1
1
openmpi/opal/runtime/opal_init.c
Josh Hursey e12ca48cd9 A number of C/R enhancements per RFC below:
http://www.open-mpi.org/community/lists/devel/2010/07/8240.php

Documentation:
  http://osl.iu.edu/research/ft/

Major Changes: 
-------------- 
 * Added C/R-enabled Debugging support. 
   Enabled with the --enable-crdebug flag. See the following website for more information: 
   http://osl.iu.edu/research/ft/crdebug/ 
 * Added Stable Storage (SStore) framework for checkpoint storage 
   * 'central' component does a direct to central storage save 
   * 'stage' component stages checkpoints to central storage while the application continues execution. 
     * 'stage' supports offline compression of checkpoints before moving (sstore_stage_compress) 
     * 'stage' supports local caching of checkpoints to improve automatic recovery (sstore_stage_caching) 
 * Added Compression (compress) framework to support 
 * Add two new ErrMgr recovery policies 
   * {{{crmig}}} C/R Process Migration 
   * {{{autor}}} C/R Automatic Recovery 
 * Added the {{{ompi-migrate}}} command line tool to support the {{{crmig}}} ErrMgr component 
 * Added CR MPI Ext functions (enable them with {{{--enable-mpi-ext=cr}}} configure option) 
   * {{{OMPI_CR_Checkpoint}}} (Fixes trac:2342) 
   * {{{OMPI_CR_Restart}}} 
   * {{{OMPI_CR_Migrate}}} (may need some more work for mapping rules) 
   * {{{OMPI_CR_INC_register_callback}}} (Fixes trac:2192) 
   * {{{OMPI_CR_Quiesce_start}}} 
   * {{{OMPI_CR_Quiesce_checkpoint}}} 
   * {{{OMPI_CR_Quiesce_end}}} 
   * {{{OMPI_CR_self_register_checkpoint_callback}}} 
   * {{{OMPI_CR_self_register_restart_callback}}} 
   * {{{OMPI_CR_self_register_continue_callback}}} 
 * The ErrMgr predicted_fault() interface has been changed to take an opal_list_t of ErrMgr defined types. This will allow us to better support a wider range of fault prediction services in the future. 
 * Add a progress meter to: 
   * FileM rsh (filem_rsh_process_meter) 
   * SnapC full (snapc_full_progress_meter) 
   * SStore stage (sstore_stage_progress_meter) 
 * Added 2 new command line options to ompi-restart 
   * --showme : Display the full command line that would have been exec'ed. 
   * --mpirun_opts : Command line options to pass directly to mpirun. (Fixes trac:2413) 
 * Deprecated some MCA params: 
   * crs_base_snapshot_dir deprecated, use sstore_stage_local_snapshot_dir 
   * snapc_base_global_snapshot_dir deprecated, use sstore_base_global_snapshot_dir 
   * snapc_base_global_shared deprecated, use sstore_stage_global_is_shared 
   * snapc_base_store_in_place deprecated, replaced with different components of SStore 
   * snapc_base_global_snapshot_ref deprecated, use sstore_base_global_snapshot_ref 
   * snapc_base_establish_global_snapshot_dir deprecated, never well supported 
   * snapc_full_skip_filem deprecated, use sstore_stage_skip_filem 

Minor Changes: 
-------------- 
 * Fixes trac:1924 : {{{ompi-restart}}} now recognizes path prefixed checkpoint handles and does the right thing. 
 * Fixes trac:2097 : {{{ompi-info}}} should now report all available CRS components 
 * Fixes trac:2161 : Manual checkpoint movement. A user can 'mv' a checkpoint directory from the original location to another and still restart from it. 
 * Fixes trac:2208 : Honor various TMPDIR varaibles instead of forcing {{{/tmp}}} 
 * Move {{{ompi_cr_continue_like_restart}}} to {{{orte_cr_continue_like_restart}}} to be more flexible in where this should be set. 
 * opal_crs_base_metadata_write* functions have been moved to SStore to support a wider range of metadata handling functionality. 
 * Cleanup the CRS framework and components to work with the SStore framework. 
 * Cleanup the SnapC framework and components to work with the SStore framework (cleans up these code paths considerably). 
 * Add 'quiesce' hook to CRCP for a future enhancement. 
 * We now require a BLCR version that supports {{{cr_request_file()}}} or {{{cr_request_checkpoint()}}} in order to make the code more maintainable. Note that {{{cr_request_file}}} has been deprecated since 0.7.0, so we prefer to use {{{cr_request_checkpoint()}}}. 
 * Add optional application level INC callbacks (registered through the CR MPI Ext interface). 
 * Increase the {{{opal_cr_thread_sleep_wait}}} parameter to 1000 microseconds to make the C/R thread less aggressive. 
 * {{{opal-restart}}} now looks for cache directories before falling back on stable storage when asked. 
 * {{{opal-restart}}} also support local decompression before restarting 
 * {{{orte-checkpoint}}} now uses the SStore framework to work with the metadata 
 * {{{orte-restart}}} now uses the SStore framework to work with the metadata 
 * Remove the {{{orte-restart}}} preload option. This was removed since the user only needs to select the 'stage' component in order to support this functionality. 
 * Since the '-am' parameter is saved in the metadata, {{{ompi-restart}}} no longer hard codes {{{-am ft-enable-cr}}}. 
 * Fix {{{hnp}}} ErrMgr so that if a previous component in the stack has 'fixed' the problem, then it should be skipped. 
 * Make sure to decrement the number of 'num_local_procs' in the orted when one goes away. 
 * odls now checks the SStore framework to see if it needs to load any checkpoint files before launching (to support 'stage'). This separates the SStore logic from the --preload-[binary|files] options. 
 * Add unique IDs to the named pipes established between the orted and the app in SnapC. This is to better support migration and automatic recovery activities. 
 * Improve the checks for 'already checkpointing' error path. 
 * A a recovery output timer, to show how long it takes to restart a job 
 * Do a better job of cleaning up the old session directory on restart. 
 * Add a local module to the autor and crmig ErrMgr components. These small modules prevent the 'orted' component from attempting a local recovery (Which does not work for MPI apps at the moment) 
 * Add a fix for bounding the checkpointable region between MPI_Init and MPI_Finalize. 

This commit was SVN r23587.

The following Trac tickets were found above:
  Ticket 1924 --> https://svn.open-mpi.org/trac/ompi/ticket/1924
  Ticket 2097 --> https://svn.open-mpi.org/trac/ompi/ticket/2097
  Ticket 2161 --> https://svn.open-mpi.org/trac/ompi/ticket/2161
  Ticket 2192 --> https://svn.open-mpi.org/trac/ompi/ticket/2192
  Ticket 2208 --> https://svn.open-mpi.org/trac/ompi/ticket/2208
  Ticket 2342 --> https://svn.open-mpi.org/trac/ompi/ticket/2342
  Ticket 2413 --> https://svn.open-mpi.org/trac/ompi/ticket/2413
2010-08-10 20:51:11 +00:00

468 строки
14 KiB
C

/*
* Copyright (c) 2004-2010 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2005 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) 2007-2010 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2007 Sun Microsystems, Inc. All rights reserved.
* Copyright (c) 2009 Oak Ridge National Labs. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
/** @file **/
#include "opal_config.h"
#include "opal/util/malloc.h"
#include "opal/util/output.h"
#include "opal/util/trace.h"
#include "opal/util/show_help.h"
#include "opal/memoryhooks/memory.h"
#include "opal/mca/base/base.h"
#include "opal/runtime/opal.h"
#include "opal/util/net.h"
#include "opal/datatype/opal_datatype.h"
#include "opal/mca/installdirs/base/base.h"
#include "opal/mca/memory/base/base.h"
#include "opal/mca/memcpy/base/base.h"
#include "opal/mca/paffinity/base/base.h"
#include "opal/mca/timer/base/base.h"
#include "opal/mca/memchecker/base/base.h"
#include "opal/dss/dss.h"
#include "opal/mca/carto/base/base.h"
#if OPAL_ENABLE_FT_CR == 1
#include "opal/mca/compress/base/base.h"
#endif
#include "opal/runtime/opal_cr.h"
#include "opal/mca/crs/base/base.h"
#include "opal/runtime/opal_progress.h"
#include "opal/event/event.h"
#include "opal/mca/backtrace/base/base.h"
#include "opal/constants.h"
#include "opal/util/error.h"
#include "opal/util/stacktrace.h"
#include "opal/util/keyval_parse.h"
#include "opal/util/sys_limits.h"
#include "opal/util/opal_sos.h"
#if OPAL_CC_USE_PRAGMA_IDENT
#pragma ident OPAL_IDENT_STRING
#elif OPAL_CC_USE_IDENT
#ident OPAL_IDENT_STRING
#endif
const char opal_version_string[] = OPAL_IDENT_STRING;
int opal_initialized = 0;
int opal_util_initialized = 0;
bool opal_profile = false;
char *opal_profile_file = NULL;
int opal_cache_line_size;
static const char *
opal_err2str(int errnum)
{
const char *retval;
switch (OPAL_SOS_GET_ERROR_CODE(errnum)) {
case OPAL_SUCCESS:
retval = "Success";
break;
case OPAL_ERROR:
retval = "Error";
break;
case OPAL_ERR_OUT_OF_RESOURCE:
retval = "Out of resource";
break;
case OPAL_ERR_TEMP_OUT_OF_RESOURCE:
retval = "Temporarily out of resource";
break;
case OPAL_ERR_RESOURCE_BUSY:
retval = "Resource busy";
break;
case OPAL_ERR_BAD_PARAM:
retval = "Bad parameter";
break;
case OPAL_ERR_FATAL:
retval = "Fatal";
break;
case OPAL_ERR_NOT_IMPLEMENTED:
retval = "Not implemented";
break;
case OPAL_ERR_NOT_SUPPORTED:
retval = "Not supported";
break;
case OPAL_ERR_INTERUPTED:
retval = "Interupted";
break;
case OPAL_ERR_WOULD_BLOCK:
retval = "Would block";
break;
case OPAL_ERR_IN_ERRNO:
retval = "In errno";
break;
case OPAL_ERR_UNREACH:
retval = "Unreachable";
break;
case OPAL_ERR_NOT_FOUND:
retval = "Not found";
break;
case OPAL_EXISTS:
retval = "Exists";
break;
case OPAL_ERR_TIMEOUT:
retval = "Timeout";
break;
case OPAL_ERR_NOT_AVAILABLE:
retval = "Not available";
break;
case OPAL_ERR_PERM:
retval = "No permission";
break;
case OPAL_ERR_VALUE_OUT_OF_BOUNDS:
retval = "Value out of bounds";
break;
case OPAL_ERR_FILE_READ_FAILURE:
retval = "File read failure";
break;
case OPAL_ERR_FILE_WRITE_FAILURE:
retval = "File write failure";
break;
case OPAL_ERR_FILE_OPEN_FAILURE:
retval = "File open failure";
break;
case OPAL_ERR_PACK_MISMATCH:
retval = "Pack data mismatch";
break;
case OPAL_ERR_PACK_FAILURE:
retval = "Data pack failed";
break;
case OPAL_ERR_UNPACK_FAILURE:
retval = "Data unpack failed";
break;
case OPAL_ERR_UNPACK_INADEQUATE_SPACE:
retval = "Data unpack had inadequate space";
break;
case OPAL_ERR_UNPACK_READ_PAST_END_OF_BUFFER:
retval = "Data unpack would read past end of buffer";
break;
case OPAL_ERR_OPERATION_UNSUPPORTED:
retval = "Requested operation is not supported on referenced data type";
break;
case OPAL_ERR_UNKNOWN_DATA_TYPE:
retval = "Unknown data type";
break;
case OPAL_ERR_BUFFER:
retval = "Buffer type (described vs non-described) mismatch - operation not allowed";
break;
case OPAL_ERR_DATA_TYPE_REDEF:
retval = "Attempt to redefine an existing data type";
break;
case OPAL_ERR_DATA_OVERWRITE_ATTEMPT:
retval = "Attempt to overwrite a data value";
break;
case OPAL_ERR_MODULE_NOT_FOUND:
retval = "Framework requires at least one active module, but none found";
break;
case OPAL_ERR_TOPO_SLOT_LIST_NOT_SUPPORTED:
retval = "OS topology does not support slot_list process affinity";
break;
case OPAL_ERR_TOPO_SOCKET_NOT_SUPPORTED:
retval = "Could not obtain socket topology information";
break;
case OPAL_ERR_TOPO_CORE_NOT_SUPPORTED:
retval = "Could not obtain core topology information";
break;
case OPAL_ERR_NOT_ENOUGH_SOCKETS:
retval = "Not enough sockets to meet request";
break;
case OPAL_ERR_NOT_ENOUGH_CORES:
retval = "Not enough cores to meet request";
break;
case OPAL_ERR_INVALID_PHYS_CPU:
retval = "Invalid physical cpu number returned";
break;
case OPAL_ERR_MULTIPLE_AFFINITIES:
retval = "Multiple methods for assigning process affinity were specified";
break;
case OPAL_ERR_SLOT_LIST_RANGE:
retval = "Provided slot_list range is invalid";
break;
default:
retval = NULL;
}
return retval;
}
int
opal_init_util(int* pargc, char*** pargv)
{
int ret;
char *error = NULL;
if( ++opal_util_initialized != 1 ) {
if( opal_util_initialized < 1 ) {
return OPAL_ERROR;
}
return OPAL_SUCCESS;
}
/* JMS See note in runtime/opal.h -- this is temporary; to be
replaced with real hwloc information soon (in trunk/v1.5 and
beyond, only). This *used* to be a #define, so it's important
to define it very early. */
opal_cache_line_size = 128;
/* initialize the memory allocator */
opal_malloc_init();
/* initialize the output system */
opal_output_init();
/* initialize install dirs code */
if (OPAL_SUCCESS != (ret = opal_installdirs_base_open())) {
fprintf(stderr, "opal_installdirs_base_open() failed -- process will likely abort (%s:%d, returned %d instead of OPAL_INIT)\n",
__FILE__, __LINE__, ret);
return ret;
}
/* initialize the help system */
opal_show_help_init();
/* initialize the OPAL SOS system */
opal_sos_init();
/* register handler for errnum -> string converstion */
if (OPAL_SUCCESS !=
(ret = opal_error_register("OPAL",
OPAL_ERR_BASE, OPAL_ERR_MAX, opal_err2str))) {
error = "opal_error_register";
goto return_error;
}
/* init the trace function */
opal_trace_init();
/* keyval lex-based parser */
if (OPAL_SUCCESS != (ret = opal_util_keyval_parse_init())) {
error = "opal_util_keyval_parse_init";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_net_init())) {
error = "opal_net_init";
goto return_error;
}
/* Setup the parameter system */
if (OPAL_SUCCESS != (ret = mca_base_param_init())) {
error = "mca_base_param_init";
goto return_error;
}
/* register params for opal */
if (OPAL_SUCCESS != (ret = opal_register_params())) {
error = "opal_register_params";
goto return_error;
}
/* pretty-print stack handlers */
if (OPAL_SUCCESS != (ret = opal_util_register_stackhandlers())) {
error = "opal_util_register_stackhandlers";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_util_init_sys_limits())) {
error = "opal_util_init_sys_limits";
goto return_error;
}
/* initialize the datatype engine */
if (OPAL_SUCCESS != (ret = opal_datatype_init ())) {
error = "opal_datatype_init";
goto return_error;
}
/* Initialize the data storage service. */
if (OPAL_SUCCESS != (ret = opal_dss_open())) {
error = "opal_dss_open";
goto return_error;
}
return OPAL_SUCCESS;
return_error:
opal_show_help( "help-opal-runtime.txt",
"opal_init:startup:internal-failure", true,
error, ret );
return ret;
}
int
opal_init(int* pargc, char*** pargv)
{
int ret;
char *error = NULL;
if( ++opal_initialized != 1 ) {
if( opal_initialized < 1 ) {
return OPAL_ERROR;
}
return OPAL_SUCCESS;
}
/* initialize util code */
if (OPAL_SUCCESS != (ret = opal_init_util(pargc, pargv))) {
return ret;
}
/* initialize the mca */
if (OPAL_SUCCESS != (ret = mca_base_open())) {
error = "mca_base_open";
goto return_error;
}
/* open the processor affinity base */
if (OPAL_SUCCESS != (ret = opal_paffinity_base_open())) {
error = "opal_paffinity_base_open";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_paffinity_base_select())) {
error = "opal_paffinity_base_select";
goto return_error;
}
/* the memcpy component should be one of the first who get
* loaded in order to make sure we ddo have all the available
* versions of memcpy correctly configured.
*/
if( OPAL_SUCCESS != (ret = opal_memcpy_base_open()) ) {
error = "opal_memcpy_base_open";
goto return_error;
}
/* open the memory manager components. Memory hooks may be
triggered before this (any time after mem_free_init(),
actually). This is a hook available for memory manager hooks
without good initialization routine support */
if (OPAL_SUCCESS != (ret = opal_memory_base_open())) {
error = "opal_memory_base_open";
goto return_error;
}
/* initialize the memory manager / tracker */
if (OPAL_SUCCESS != (ret = opal_mem_hooks_init())) {
error = "opal_mem_hooks_init";
goto return_error;
}
/* initialize the memory checker, to allow early support for annotation */
if (OPAL_SUCCESS != (ret = opal_memchecker_base_open())) {
error = "opal_memchecker_base_open";
goto return_error;
}
/* select the memory checker */
if (OPAL_SUCCESS != (ret = opal_memchecker_base_select())) {
error = "opal_memchecker_base_select";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_backtrace_base_open())) {
error = "opal_backtrace_base_open";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_timer_base_open())) {
error = "opal_timer_base_open";
goto return_error;
}
/* setup the carto framework */
if (OPAL_SUCCESS != (ret = opal_carto_base_open())) {
error = "opal_carto_base_open";
goto return_error;
}
if (OPAL_SUCCESS != (ret = opal_carto_base_select())) {
error = "opal_carto_base_select";
goto return_error;
}
/*
* Need to start the event and progress engines if noone else is.
* opal_cr_init uses the progress engine, so it is lumped together
* into this set as well.
*/
/*
* Initialize the event library
*/
if (OPAL_SUCCESS != (ret = opal_event_init())) {
error = "opal_event_init";
goto return_error;
}
/*
* Initialize the general progress engine
*/
if (OPAL_SUCCESS != (ret = opal_progress_init())) {
error = "opal_progress_init";
goto return_error;
}
/* we want to tick the event library whenever possible */
opal_progress_event_users_increment();
#if OPAL_ENABLE_FT_CR == 1
/*
* Initialize the compression framework
* Note: Currently only used in C/R so it has been marked to only
* initialize when C/R is enabled. If other places in the code
* wish to use this framework, it is safe to remove the protection.
*/
if( OPAL_SUCCESS != (ret = opal_compress_base_open()) ) {
error = "opal_compress_base_open() failed";
goto return_error;
}
if( OPAL_SUCCESS != (ret = opal_compress_base_select()) ) {
error = "opal_compress_base_select() failed";
goto return_error;
}
#endif
/*
* Initalize the checkpoint/restart functionality
* Note: Always do this so we can detect if the user
* attempts to checkpoint a non checkpointable job,
* otherwise the tools may hang or not clean up properly.
*/
if (OPAL_SUCCESS != (ret = opal_cr_init() ) ) {
error = "opal_cr_init() failed";
goto return_error;
}
return OPAL_SUCCESS;
return_error:
opal_show_help( "help-opal-runtime.txt",
"opal_init:startup:internal-failure", true,
error, ret );
return ret;
}