ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
f5dfc005a4
openmpi/orte/mca/odls/base/odls_base_open.c

153 строки
4.8 KiB
C
Исходник Обычный вид История

Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
/*
* Copyright (c) 2004-2005 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.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "orte_config.h"
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
#include "orte/constants.h"
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
#include "opal/mca/mca.h"
#include "opal/mca/base/base.h"
#include "opal/mca/base/mca_base_param.h"
#include "opal/util/output.h"
#include "opal/util/trace.h"
Squeeeeeeze the launch message. This is the message sent to the daemons that provides all the data required for launching their local procs. In reorganizing the ODLS framework, I discovered that we were sending a significant amount of unnecessary and repeated data. This commit resolves this by: 1. taking advantage of the fact that we no longer create the launch message via a GPR trigger. In earlier times, we had the GPR create the launch message based on a subscription. In that mode of operation, we could not guarantee the order in which the data was stored in the message - hence, we had no choice but to parse the message in a loop that checked each value against a list of possible "keys" until the corresponding value was found. Now, however, we construct the message "by hand", so we know precisely what data is in each location in the message. Thus, we no longer need to send the character string "keys" for each data value any more. This represents a rather large savings in the message size - to give you an example, we typically would use a 30-char "key" for a 2-byte data value. As you can see, the overhead can become very large. 2. sending node-specific data only once. Again, because we used to construct the message via subscriptions that were done on a per-proc basis, the data for each node (e.g., the daemon's name, whether or not the node was oversubscribed) would be included in the data for each proc. Thus, the node-specific data was repeated for every proc. Now that we construct the message "by hand", there is no reason to do this any more. Instead, we can insert the data for a specific node only once, and then provide the per-proc data for that node. We therefore not only save all that extra data in the message, but we also only need to parse the per-node data once. The savings become significant at scale. Here is a comparison between the revised trunk and the trunk prior to this commit (all data was taken on odin, using openib, 64 nodes, unity message routing, tested with application consisting of mpi_init/mpi_barrier/mpi_finalize, all execution times given in seconds, all launch message sizes in bytes): Per-node scaling, taken at 1ppn: #nodes original trunk revised trunk time size time size 1 0.10 819 0.09 564 2 0.14 1070 0.14 677 3 0.15 1321 0.14 790 4 0.15 1572 0.15 903 8 0.17 2576 0.20 1355 16 0.25 4584 0.21 2259 32 0.28 8600 0.27 4067 64 0.50 16632 0.39 7683 Per-proc scaling, taken at 64 nodes ppn original trunk revised trunk time size time size 1 0.50 16669 0.40 7720 2 0.55 32733 0.54 11048 3 0.87 48797 0.81 14376 4 1.0 64861 0.85 17704 Condensing those numbers, it appears we gained: per-node message size: 251 bytes/node -> 113 bytes/node per-proc message size: 251 bytes/proc -> 52 bytes/proc per-job message size: 568 bytes/job -> 399 bytes/job (job-specific data such as jobid, override oversubscribe flag, total #procs in job, total slots allocated) The fact that the two pre-commit trunk numbers are the same confirms the fact that each proc was containing the node data as well. It isn't quite the 10x message reduction I had hoped to get, but it is significant and gives much better scaling. Note that the timing info was, as usual, pretty chaotic - the numbers cited here were typical across several runs taken after the initial one to avoid NFS file positioning influences. Also note that this commit removes the orte_process_info.vpid_start field and the handful of places that passed that useless value. By definition, all jobs start at vpid=0, so all we were doing is passing "0" around. In fact, many places simply hardwired it to "0" anyway rather than deal with it. This commit was SVN r16428.
2007-10-11 19:57:26 +04:00
#include "opal/util/argv.h"
Fully implement the inbound binomial allgather for daemon-based collectives. Supports both modex and barrier operations. Comm_spawn still uses the rank=0 method - shifting that algo to the daemons is under study. This commit was SVN r18115.
2008-04-10 02:10:53 +04:00
#include "opal/class/opal_value_array.h"
Add a binomial tree-based launch to ssh, turned "on" only when the plm_rsh_tree_spawned mca param is set to a non-zero value. This probably isn't a very optimized capability, but it does execute a tree-based launch that may scale better than linear at high node counts. Add the daemon map capability to the ODLS to create and save a map of daemon vpid vs nodename from the launch message. Cleanup a few places in the base plm launch support where we didn't adequately protect rml recv's from potentially executing sends. This commit was SVN r18143.
2008-04-14 22:26:08 +04:00
#include "opal/class/opal_pointer_array.h"
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
#include "opal/dss/dss.h"
Fully implement the inbound binomial allgather for daemon-based collectives. Supports both modex and barrier operations. Comm_spawn still uses the rank=0 method - shifting that algo to the daemons is under study. This commit was SVN r18115.
2008-04-10 02:10:53 +04:00
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
#include "orte/mca/errmgr/errmgr.h"
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
#include "orte/mca/plm/plm_types.h"
#include "orte/util/name_fns.h"
#include "orte/runtime/orte_globals.h"
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
#include "orte/mca/odls/base/base.h"
#include "orte/mca/odls/base/odls_private.h"
/*
* The following file was created by configure. It contains extern
* statements and the definition of an array of pointers to each
* component's public mca_base_component_t struct.
*/
#include "orte/mca/odls/base/static-components.h"
/*
* Instantiate globals
*/
orte_odls_base_module_t orte_odls;
This commit accomplishes two things: 1. Fix the "hang" condition when an application isn't found. It turned out that the ODLS had some difficulty with the process actually not having been started - hence, it never called the waitpid callback. As a result, the "terminated" trigger didn't fire, and so mpirun didn't wake up. With this change, the HNP's errmgr forces the issue by causing the trigger to fire itself when an abort condition occurs. 2. Shift the recording of the pid and the nodename from mpi_init to the orted launcher. This allows programs such as Eclipse PTP to get the pids even for non-MPI applications. In the case of bproc, the pls handles this chore since we don't use orteds in that system. This commit was SVN r12558.
2006-11-11 07:03:45 +03:00
/* instance the child list object */
static void orte_odls_child_constructor(orte_odls_child_t *ptr)
{
ptr->name = NULL;
Compute and pass the local_rank and local number of procs (in that proc's job) on the node. To be precise, given this hypothetical launching pattern: host1: vpids 0, 2, 4, 6 host2: vpids 1, 3, 5, 7 The local_rank for these procs would be: host1: vpids 0->local_rank 0, v2->lr1, v4->lr2, v6->lr3 host2: vpids 1->local_rank 0, v3->lr1, v5->lr2, v7->lr3 and the number of local procs on each node would be four. If vpid=0 then does a comm_spawn of one process on host1, the values of the parent job would remain unchanged. The local_rank of the child process would be 0 and its num_local_procs would be 1 since it is in a separate jobid. I have verified this functionality for the rsh case - need to verify that slurm and other cases also get the right values. Some consolidation of common code is probably going to occur in the SDS components to make this simpler and more maintainable in the future. This commit was SVN r14706.
2007-05-21 18:30:10 +04:00
ptr->local_rank = ORTE_VPID_INVALID;
This commit accomplishes two things: 1. Fix the "hang" condition when an application isn't found. It turned out that the ODLS had some difficulty with the process actually not having been started - hence, it never called the waitpid callback. As a result, the "terminated" trigger didn't fire, and so mpirun didn't wake up. With this change, the HNP's errmgr forces the issue by causing the trigger to fire itself when an abort condition occurs. 2. Shift the recording of the pid and the nodename from mpi_init to the orted launcher. This allows programs such as Eclipse PTP to get the pids even for non-MPI applications. In the case of bproc, the pls handles this chore since we don't use orteds in that system. This commit was SVN r12558.
2006-11-11 07:03:45 +03:00
ptr->pid = 0;
ptr->app_idx = -1;
ptr->alive = false;
Improve the scalability of the modex operation and fix a bug reported by Tim P The bug was a race condition in the barrier operation that caused the barrier in MPI_Finalize to fail on very short programs. Scalaiblity was improved by using the daemons to aggregate modex and barrier messages before sending them to the rank=0 proc. Improvement is proportional to ppn, of course, but there really wasn't a scaling problem at low ppn anyway. This modification also paves the way for better allgather operations since now all the data for each node is sitting at the daemon level, and the daemons are now aware that a collective operation on the OOB is underway (so they -can- participate in a collective of their own to support it). Also added better diagnostics to map out the timing associated with MPI_Init - turned on by -mca orte_timing 1. This commit was SVN r17988.
2008-03-27 18:17:53 +03:00
ptr->coll_recvd = false;
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
/* set the default state to "failed to start" so
* we can correctly report should something
* go wrong during launch
*/
ptr->state = ORTE_PROC_STATE_FAILED_TO_START;
Teach the odls to properly test for and report failed-to-start for application processes. Test for system limits (where known) prior to doing things like fork and pipe since some systems aren't very nice about it when we try to exceed such limits. This commit was SVN r14494.
2007-04-24 22:54:45 +04:00
ptr->exit_code = 0;
Roll in the Voltaire core/socket/etc process mapping implementation. Only change I made was to cleanup some of the diagnostic output in the odls_default component so it uses the -mca odls_base_verbose parameter. You will not see any impact from this change unless you use the syntax described in ticket #1023. I've tried as many of the RAS components as possible and saw no problem - there may be issues with other RAS components that would not compile on any of my systems. Anything that appears should be trivial to fix. This commit was SVN r15427.
2007-07-14 19:14:07 +04:00
ptr->cpu_set = 0xffffffff;
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
ptr->rml_uri = NULL;
Support for new RMAPS rank mapping component This commit was SVN r17860.
2008-03-18 12:39:07 +03:00
ptr->slot_list = NULL;
This commit accomplishes two things: 1. Fix the "hang" condition when an application isn't found. It turned out that the ODLS had some difficulty with the process actually not having been started - hence, it never called the waitpid callback. As a result, the "terminated" trigger didn't fire, and so mpirun didn't wake up. With this change, the HNP's errmgr forces the issue by causing the trigger to fire itself when an abort condition occurs. 2. Shift the recording of the pid and the nodename from mpi_init to the orted launcher. This allows programs such as Eclipse PTP to get the pids even for non-MPI applications. In the case of bproc, the pls handles this chore since we don't use orteds in that system. This commit was SVN r12558.
2006-11-11 07:03:45 +03:00
}
static void orte_odls_child_destructor(orte_odls_child_t *ptr)
{
if (NULL != ptr->name) free(ptr->name);
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
if (NULL != ptr->rml_uri) free(ptr->rml_uri);
Support for new RMAPS rank mapping component This commit was SVN r17860.
2008-03-18 12:39:07 +03:00
if (NULL != ptr->slot_list) free(ptr->slot_list);
This commit accomplishes two things: 1. Fix the "hang" condition when an application isn't found. It turned out that the ODLS had some difficulty with the process actually not having been started - hence, it never called the waitpid callback. As a result, the "terminated" trigger didn't fire, and so mpirun didn't wake up. With this change, the HNP's errmgr forces the issue by causing the trigger to fire itself when an abort condition occurs. 2. Shift the recording of the pid and the nodename from mpi_init to the orted launcher. This allows programs such as Eclipse PTP to get the pids even for non-MPI applications. In the case of bproc, the pls handles this chore since we don't use orteds in that system. This commit was SVN r12558.
2006-11-11 07:03:45 +03:00
}
OBJ_CLASS_INSTANCE(orte_odls_child_t,
opal_list_item_t,
orte_odls_child_constructor,
orte_odls_child_destructor);
Fully implement the inbound binomial allgather for daemon-based collectives. Supports both modex and barrier operations. Comm_spawn still uses the rank=0 method - shifting that algo to the daemons is under study. This commit was SVN r18115.
2008-04-10 02:10:53 +04:00
/* instance the job list object */
static void orte_odls_job_constructor(orte_odls_job_t *ptr)
{
ptr->jobid = ORTE_JOBID_INVALID;
ptr->dp = 0;
OBJ_CONSTRUCT(&ptr->daemons, opal_value_array_t);
opal_value_array_init(&ptr->daemons, sizeof(orte_vpid_t));
}
static void orte_odls_job_destructor(orte_odls_job_t *ptr)
{
OBJ_DESTRUCT(&ptr->daemons);
}
OBJ_CLASS_INSTANCE(orte_odls_job_t,
opal_list_item_t,
orte_odls_job_constructor,
orte_odls_job_destructor);
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
/*
* Framework global variables
*/
orte_odls_base_t orte_odls_base;
orte_odls_globals_t orte_odls_globals;
/**
* Function for finding and opening either all MCA components, or the one
* that was specifically requested via a MCA parameter.
*/
int orte_odls_base_open(void)
{
Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer This commit was SVN r17632.
2008-02-28 04:57:57 +03:00
/* Debugging / verbose output. Always have stream open, with
verbose set by the mca open system... */
orte_odls_globals.output = opal_output_open(NULL);
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
Fix invalid MCA 'base' names so they appear in ompi_info. A subset of this patch needs to be applied to v1.2 Refs trac:928 This commit was SVN r15918. The following Trac tickets were found above: Ticket 928 --> https://svn.open-mpi.org/trac/ompi/ticket/928
2007-08-18 07:05:45 +04:00
mca_base_param_reg_int_name("odls", "base_sigkill_timeout",
Need to revise the display-map-at-launch option so it is active not only for the initial launch, but applies to any subsequent comm_spawn events too. Add placeholders for the new orte tools. These don't actually do anything yet - in fact, I have set the .ompi_ignore so that you won't compile them (I have set a .ompi_unignore for me). Please let me know if you encounter any trouble with this - the ompi_ignore's should protect everyone. This commit was SVN r12616.
2006-11-17 05:58:46 +03:00
"Time to wait for a process to die after issuing a kill signal to it",
false, false, 1, &orte_odls_globals.timeout_before_sigkill);
Bring in the code for routing xcast stage gate messages via the local orteds. This code is inactive unless you specifically request it via an mca param oob_xcast_mode (can be set to "linear" or "direct"). Direct mode is the old standard method where we send messages directly to each MPI process. Linear mode sends the xcast message via the orteds, with the HNP sending the message to each orted directly. There is a binomial algorithm in the code (i.e., the HNP would send to a subset of the orteds, which then relay it on according to the typical log-2 algo), but that has a bug in it so the code won't let you select it even if you tried (and the mca param doesn't show, so you'd *really* have to try). This also involved a slight change to the oob.xcast API, so propagated that as required. Note: this has *only* been tested on rsh, SLURM, and Bproc environments (now that it has been transferred to the OMPI trunk, I'll need to re-test it [only done rsh so far]). It should work fine on any environment that uses the ORTE daemons - anywhere else, you are on your own... :-) Also, correct a mistake where the orte_debug_flag was declared an int, but the mca param was set as a bool. Move the storage for that flag to the orte/runtime/params.c and orte/runtime/params.h files appropriately. This commit was SVN r14475.
2007-04-23 22:41:04 +04:00
Reorganize the odls framework to place common functionality in the base, thus making maintenance easier. We still need this to be a framework as some environments (e.g., bproc) require significantly different functionality. However, there is quite a bit of commonality across the components, so this ensures that fixes in one get propagated across the others. This patch also fixes a minor bug discovered along the way: we had "lost" the passing of the oversubscribed condition flag from the mapper to the orteds. Thus, we were not setting sched_yield correctly when in oversubscribed conditions (except when a hostfile was specified - different logic there because we treat the number of slots allocated on the node as "uncertain") I did not modify the process component in this patch - I will send a proposed patch to the maintainers of that component so they can review it first. This commit was SVN r16418.
2007-10-10 19:02:10 +04:00
/* initialize globals */
OBJ_CONSTRUCT(&orte_odls_globals.mutex, opal_mutex_t);
OBJ_CONSTRUCT(&orte_odls_globals.cond, opal_condition_t);
OBJ_CONSTRUCT(&orte_odls_globals.children, opal_list_t);
Fully implement the inbound binomial allgather for daemon-based collectives. Supports both modex and barrier operations. Comm_spawn still uses the rank=0 method - shifting that algo to the daemons is under study. This commit was SVN r18115.
2008-04-10 02:10:53 +04:00
OBJ_CONSTRUCT(&orte_odls_globals.jobs, opal_list_t);
Add a binomial tree-based launch to ssh, turned "on" only when the plm_rsh_tree_spawned mca param is set to a non-zero value. This probably isn't a very optimized capability, but it does execute a tree-based launch that may scale better than linear at high node counts. Add the daemon map capability to the ODLS to create and save a map of daemon vpid vs nodename from the launch message. Cleanup a few places in the base plm launch support where we didn't adequately protect rml recv's from potentially executing sends. This commit was SVN r18143.
2008-04-14 22:26:08 +04:00
/* initialize and setup the daemonmap */
OBJ_CONSTRUCT(&orte_daemonmap, opal_pointer_array_t);
opal_pointer_array_init(&orte_daemonmap, 8, INT32_MAX, 8);
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
/* Open up all available components */
if (ORTE_SUCCESS !=
mca_base_components_open("odls", orte_odls_globals.output,
mca_odls_base_static_components,
&orte_odls_base.available_components, true)) {
return ORTE_ERROR;
}
set orte_odls_base.components_available to false if no odls components are available. Startup now works if no odls components are availble. This commit was SVN r13339.
2007-01-27 18:37:13 +03:00
/* are there components available for use ? -
* orte_odls_base.available_components is always initialized */
if(0 < opal_list_get_size(&(orte_odls_base.available_components))) {
orte_odls_base.components_available = true;
} else {
orte_odls_base.components_available = false;
}
Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done. This commit was SVN r11661.
2006-09-15 01:29:51 +04:00
/* All done */
return ORTE_SUCCESS;
}
Копировать постоянную ссылку