ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
openmpi/orte/mca/rml/oob/rml_oob_component.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

516 строки
16 KiB
C
Исходник Ответственный История

/*
* Copyright (c) 2004-2007 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 Cisco Systems, Inc. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "orte_config.h"
#include "orte/constants.h"
#include "orte/util/output.h"
#include "opal/mca/base/base.h"
#include "opal/mca/base/mca_base_param.h"
#include "orte/mca/rml/base/base.h"
#include "orte/mca/routed/routed.h"
#include "orte/mca/errmgr/errmgr.h"
#include "orte/util/name_fns.h"
#include "orte/runtime/orte_globals.h"
#include "orte/mca/oob/oob.h"
#include "orte/mca/oob/base/base.h"
#include "rml_oob.h"
static orte_rml_module_t* rml_oob_init(int* priority);
static int rml_oob_open(void);
static int rml_oob_close(void);
static void rml_oob_recv_route_callback(int status,
struct orte_process_name_t* peer,
struct iovec* iov,
int count,
orte_rml_tag_t tag,
void *cbdata);
static void rml_oob_queued_progress(int fd, short event, void *arg);
/**
* component definition
*/
orte_rml_component_t mca_rml_oob_component = {
/* First, the mca_base_component_t struct containing meta
information about the component itself */
{
/* Indicate that we are a rml v1.0.0 component (which also
implies a specific MCA version) */
ORTE_RML_BASE_VERSION_1_0_0,
"oob", /* MCA component name */
ORTE_MAJOR_VERSION, /* MCA component major version */
ORTE_MINOR_VERSION, /* MCA component minor version */
ORTE_RELEASE_VERSION, /* MCA component release version */
rml_oob_open, /* component open */
rml_oob_close, /* component close */
},
/* Next the MCA v1.0.0 component meta data */
{
/* The component is checkpoint ready */
MCA_BASE_METADATA_PARAM_CHECKPOINT
},
rml_oob_init
};
orte_rml_oob_module_t orte_rml_oob_module = {
{
orte_rml_oob_init,
orte_rml_oob_fini,
orte_rml_oob_get_uri,
orte_rml_oob_set_uri,
orte_rml_oob_get_new_name,
orte_rml_oob_ping,
orte_rml_oob_send,
orte_rml_oob_send_nb,
orte_rml_oob_send_buffer,
orte_rml_oob_send_buffer_nb,
orte_rml_oob_recv,
orte_rml_oob_recv_nb,
orte_rml_oob_recv_buffer,
orte_rml_oob_recv_buffer_nb,
orte_rml_oob_recv_cancel,
orte_rml_oob_add_exception,
orte_rml_oob_del_exception,
orte_rml_oob_ft_event
}
};
static int
rml_oob_open(void)
{
int rc;
if (ORTE_SUCCESS != (rc = mca_oob_base_open())) {
ORTE_ERROR_LOG(rc);
return rc;
}
return rc;
}
static int
rml_oob_close(void)
{
int rc;
if (ORTE_SUCCESS != (rc = mca_oob_base_close())) {
return rc;
}
return rc;
}
static orte_rml_module_t*
rml_oob_init(int* priority)
{
if (mca_oob_base_init() != ORTE_SUCCESS)
return NULL;
*priority = 1;
OBJ_CONSTRUCT(&orte_rml_oob_module.exceptions, opal_list_t);
OBJ_CONSTRUCT(&orte_rml_oob_module.exceptions_lock, opal_mutex_t);
OBJ_CONSTRUCT(&orte_rml_oob_module.queued_routing_messages, opal_list_t);
OBJ_CONSTRUCT(&orte_rml_oob_module.queued_lock, opal_mutex_t);
/* Set default timeout for queued messages to be 1/2 second */
orte_rml_oob_module.timeout.tv_sec = 0;
orte_rml_oob_module.timeout.tv_usec = 500000;
orte_rml_oob_module.timer_event = malloc(sizeof(opal_event_t));
if (NULL == orte_rml_oob_module.timer_event) {
return NULL;
}
opal_evtimer_set(orte_rml_oob_module.timer_event, rml_oob_queued_progress,
NULL);
orte_rml_oob_module.active_oob = &mca_oob;
orte_rml_oob_module.active_oob->oob_exception_callback =
orte_rml_oob_exception_callback;
return &orte_rml_oob_module.super;
}
static struct iovec route_recv_iov[1];
int
orte_rml_oob_init(void)
{
int ret;
ret = orte_rml_oob_module.active_oob->oob_init();
route_recv_iov[0].iov_base = NULL;
route_recv_iov[0].iov_len = 0;
ret = orte_rml_oob_module.active_oob->oob_recv_nb(ORTE_NAME_WILDCARD,
route_recv_iov, 1,
ORTE_RML_TAG_RML_ROUTE,
ORTE_RML_ALLOC|ORTE_RML_PERSISTENT,
rml_oob_recv_route_callback,
NULL);
/* enable the base receive to get updates on contact info */
orte_rml_base_comm_start();
return ret;
}
int
orte_rml_oob_fini(void)
{
opal_list_item_t *item;
while (NULL !=
(item = opal_list_remove_first(&orte_rml_oob_module.exceptions))) {
OBJ_RELEASE(item);
}
OBJ_DESTRUCT(&orte_rml_oob_module.exceptions);
OBJ_DESTRUCT(&orte_rml_oob_module.exceptions_lock);
OBJ_DESTRUCT(&orte_rml_oob_module.queued_routing_messages);
OBJ_DESTRUCT(&orte_rml_oob_module.queued_lock);
orte_rml_oob_module.active_oob->oob_exception_callback = NULL;
/* clear the base receive */
orte_rml_base_comm_stop();
return ORTE_SUCCESS;
}
int
orte_rml_oob_ft_event(int state) {
int exit_status = ORTE_SUCCESS;
int ret;
if(OPAL_CRS_CHECKPOINT == state) {
;
}
else if(OPAL_CRS_CONTINUE == state) {
;
}
else if(OPAL_CRS_RESTART == state) {
;
}
else if(OPAL_CRS_TERM == state ) {
;
}
else {
;
}
if( ORTE_SUCCESS !=
(ret = orte_rml_oob_module.active_oob->oob_ft_event(state)) ) {
ORTE_ERROR_LOG(ret);
exit_status = ret;
goto cleanup;
}
if(OPAL_CRS_CHECKPOINT == state) {
;
}
else if(OPAL_CRS_CONTINUE == state) {
;
}
else if(OPAL_CRS_RESTART == state) {
if( ORTE_SUCCESS != (ret = mca_oob_base_close())) {
ORTE_ERROR_LOG(ret);
exit_status = ret;
goto cleanup;
}
if( ORTE_SUCCESS != (ret = mca_oob_base_open())) {
ORTE_ERROR_LOG(ret);
exit_status = ret;
goto cleanup;
}
if( ORTE_SUCCESS != (ret = mca_oob_base_init())) {
ORTE_ERROR_LOG(ret);
exit_status = ret;
goto cleanup;
}
}
else if(OPAL_CRS_TERM == state ) {
;
}
else {
;
}
cleanup:
return exit_status;
}
static void
msg_construct(orte_rml_oob_msg_t *msg)
{
OBJ_CONSTRUCT(&msg->msg_lock, opal_mutex_t);
OBJ_CONSTRUCT(&msg->msg_cond, opal_condition_t);
msg->msg_status = 0;
msg->msg_complete = false;
msg->msg_persistent = false;
OBJ_CONSTRUCT(&msg->msg_recv_buffer, opal_buffer_t);
msg->msg_data = NULL;
}
static void
msg_destruct(orte_rml_oob_msg_t *msg)
{
if (NULL != msg->msg_data) free(msg->msg_data);
OBJ_DESTRUCT(&msg->msg_recv_buffer);
OBJ_DESTRUCT(&msg->msg_lock);
OBJ_DESTRUCT(&msg->msg_cond);
}
OBJ_CLASS_INSTANCE(orte_rml_oob_msg_t, opal_object_t,
msg_construct, msg_destruct);
static void
queued_msg_construct(orte_rml_oob_queued_msg_t *msg)
{
msg->payload[0].iov_base = NULL;
msg->payload[0].iov_len = 0;
}
static void
queued_msg_destruct(orte_rml_oob_queued_msg_t *msg)
{
if (NULL != msg->payload[0].iov_base) free(msg->payload[0].iov_base);
}
OBJ_CLASS_INSTANCE(orte_rml_oob_queued_msg_t, opal_list_item_t,
queued_msg_construct, queued_msg_destruct);
static void
rml_oob_recv_route_queued_send_callback(int status,
struct orte_process_name_t* peer,
struct iovec* iov,
int count,
orte_rml_tag_t tag,
void* cbdata)
{
orte_rml_oob_queued_msg_t *qmsg = (orte_rml_oob_queued_msg_t*) cbdata;
OBJ_RELEASE(qmsg);
}
static void
rml_oob_queued_progress(int fd, short event, void *arg)
{
orte_rml_oob_queued_msg_t *qmsg;
orte_rml_oob_msg_header_t *hdr;
int real_tag;
int ret;
orte_process_name_t next, origin;
while (true) {
OPAL_THREAD_LOCK(&orte_rml_oob_module.queued_lock);
qmsg = (orte_rml_oob_queued_msg_t*) opal_list_remove_first(&orte_rml_oob_module.queued_routing_messages);
OPAL_THREAD_UNLOCK(&orte_rml_oob_module.queued_lock);
if (NULL == qmsg) break;
hdr = (orte_rml_oob_msg_header_t*) qmsg->payload[0].iov_base;
origin = hdr->origin;
next = orte_routed.get_route(&hdr->destination);
if (next.vpid == ORTE_VPID_INVALID) {
orte_output(0,
"%s tried routing message to %s, can't find route",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->destination));
abort();
}
if (OPAL_EQUAL == orte_util_compare_name_fields(ORTE_NS_CMP_ALL, &next, ORTE_PROC_MY_NAME)) {
orte_output(0, "%s trying to get message to %s, routing loop",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->destination));
abort();
}
if (OPAL_EQUAL == orte_util_compare_name_fields(ORTE_NS_CMP_ALL, &next, &hdr->destination)) {
real_tag = hdr->tag;
} else {
real_tag = ORTE_RML_TAG_RML_ROUTE;
}
ORTE_OUTPUT_VERBOSE((1, orte_rml_base_output,
"%s routing message from %s for %s to %s (tag: %d)",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->origin),
ORTE_NAME_PRINT(&hdr->destination),
ORTE_NAME_PRINT(&next),
hdr->tag));
ORTE_RML_OOB_MSG_HEADER_HTON(*hdr);
ret = orte_rml_oob_module.active_oob->oob_send_nb(&next,
&origin,
qmsg->payload,
1,
real_tag,
0,
rml_oob_recv_route_queued_send_callback,
qmsg);
if (ORTE_SUCCESS != ret) {
if (ORTE_ERR_ADDRESSEE_UNKNOWN == ret) {
/* still no route -- try again */
OPAL_THREAD_LOCK(&orte_rml_oob_module.queued_lock);
opal_list_append(&orte_rml_oob_module.queued_routing_messages,
&qmsg->super);
if (1 == opal_list_get_size(&orte_rml_oob_module.queued_routing_messages)) {
opal_evtimer_add(orte_rml_oob_module.timer_event,
&orte_rml_oob_module.timeout);
}
OPAL_THREAD_UNLOCK(&orte_rml_oob_module.queued_lock);
} else {
orte_output(0,
"%s failed to send message to %s: %s (rc = %d)",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&next),
ORTE_ERROR_NAME(ret),
ret);
abort();
}
}
}
}
static void
rml_oob_recv_route_send_callback(int status,
struct orte_process_name_t* peer,
struct iovec* iov,
int count,
orte_rml_tag_t tag,
void* cbdata)
{
}
static void
rml_oob_recv_route_callback(int status,
struct orte_process_name_t* peer,
struct iovec* iov,
int count,
orte_rml_tag_t tag,
void *cbdata)
{
orte_rml_oob_msg_header_t *hdr =
(orte_rml_oob_msg_header_t*) iov[0].iov_base;
int real_tag;
int ret;
orte_process_name_t next, origin;
/* BWB -- propogate errors here... */
assert(status >= 0);
ORTE_RML_OOB_MSG_HEADER_NTOH(*hdr);
origin = hdr->origin;
next = orte_routed.get_route(&hdr->destination);
if (next.vpid == ORTE_VPID_INVALID) {
orte_output(0,
"%s tried routing message to %s, can't find route",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->destination));
abort();
}
if (OPAL_EQUAL == orte_util_compare_name_fields(ORTE_NS_CMP_ALL, &next, ORTE_PROC_MY_NAME)) {
orte_output(0, "%s trying to get message to %s, routing loop",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->destination));
abort();
}
if (OPAL_EQUAL == orte_util_compare_name_fields(ORTE_NS_CMP_ALL, &next, &hdr->destination)) {
real_tag = hdr->tag;
} else {
real_tag = ORTE_RML_TAG_RML_ROUTE;
}
ORTE_OUTPUT_VERBOSE((1, orte_rml_base_output,
"%s routing message from %s for %s to %s (tag: %d)",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&hdr->origin),
ORTE_NAME_PRINT(&hdr->destination),
ORTE_NAME_PRINT(&next),
hdr->tag));
ORTE_RML_OOB_MSG_HEADER_HTON(*hdr);
ret = orte_rml_oob_module.active_oob->oob_send_nb(&next,
&origin,
iov,
count,
real_tag,
0,
rml_oob_recv_route_send_callback,
NULL);
if (ORTE_SUCCESS != ret) {
if (ORTE_ERR_ADDRESSEE_UNKNOWN == ret) {
/* no route -- queue and hope we find a route */
orte_rml_oob_queued_msg_t *qmsg = OBJ_NEW(orte_rml_oob_queued_msg_t);
ORTE_OUTPUT_VERBOSE((1, orte_rml_base_output,
"%s: no OOB information for %s. Queuing for later.",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&next)));
ORTE_RML_OOB_MSG_HEADER_NTOH(*hdr);
qmsg->payload[0].iov_base = malloc(iov[0].iov_len);
if (NULL == qmsg->payload[0].iov_base) abort();
qmsg->payload[0].iov_len = iov[0].iov_len;
memcpy(qmsg->payload[0].iov_base, iov[0].iov_base, iov[0].iov_len);
OPAL_THREAD_LOCK(&orte_rml_oob_module.queued_lock);
opal_list_append(&orte_rml_oob_module.queued_routing_messages,
&qmsg->super);
if (1 == opal_list_get_size(&orte_rml_oob_module.queued_routing_messages)) {
opal_evtimer_add(orte_rml_oob_module.timer_event,
&orte_rml_oob_module.timeout);
}
OPAL_THREAD_UNLOCK(&orte_rml_oob_module.queued_lock);
} else {
orte_output(0,
"%s failed to send message to %s: %s (rc = %d)",
ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
ORTE_NAME_PRINT(&next),
opal_strerror(ret),
ret);
abort();
}
}
}
Показать git blame Копировать постоянную ссылку