Update to track master branch

There have been a number of commits that didn't make the transition to this release branch. This was usually due to their being things in the commit that would have provided a positive change here, but also things that would bring in new material not intended for this release. Unfortunately, that makes this branch difficult to maintain. So let's sync the branch up to the HEAD of master and then "prune" away the pieces we know we don't want here. Signed-off-by: Ralph Castain <rhc@pmix.org> Includes (not an exhaustive list): (cherry picked from commit 84a55e1) (cherry picked from commit 70ed111) (cherry picked from commit 19eba9e) (cherry picked from commit 1e4edb5)
openpmix · Sep 14, 2023 · ada0565 · ada0565
1 parent 1c38f31
commit ada0565
Show file tree

Hide file tree

Showing 170 changed files with 8,137 additions and 4,461 deletions.
diff --git a/.gitignore b/.gitignore
@@ -71,7 +71,6 @@ vc70.pdb
 .hgignore_local
 stamp-h?
 AUTHORS
-docs-venv/
 
 ar-lib
 ylwrap
@@ -149,6 +148,7 @@ src/include/prte_config.h.in
 src/include/version.h
 src/include/prte_frameworks.c
 src/include/prte_frameworks.h
+src/include/frameworks.h
 src/include/prte_config.h
 
 src/mca/rmaps/rank_file/rmaps_rank_file_lex.c
@@ -161,6 +161,7 @@ src/tools/prte/prte
 src/tools/pcc/pcc
 src/tools/pcc/pcc-wrapper-data.txt
 src/tools/pterm/pterm
+src/tools/psched/psched
 
 src/util/hostfile/hostfile_lex.c
 src/util/keyval/keyval_lex.c
@@ -188,7 +189,12 @@ docs/_build
 docs/_static
 docs/_static/css/custom.css
 docs/_templates
+src/docs/_build
+src/docs/mca/mca.rst
+src/docs/mca/help*rst
+__pycache__
 
 # Common Python virtual environment directory names
 venv
 py??
+docs-venv
diff --git a/LICENSE b/LICENSE
@@ -1,14 +1,14 @@
 Most files in this release are marked with the copyrights of the
 organizations who have edited them.  The copyrights below are in no
 particular order and generally reflect members of the PMIx and
-PRRTE core teams who have contributed code to this release.
+Open MPI core teams who have contributed code to this release.
 The copyrights for code used under license from other parties are
 included in the corresponding files.
 
 Copyright (c) 2004-2010 The Trustees of Indiana University and Indiana
                         University Research and Technology
                         Corporation.  All rights reserved.
-Copyright (c) 2004-2023 The University of Tennessee and The University
+Copyright (c) 2004-2017 The University of Tennessee and The University
                         of Tennessee Research Foundation.  All rights
                         reserved.
 Copyright (c) 2004-2010 High Performance Computing Center Stuttgart,
@@ -53,13 +53,12 @@ Copyright (c) 2014-2015 Hewlett-Packard Development Company, LP.  All
                         rights reserved.
 Copyright (c) 2013-2017 Research Organization for Information Science (RIST).
                         All rights reserved.
-Copyright (c) 2017-2023 Amazon.com, Inc. or its affiliates.  All Rights
+Copyright (c) 2017-2018 Amazon.com, Inc. or its affiliates.  All Rights
                         reserved.
 Copyright (c) 2018      DataDirect Networks. All rights reserved.
-Copyright (c) 2020-2023 Nanook Consulting. All rights reserved.
-Copyright (c) 2022-2023 Triad National Security, LLC. All rights reserved
 Copyright (c) 2023      Advanced Micro Devices, Inc. All rights reserved.
 
+Copyright (c) 2023      Nanook Consulting.  All rights reserved.
 $COPYRIGHT$
 
 Additional copyrights may follow

diff --git a/Makefile.am b/Makefile.am
@@ -14,7 +14,9 @@
 # Copyright (c) 2014-2019 Intel, Inc.  All rights reserved.
 # Copyright (c) 2017-2018 Amazon.com, Inc. or its affiliates.
 #                         All Rights reserved.
-# Copyright (c) 2021-2022 Nanook Consulting.  All rights reserved.
+# Copyright (c) 2021-2023 Nanook Consulting.  All rights reserved.
+# Copyright (c) 2023      Jeffrey M. Squyres.  All rights reserved.
+#
 # $COPYRIGHT$
 #
 # Additional copyrights may follow
@@ -27,13 +29,6 @@ EXTRA_DIST = README.md VERSION LICENSE autogen.pl
 
 include examples/Makefile.include
 
-# dist-hook:
-#	env LS_COLORS= sh "$(top_srcdir)/config/distscript.sh" "$(top_srcdir)" "$(distdir)" "$(PRTE_REPO_REV)"
-#	@if test ! -s $(distdir)/AUTHORS ; then \
-#		echo "AUTHORS file is empty; aborting distribution"; \
-#		exit 1; \
-#	fi
-
 # Check for common symbols.  Use a "-hook" to increase the odds that a
 # developer will see it at the end of their installation process.
 install-exec-hook:
@@ -46,15 +41,3 @@ install-exec-hook:
 	fi
 
 ACLOCAL_AMFLAGS = -I config
-
-# Use EXTRA_DIST and an explicit target (with a FORCE hack so that
-# it's always run) rather than a dist-hook because there's some magic
-# extra logic in Automake that will add AUTHORS to EXTRA_DIST if the
-# file exists when Automake is run.  Once we're explicit (to avoid odd
-# copy behavior), it's easier to always build AUTHORS here, rather
-# than trying to handle the EXTRA_DIST dependency from a clean repo
-# (no AUTHORS file present) and use dist-hook to run every time.
-#AUTHORS: FORCE
-#	$(PERL) "$(top_srcdir)/contrib/dist/make-authors.pl" --skip-ok --quiet --srcdir="$(top_srcdir)"
-
-#FORCE:
diff --git a/Makefile.prte-rules b/Makefile.prte-rules
@@ -2,6 +2,7 @@
 # Copyright (c) 2008-2022 Cisco Systems, Inc.  All rights reserved.
 # Copyright (c) 2008      Sun Microsystems, Inc.  All rights reserved.
 # Copyright (c) 2020      Intel, Inc.  All rights reserved.
+# Copyright (c) 2023      Nanook Consulting.  All rights reserved.
 # $COPYRIGHT$
 #
 # Additional copyrights may follow
@@ -21,3 +22,11 @@ prte__v_SPHINX_HTML_0 = @echo "  GENERATE HTML docs";
 PRTE_V_SPHINX_MAN = $(prte__v_SPHINX_MAN_$V)
 prte__v_SPHINX_MAN_ = $(prte__v_SPHINX_MAN_$AM_DEFAULT_VERBOSITY)
 prte__v_SPHINX_MAN_0 = @echo "  GENERATE man pages";
+
+PRTE_V_TXT = $(prte__v_SPHINX_TXT_$V)
+prte__v_TXT_ = $(prte__v_SPHINX_TXT_$AM_DEFAULT_VERBOSITY)
+prte__v_TXT_0 = @echo "  GENERATE text files";
+
+PRTE_V_LN_S = $(prte__v_LN_S_$V)
+prte__v_LN_S_ = $(prte__v_LN_S_$AM_DEFAULT_VERBOSITY)
+prte__v_LN_S_0 = @echo "  LN_S    " `basename $@`;
diff --git a/NEWS b/NEWS
@@ -19,10 +19,10 @@ Copyright (c) 2012      Oak Ridge National Labs.  All rights reserved.
 Copyright (c) 2012      Sandia National Laboratories.  All rights reserved.
 Copyright (c) 2012      University of Houston. All rights reserved.
 Copyright (c) 2013      NVIDIA Corporation.  All rights reserved.
-Copyright (c) 2013-2019 Intel, Inc.  All rights reserved.
+# Copyright (c) 2013-2019 Intel, Inc.  All rights reserved.
 Copyright (c) 2018-2001 Amazon.com, Inc. or its affiliates.  All Rights
                         reserved.
-Copyright (c) 2021-2023 Nanook Consulting.  All rights reserved.
+# Copyright (c) 2021-2023 Nanook Consulting.  All rights reserved.
 $COPYRIGHT$
 
 Additional copyrights may follow
@@ -36,89 +36,7 @@ bug fixes (and other actions) for each version of PRRTE since
 version 1.0.
 
 
-3.0.1 -- TBD
-------------------------------------
-PR #1696: Always set waitpid-fired flag when proc exits
-PR #1694: Fix passing of envars for spawn
-PR #1692: Multiple commits
-  - Plug small memory leaks
-  - Adhere to project's variable declaration convention
-  - Cleanup initial implementation of allocate and session control
-  - Protect one more place for session_control
-  - Fix the routing for non-tree-spawn launch
-  - Cleanup prte_info output
-  - Ensure we exit cleanly when a daemon fails to start
-  - Store prted URI under correct name
-  - Remove duplicate computation of routing tree
-  - Minor cleanup of verbose output
-  - Send direct to HNP if we don't know a route
-  - Cleanup some mangled formatting
-  - Cleanup some debug output
-  - Update the pmix server integration
-PR #1686: Update config/oac submodule
-PR #1682: Multiple commits
-  - Remove wrapper definitions
-  - Extend example to check for sessionID and jobID
-PR #1678: Multiple commits
-  - Add --debug-daemon to prted options
-  - Add the rml and routed types to prte_info array
-PR #1674: Ensure cmd line errors provide error messages
-PR #1672: Multiple commits
-  - Correct default binding for map-by node/slot
-  - Provide better error message for cross-package binding
-  - Add missing command line directives
-  - Fix minor issues found in tests
-  - Silence compiler warning in test/connect
-  - Silence compiler warnings in examples
-  - Use exit macros in example dmodex
-  - Remove antiquated code that can cause problems
-  - Update sphinx requirements
-  - Remove the PMIX_SIZE_ESTIMATE support
-PR #1660: Correct function name when using older hwloc versions
-PR #1658: Multiple commits
-  - Properly handle overload-allowed
-  - Define MIN for environments that lack it
-  - Fix a typo in the README
-  - Fix Sphinx parsing error
-PR #1644: Multiple commits
-  - Add a spawn_multiple test
-  - Provide an estimate of the size of registration data
-  - Provide size estimates for collective operations that return data
-  - Add an option to display the available CPUs
-  - Fix generate_dist type
-  - Protect against missing attribute definition
-  - Set distances array type
-  - Add "parseable" display qualifier
-  - Generate distances by default
-  - Show help message when mapping too low
-  - Fix round-robin by obj with multiple cpus/rank
-  - Fix map-by slot and node for multiple cpus/rank
-PR #1633: First cut of Sphinx/ReadTheDocs docs
-PR #1626: Remove unused variable
-PR #1622: Multiple commits
-  - Don't error out on binding if not required
-  - Check for and track scheduler attachment
-  - Accept local group op upcalls
-  - Add an IOF stress test
-  - Protect against earlier PMIx versions
-  - Pass our event base for use by the PMIx library as an aux base
-  - Update to compatibility with PMIx master branch
-  - Silence warning - add return values
-  - Begin adding support for new server module fn pointer
-  - Designate the framework project as PRTE
-  - Minor touchups of the "fault" example
-  - Include topology in allocated node info
-  - Remove debug
-  - Extend query support to include allocation info
-  - Update version to 3.0.1
-PR #1600: Multiple commits
-  - Minor change to check_os_flavors
-  - Don't emit an error output if not needed
-PR #1597: Fix oac_check_package.m4
-PR #1595: build: fix bashisms in configure
-
-
-3.0.0 -- 13 Nov 2022
+3.0.0: TBD
 ------------------------------------
 **** NOTE: This release represents a major update to the PRRTE
 code base from prior releases. Many of the changes will be
@@ -134,160 +52,18 @@ may (likely) impact users. These include:
     is detected by PRRTE thru the environment.
   * a more extensive "--help" support system.
   * reduction in the "--rank-by" options.
-  * addition of new "--output", "--display", and "--runtime-options"
-    command line directives
+  * addition of new "--output" options
   * removal of the "mindist" mapper due to lack of a maintainer
-  * remove of man pages for PRRTE tools - this has been replaced
-    with a hierarchical help system. Thus, the command "prun --help"
-    will output the former man page for the prun command. Users
-    can then access more detailed help on any command line option
-    by simply adding it to the help command - i.e., "prun --help <option>"
 There are quite a few more changes than listed here, but they mostly
 pertain to code internals. However, these dictate that PRRTE v3.0.0
-require use of PMIx v4.2.2 or above to compile. Note that this doesn't
+require use of PMIx v4.2 or above to compile. Note that this doesn't
 impact the PMIx level associated with any applications executed by
 PRRTE - the only requirement is that applications link against a
 PMIx version that meets the PMIx cross-version requirements (i.e., a
 minimum of PMIx v2.5).
 
 Detailed changes:
 
-PR #1591: Multiple commits
-  - fix conflict
-  - remove unused vars in tools area
-  - adjust prterun to prefix ns: for dvm option
-  - Cleanup the --dvm option a bit
-  - Some more minor cleanups of --dvm
-PR #1586: Multiple commits
-  - Extend PBS allocation detection to cover Cobalt
-  - Broaden ability to require PRRTE to discover slots
-  - Fix default slots count when using --use-hwthread-cpus
-  - Cleanup schizo/prte help files
-  - schizo-ompi: add examples for verbose [dvm] block
-  - schizo:ompi: require arg when specifying dvm
-  - rank_file/LSF: Fix LSB_AFFINITY_HOSTFILE ranking policy
-  - Fix the load-balancing and span round-robin mappers
-  - doc: Fix --rank-by help output
-PR #1575: Ensure we init the available node bit tracker
-PR #1572: Fix some default map/rank policy settings
-PR #1570: Multiple commits
-  - rmaps/base: Fix segv on failed-map
-  - ompi/schizo: Fix deprecated message for rankfile
-  - Rankfile slots fix and LSF support fix for LSB_AFFINITY_HOSTFILE
-PR #1564: Add an example that retrieves node IDs
-PR #1562: Remove check for resilience that caused a segfault
-PR #1557: Update NEWS
-PR #1556: Multiple commits
-  - Support query of pset membership
-  - Cleanup grpcomm cruft
-  - remove some unused vars
-  - squash warnings
-  - unused var
-  - more unused vars
-  - remove unused vars and fix rc/ret typo
-  - misc unused var cleanups
-  - Reduce debugger confusion
-  - Pickup missing changes
-PR #1548: Multiple commits
-  - alps fixes for mca move
-  - schizo/ompi: Fix --report-pid/sid.
-  - Change --stop-in-* to take optional arguments.
-  - Update Open MPI mpirun help text
-  - Fix memory leaks in RML and at job termination.
-  - plm/tm: Fix build breakage
-PR #1539: Multiple commits
-  - Fix a typo and expand debugger example range to cover MPI
-  - ras/lsf: Fix build breakage
-  - Catch some more component updates
-  - Allow prterun to act as prun
-PR #1534: Multiple commits
-  - Fix print statement
-  - Push launch-agent CLI into the env
-  - Fix mapping by pe-list when oversubscribed
-PR #1530: Actually support the output-proctable option
-PR #1528: Add missing commits and new additions
-  - Complete help text on notifications
-  - BuildRequires: gcc
-  - Plug a memory leak
-  - schizo/ompi: Fix --use-hwthead-cpus option
-  - Fix --preload-binary
-PR #1522: Roll version to rc2
-PR #1518: schizo_ompi: use the OMPI_MCA_PREFIXES env var
-PR #1517: Multiple commits
-  - Remove the --with-ft configure option
-  - Replace prte_output with pmix_output
-  - Cleanup code clause identification
-  - Fix a few minor things
-  - Fix setting of runtime options
-  - Update the schizo/ompi component
-  - ompi/schizo: Fix --do-not-launch and --show-progress
-  - schizo/ompi: Fix --display topo
-  - schizo/ompi: Remove --display-devel-allocation
-  - Added specfile BuildRequires
-  - Add action to check commits on pull requests
-  - Restore the PRRTE application-level fault tolerance modes
-  - Update minimum PMIx version to v4.2.2
-PR #1501: Multiple commits
-  - Delete PRRTE MCA infrastructure
-  - Remove stale m4
-  - Handle default MCA params
-  - Rename component libraries and component struct names
-PR #1498: Remove leftover BINDTONONE SET debug code
-PR #1495: Multiple commits
-  - Revise passing of runtime options
-  - Add/modify some spawn-related tests
-  - Update to use PMIx CLI option macro and "check true"
-PR #1491: Multiple commits
-  - stream-buffering is an OMPI option, so make it an MCA
-  - Deprecated the --stream-buffering CLI option
-  - Fix hetero topology operations
-PR #1487: Use correct target path for Flex check
-PR #1485: Fix bug in not updating cached daemon topologies
-PR #1483: Fix oversubscription to not require binding
-PR #1481: Multiple commits
-  - Add support for querying process group attributes
-  - Initial work on --enable-devel-check
-  - Fix free in hostfile
-  - Continue progress on enable-devel-check
-PR #1474: Multiple commits
-  - Upgrade the cmd line option checker
-  - Enable the "show-progress" runtime option
-  - Fix the stack trace handshake
-  - Add more debug to the node list construction procedure
-PR #1469: Multiple commits
-  - Raise the minimum required PMIx level to v4.2.1
-  - Separate out the runtime controls options
-PR #1466: Multiple commits
-  - Ensure tools get string explanation for non-norm term
-  - stop-in-app does not require an argument
-  - Fix compile error with FT enabled
-  - schizo/ompi: Fix --do-not-launch
-  - Fix typo in help message
-  - schizo/ompi: Add translation for --merge-stderr-to-stdout
-  - java: add ease-of-use feature for ompi java apps
-  - Fix second pass on map-by obj
-  - Remove stale references to alloc-dev as display options
-  - Get the "--display topo" option to work
-PR #1456: Multiple commits
-  - Fix synthetic topo support and too early sanity check
-  - Correct the support check logic
-  - Remove the "report" qualifier from "--bind-to"
-  - Fix bind-upward check and cleanup bind output
-PR #1450: Update current state of PRRTE help docs
-PR #1448: Provide upgrade warning for ancient Slurm versions
-PR #1447: Fix app/local rank computation
-PR #1444: Fix keepalive CLI option
-PR #1441: Include the PMIx libdir in the ssh prefix operation
-PR #1440: Do not error out if lib is symlinked to lib64
-PR #1437: Avoid loopback interfaces if not needed
-PR #1433: Update the job object's binding policy
-PR #1431: Fix the tm configure logic
-PR #1429: Update help docs
-PR #1427: Cleanup the PRRTE rpm spec file
-PR #1423: Fix typo in ppr mapper
-PR #1417: Remove stale file reference
-PR #1415: Update release files for rc1
-PR #1414: Remove stale files
 PR #1412: Restrict PRRTE to PMIx v4.2+
 PR #1410: Remove schizo/hydra component
 PR #1409: Add new output options to help string