS;
}
/*
* Back end to MPI_FILE_CLOSE.
*/
int ompi_file_close(ompi_file_t **file)
{
(*file)->f_flags |= OMPI_FILE_ISCLOSED;
mca_io_base_component_del(&((*file)->f_io_selected_component));
mca_io_base_request_return(*file);
OBJ_RELEASE(*file);
*file = &ompi_mpi_file_null;
return OMPI_SUCCESS;
}
/*
* Shut down the MPI_File bookkeeping
*/
int ompi_file_finalize(void)
{
int i, max;
size_t num_unnamed;
ompi_file_t *file;
/* Release MPI_FILE_NULL. Do this so that we don't get a bogus leak
report on it. Plus, it's statically allocated, so we don't want
to call OBJ_RELEASE on it. */
OBJ_DESTRUCT(&ompi_mpi_file_null);
opal_pointer_array_set_item(&ompi_file_f_to_c_table, 0, NULL);
/* Iterate through all the file handles and destroy them. Note
that this also takes care of destroying MPI_FILE_NULL. */
max = opal_pointer_array_get_size(&ompi_file_f_to_c_table);
for (num_unnamed = i = 0; i < max; ++i) {
file = (ompi_file_t *)opal_pointer_array_get_item(&ompi_file_f_to_c_table, i);
/* If the file was closed but still exists because the user
told us to never free handles, then do an OBJ_RELEASE it
and all is well. Then get the value again and see if it's
actually been freed. */
if (NULL != file && ompi_debug_no_free_handles &&
0 == (file->f_flags & OMPI_FILE_ISCLOSED)) {
OBJ_RELEASE(file);
file = (ompi_file_t *)opal_pointer_array_get_item(&ompi_file_f_to_c_table, i);
}
if (NULL != file) {
/* If the user wanted warnings about MPI object leaks,
print out a message */
if (ompi_debug_show_handle_leaks) {
++num_unnamed;
}
OBJ_RELEASE(file);
}
/* Don't bother setting each element back down to NULL; it
would just take a lot of thread locks / unlocks and since
we're destroying everything, it isn't worth it */
}
if (num_unnamed > 0) {
orte_output(0, "WARNING: %lu unnamed MPI_File handles still allocated at MPI_FINALIZE", (unsigned long)num_unnamed);
}
OBJ_DESTRUCT(&ompi_file_f_to_c_table);
/* All done */
return OMPI_SUCCESS;
}
/*
* Constructor
*/
static void file_constructor(ompi_file_t *file)
{
/* Initialize the MPI_FILE_OPEN params */
file->f_comm = NULL;
file->f_filename = NULL;
file->f_amode = 0;
file->f_info = NULL;
/* Initialize flags */
file->f_flags = 0;
/* Initialize the fortran <--> C translation index */
file->f_f_to_c_index = opal_pointer_array_add(&ompi_file_f_to_c_table,
file);
/* Initialize the error handler. Per MPI-2:9.7 (p265), the
default error handler on file handles is the error handler on
MPI_FILE_NULL, which starts out as MPI_ERRORS_RETURN (but can
be changed by invoking MPI_FILE_SET_ERRHANDLER on
MPI_FILE_NULL). */
file->errhandler_type = OMPI_ERRHANDLER_TYPE_FILE;
if (file != &ompi_mpi_file_null) {
file->error_handler = ompi_mpi_file_null.error_handler;
} else {
file->error_handler = &ompi_mpi_errors_return;
}
OBJ_RETAIN(file->error_handler);
/* Initialize the module */
file->f_io_version = MCA_IO_BASE_V_NONE;
memset(&(file->f_io_selected_module), 0,
sizeof(file->f_io_selected_module));
file->f_io_selected_data = NULL;
/* Construct the io request freelist */
OBJ_CONSTRUCT(&file->f_io_requests, opal_list_t);
/* If the user doesn't want us to ever free it, then add an extra
RETAIN here */
if (ompi_debug_no_free_handles) {
OBJ_RETAIN(&(file->super));
}
}
/*
* Destructor
*/
static void file_destructor(ompi_file_t *file)
{
/* Finalize the module */
switch (file->f_io_version) {
case MCA_IO_BASE_V_1_0_0:
file->f_io_selected_module.v1_0_0.io_module_file_close(file);
break;
default:
/* Should never get here */
break;
}
/* Finalize the data members */
if (NULL != file->f_comm) {
OBJ_RELEASE(file->f_comm);
#if OMPI_ENABLE_DEBUG
file->f_comm = NULL;
#endif
}
if (NULL != file->f_filename) {
free(file->f_filename);
#if OMPI_ENABLE_DEBUG
file->f_filename = NULL;
#endif
}
if (NULL != file->error_handler) {
OBJ_RELEASE(file->error_handler);
#if OMPI_ENABLE_DEBUG
file->error_handler = NULL;
#endif
}
if (NULL != file->f_info) {
OBJ_RELEASE(file->f_info);
#if OMPI_ENABLE_DEBUG
file->f_info = NULL;
#endif
}
/* Destruct the io request freelist */
OBJ_DESTRUCT(&file->f_io_requests);
/* Reset the f_to_c table entry */
if (MPI_UNDEFINED != file->f_f_to_c_index &&
NULL != opal_pointer_array_get_item(&ompi_file_f_to_c_table,
file->f_f_to_c_index)) {
opal_pointer_array_set_item(&ompi_file_f_to_c_table,
file->f_f_to_c_index, NULL);
}
}
Показать git blame Копировать постоянную ссылку