1
1
openmpi/opal/util
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
..
keyval Back out r15158 because it apparently breaks with recent versions of 2007-06-21 21:14:10 +00:00
arch.c Shift the architecture calculation from the ompi/datatype engine to the opal/util area. This allows us to compute the architecture earlier in the launch and communicate it outside of the modex. 2008-04-17 20:43:56 +00:00
arch.h Make the symbols be visible. 2008-04-18 00:26:17 +00:00
argv.c Add new opal_argv_split_with_empty() function. opal_argv_split() function 2007-08-01 12:08:11 +00:00
argv.h Add new opal_argv_split_with_empty() function. opal_argv_split() function 2007-08-01 12:08:11 +00:00
basename.c Fix a bunch of issues exposed by Ken Cain in getting Open MPI to work with 2007-07-10 03:46:57 +00:00
basename.h On windows make sure we add the initial path separator only if we are 2007-07-01 17:51:34 +00:00
bit_ops.h Update the copyright notices for IU and UTK. 2005-11-05 19:57:48 +00:00
cmd_line.c Coverty blame list: 2007-09-14 02:08:37 +00:00
cmd_line.h Change a bunch of 'OMPI' names to OPAL names, since this is opal. 2007-03-29 01:00:33 +00:00
convert.c Latest and greatest. Now OPAL is ready for the Windows prime-time. 2006-08-23 00:29:35 +00:00
convert.h On windows make sure we add the initial path separator only if we are 2007-07-01 17:51:34 +00:00
crc.c Look like a big commit but in fact it address only one issue. The way we're working with 2006-10-17 20:20:58 +00:00
crc.h We always have to include the <PROJECT>_config.h as first include. 2006-10-20 07:01:52 +00:00
daemon_init.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
daemon_init.h Latest and greatest. Now OPAL is ready for the Windows prime-time. 2006-08-23 00:29:35 +00:00
error.c Merge in the ipv6 work from /tmp/ipv6-merge. 2007-04-25 01:55:40 +00:00
error.h Latest and greatest. Now OPAL is ready for the Windows prime-time. 2006-08-23 00:29:35 +00:00
few.c One typo and one UNIX conversion of EOL. 2006-08-25 23:22:35 +00:00
few.h Change a bunch of 'OMPI' names to OPAL names, since this is opal. 2007-03-29 01:00:33 +00:00
if.c Implement the seq rmaps module that sequentially maps process ranks to a list hosts in a hostfile. 2008-04-17 13:50:59 +00:00
if.h * Mark some of the structures passed into the if and net code as const when 2007-07-22 19:19:01 +00:00
keyval_parse.c One big step forward. Mostly explicit casting for Windows plus the 2006-08-22 20:07:42 +00:00
keyval_parse.h Windows friendly version. We have to split the OMPI_DECLSPEC in at least 3 2006-08-20 15:54:04 +00:00
Makefile.am Shift the architecture calculation from the ompi/datatype engine to the opal/util area. This allows us to compute the architecture earlier in the launch and communicate it outside of the modex. 2008-04-17 20:43:56 +00:00
malloc.c Hide what should be hidden, and expose the others. Plus some indentation. 2008-03-18 03:00:08 +00:00
malloc.h Hide what should be hidden, and expose the others. Plus some indentation. 2008-03-18 03:00:08 +00:00
net.c Few clean-ups, few indentations. Nothing really important. 2007-08-04 00:44:23 +00:00
net.h * Mark some of the structures passed into the if and net code as const when 2007-07-22 19:19:01 +00:00
num_procs.c Protect unistd.h usage. 2007-02-07 22:36:50 +00:00
num_procs.h - this symbol is required by components 2007-02-28 16:38:59 +00:00
numtostr.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
numtostr.h Change a bunch of 'OMPI' names to OPAL names, since this is opal. 2007-03-29 01:00:33 +00:00
opal_environ.c Use the correct prefix. 2008-03-31 21:42:59 +00:00
opal_environ.h Add two convenience functions in order to make sure we get these 2008-03-13 17:56:44 +00:00
opal_getcwd.c Coverty fix: Replace strcpy by strncpy. 2008-03-25 22:39:24 +00:00
opal_getcwd.h Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. 2008-02-28 01:57:57 +00:00
opal_pty.c * better support platforms without any concept of a terminal at all 2006-03-31 18:51:06 +00:00
opal_pty.h Some platforms (such as FreeBSD) need libutil.h included for openpty. 2007-09-19 21:59:22 +00:00
os_dirpath.c Coverty blame list: 2007-09-14 02:08:37 +00:00
os_dirpath.h On windows make sure we add the initial path separator only if we are 2007-07-01 17:51:34 +00:00
os_path.c - remove wrong va_end 2007-08-07 13:32:05 +00:00
os_path.h - Further attributes mainly for opal/* functions, marking 2007-03-20 13:01:32 +00:00
output.c This commit represents a bunch of work on a Mercurial side branch. As 2008-05-13 20:00:55 +00:00
output.h This commit represents a bunch of work on a Mercurial side branch. As 2008-05-13 20:00:55 +00:00
path.c Don't allocate the temporary buffer on the stack. It get way too much 2007-09-14 02:09:38 +00:00
path.h This one was nice ... and it didn't even trigger a compile warning. 2007-07-01 17:59:46 +00:00
pow2.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
pow2.h Latest and greatest. Now OPAL is ready for the Windows prime-time. 2006-08-23 00:29:35 +00:00
printf.c - add missing va_end 2007-08-07 12:25:20 +00:00
printf.h Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately. 2008-02-28 01:57:57 +00:00
qsort.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
qsort.h Update the copyright notices for IU and UTK. 2005-11-05 19:57:48 +00:00
show_help_lex.h George found a great way to avoid warnings from flex for that unused 2006-08-28 13:44:37 +00:00
show_help_lex.l This commit represents a bunch of work on a Mercurial side branch. As 2008-05-13 20:00:55 +00:00
show_help.c This commit represents a bunch of work on a Mercurial side branch. As 2008-05-13 20:00:55 +00:00
show_help.h This commit represents a bunch of work on a Mercurial side branch. As 2008-05-13 20:00:55 +00:00
stacktrace.c Fixes for building on kFreeBSD. Re #1105 2007-07-24 23:19:45 +00:00
stacktrace.h Change a bunch of 'OMPI' names to OPAL names, since this is opal. 2007-03-29 01:00:33 +00:00
strncpy.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
strncpy.h - Further attributes mainly for opal/* functions, marking 2007-03-20 13:01:32 +00:00
sys_limits.c Get rid of structure rlimit only on Windows, which doesn't provide such structure definition. 2007-06-04 12:27:40 +00:00
sys_limits.h Add ability to read system limits on number of children, open files, and file size from the local OS - to be used in failed-to-start scenarios 2007-04-23 18:53:47 +00:00
trace.c Next step in the project split, mainly source code re-arranging 2006-02-12 01:33:29 +00:00
trace.h Latest and greatest. Now OPAL is ready for the Windows prime-time. 2006-08-23 00:29:35 +00:00