Generate bigcount interfaces for Fortran and C #12226

jtronge · 2024-01-10T21:16:33Z

This adds scripts for generating the C and Fortran mpi_f08 API bindings from template files, while also generating bigcount interfaces for those that require them. On the Fortran side it also adds support for TS 29113 when possible, allowing for better Fortran array handling that matches the standard (some files were imported from PR #10302).

Python >=3.6 is required for running these scripts, which is only necessary when the binding files have not already been generated. Users of the distribution tarball should not need to generate these files and thus should not require Python.

We used https://github.com/cea-hpc/pcvs-benchmarks and the MPI4PY test suite to help ensure all big count interfaces (C and Fortran) are being generated.

PR #12033 is a previous version of this focused specifically on ABI support.

jsquyres

I haven't finished reading / understanding / reviewing both generate_bindings.py scripts yet, but since I'm so late for this and the clock is ticking on @jtronge's availability, I'm going to submit what I have so far.

jsquyres · 2024-01-30T20:12:49Z

ompi/mpi/fortran/use-mpi-f08/Makefile.am


 lib@OMPI_LIBMPI_NAME@_usempif08_la_SOURCES = \
        $(mpi_api_files) \
        mpi-f08.F90

 # These are generated; do not ship them
-nodist_lib@OMPI_LIBMPI_NAME@_usempif08_la_SOURCES =
+# nodist_lib@OMPI_LIBMPI_NAME@_usempif08_la_SOURCES =


Go ahead and delete this line; don't just comment it out.

jsquyres · 2024-01-30T20:13:41Z

ompi/mpi/fortran/use-mpi-f08/Makefile.am


 # JMS Somehow this variable substitution isn't quite working, and I
 # don't have time to figure it out.  So just wholesale copy the file
 # list.  :-(
-#pmpi_api_files = $(mpi_api_files:%=profile/p%)
+#pmpi_api_files = $(mpi_api_files:%=p%)


If you can figure out why this substitution isn't working, go ahead and uncomment it. Otherwise, it would probably be ok to delete this whole commented-out block.

jsquyres · 2024-02-07T14:47:31Z

ompi/include/mpi.h.in

@@ -1389,7 +1389,7 @@ OMPI_DECLSPEC extern struct ompi_predefined_datatype_t ompi_mpi_ub;
 /*
 * MPI API
 */
-
+#ifndef OMPI_NO_MPI_PROTOTYPES


This is the only file in the whole tree where I see this macro used -- where does it come from?

If we're using it in our infrastructure, we prefer to always define boolean-like macros to be 0 or 1 (vs. defined or undefined). There's less of a chance for error that way.

This is left over from the standard ABI code; I can remove it for now.

IIRC the reason I didn't use a 0/1 macro there is because it would require adding the macro in mpicc or forcing the end user to add it. Both the standard ABI and the ompi ABI code were using mpi.h since some types defined there needed to be used in both versions, but the prototypes cannot be the same.

When we start working again on the standard ABI code, it might be better to refactor mpi.h out into multiple files to make this easier.

Ok, gotcha -- those are good reasons. If you end up keeping something like this, having a comment to explain the deviation from our norm (i.e., using #ifndef) would be good.

jsquyres · 2024-02-07T14:48:21Z

ompi/mpi/c/Makefile.am

@@ -41,6 +41,36 @@ endif

 headers = bindings.h

+prototype_sources = \


Probably worth putting a comment before this line explaining what this macro is for -- it may not be intuitive to the casual code reader that these are all the files with bigcount APIs.

jsquyres · 2024-02-07T14:49:26Z

ompi/mpi/c/Makefile.am

@@ -60,7 +90,7 @@ endif
 # List of all C files that have profile versions
 #


Might also be good to include in this comment before the interface_profile_sources that the ompi_FOO.c files have the ompi_ prefix because they were generated.

jsquyres · 2024-02-07T15:46:48Z

ompi/mpi/c/generate_bindings.py

+
+def main():
+    if len(sys.argv) < 2:
+        # Fix required for Python 3.6


What does this comment mean -- does this mean that this script does not work if Python is < v3.6?

This a fix for Python v3.6, since the argparse package has some limitations for that version. I have not tested this code below v3.6, unless the CI is using older versions.

jsquyres · 2024-02-07T15:52:09Z

ompi/mpi/c/generate_bindings.py

+        sys.exit(1)
+
+    parser = argparse.ArgumentParser(description='generate ABI header file and conversion code')
+    subparsers = parser.add_subparsers()


I'm a bit confused by this CLI interface. Running generate_bindings.py --help shows this:

$ ./ompi/mpi/c/generate_bindings.py --help usage: generate_bindings.py [-h] {header,source} ... generate ABI header file and conversion code positional arguments: {header,source} options: -h, --help show this help message and exit

Which doesn't give me a lot of information about how to use this script. In ompi/mpi/c/Makefile.am, I see it used like this:

$(PYTHON) $(srcdir)/generate_bindings.py source ompi $< > $@

I think that means it's falling down into the "source" sub-parser.

What's the header sub-parser for? Is that for future functionality or something?

Would it be possible to use more traditional --foo=bar kinds of CLI arguments, perchance?

I can try to make --help a little more descriptive about the sub-parsers. The source sub-parser is designed to generate the source code from a template, in this case for the ompi ABI. The header sub-parser is designed to auto-generate the prototype definitions, although it was only used previously for the standard ABI code (it could be useful for the ompi ABI as well).

The standard ABI functionality was completely removed from the Makefiles, but I've left the functionality in the script, that's why some of these arguments are still there but unused.

jsquyres · 2024-02-07T15:52:33Z

ompi/mpi/c/generate_bindings.py

+        print('ERROR: missing subparser argument (see --help)')
+        sys.exit(1)
+
+    parser = argparse.ArgumentParser(description='generate ABI header file and conversion code')


Is this generating ABI files, or BigCount stuff? I thought this PR was about BigCount and ABI was future...?

...after reading a bunch more of generate_bindings.py, I see that it looks like a lot of the ABI functionality is included, but it doesn't look like it's complete...? E.g., I don't see any other kind of ABI infrastructure (e.g., creating the other libraries, stacking the libraries and implementations like we talked about, etc.).

If we're too late in the game to fully dis-entangle ABI and bigcount into 2 wholly separate commits (with one almost certainly building upon the other), it would be good to explain that this commit has elements of stuff that will be used in an upcoming ABI commit and it wasn't separated because of lack of time blah blah blah. Otherwise, the reader (e.g., me) is left wondering why a "Bigcount" commit has a bunch of unused / half-done ABI stuff included.

Sorry I should have been more clear with my commit message. I removed all ABI-specific changes from the make-system, but the script still includes the unused ABI code. This seemed the best route to take, instead of completely removing the extra ABI code, which is not affecting how ompi is being built right now. I can remove it though.

jsquyres · 2024-02-07T15:54:24Z

ompi/mpi/c/generate_bindings.py

+    args = parser.parse_args()
+
+    # Always add the header
+    print('/* THIS FILE WAS AUTOGENERATED BY ompi/mpi/c/abi.py. DO NOT EDIT BY HAND. */')


Additionally, it might be better to have a traditional --outfile=FILENAME kind of CLI arg that indicates the name of the file that you want to write to. This leaves stdout available for info/verbose/debug kinds of output.

jsquyres · 2024-02-07T16:11:53Z

ompi/mpi/c/generate_bindings.py

+* Functions requiring a bigcount implementation should have type COUNT in
+  place of MPI_Count or int for each count parameter. Bigcount functions will
+  be generated automatically for any function that includes a COUNT type.
+"""


Please add a larger comment here at the top explaining the purpose of this script, and at least a high-level overview of what it does / how it works. This is a very large script; having an intro to it at the top would be most helpful to the reader (e.g., me).

jsquyres · 2024-02-07T16:15:30Z

Should #12033 be closed (if this PR wholly replaces it)?

jtronge · 2024-02-07T17:14:40Z

I think I'll try to remove the extra standard ABI code from the script, as well as the additional items you caught @jsquyres.

I don't think this code completely replaces PR #12033, especially since that one includes the refactor of libmpi into multiple libraries that we discussed.

hppritcha · 2024-02-08T21:23:34Z

config/ompi_setup_mpi_fortran.m4

+    AS_IF([test $OMPI_TRY_FORTRAN_BINDINGS -ge $OMPI_FORTRAN_USEMPIF08_BINDINGS],
+          [OMPI_FORTRAN_CHECK_TS([OMPI_FORTRAN_HAVE_TS=1])])
+
+    AC_SUBST(OMPI_MPI_SUBARRAYS_SUPPORTED)


I think lines 458 and 459 should b moved below line 472 since it is there that we may be setting OMPI_MPI_SUBARRAYS_SUPPORTED and OMPI_MPI_ASYNC_PROTECTS_NONBLOCKING to true.

hppritcha · 2024-02-12T17:29:45Z

docs/developers/bindings.rst

+The Fortran file ``api_f08_generated.F90`` contains all the internal subroutine
+definitions, each of which makes a call into corresponding C functions. Two
+different C files are generated: ``api_f08_ts_generated.c`` contains support
+for compilers with TS 29113 support, allowing the use of ``CFI_cdesc_t`` types;


might be nice to add a hyperlink here, e.g. https://fortranwiki.org/fortran/show/Fortran+2018

jtronge · 2024-02-14T22:44:16Z

/azp run

github-actions · 2024-04-01T21:15:56Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-01T21:36:33Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-01T21:37:14Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-03T15:42:53Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-16T21:18:09Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-17T16:35:50Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-17T17:56:04Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-18T16:37:38Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-22T16:44:26Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-22T16:44:41Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-23T14:39:57Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-24T17:52:23Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

github-actions · 2024-04-24T21:10:27Z

Hello! The Git Commit Checker CI bot found a few problems with this PR:

5127029: Add ELEMENT_COUNT for MPI_Get_elements_x

check_signed_off: does not contain a valid Signed-off-by line

fc151a8: Add DATATYPE, STATUS, and GREQUEST_ types*

check_signed_off: does not contain a valid Signed-off-by line

e673867: Add int, char, and MPI_Info-related types

check_signed_off: does not contain a valid Signed-off-by line

01fa44e: Add PARTITIONED_COUNT type

check_signed_off: does not contain a valid Signed-off-by line

Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks!

hppritcha · 2024-08-21T19:08:12Z

bot:ompi:retest

hppritcha · 2024-09-29T14:31:13Z

bot:ompi:retest

hppritcha · 2024-10-02T17:32:45Z

odd the jenkins community CI is failing in an autogen.pl run:

3. Running template-generating scripts

=== ompi/include/mpif-values.pl
Cannot find executable ompi/include/mpif-values.pl!
Aborting.

hppritcha · 2024-10-02T17:33:00Z

bot:ompi:retest

related to PR open-mpi#12226 don't pay much attention to the top level MPI_Buffer_detach etc. code as it will be redone in open-mpi#12226. Signed-off-by: Howard Pritchard <[email protected]>

This adds scripts for generating the C API bindings from template files, while also generating bigcount interfaces for those that require them. The binding script also include initial support for the mpi_f08 Fortran bindings, but doesn't yet make any changes to fortran/use-mpi-f08 Python >=3.6 is required for running these scripts, which is only necessary when the binding files have not already been generated. Users of the distribution tarball should not need to generate these files and thus should not require Python. Co-authored-by: mphinney1100 <[email protected]> Co-authored-by: Howard Pritchard <[email protected]> Signed-off-by: Jake Tronge <[email protected]>

hppritcha · 2024-10-19T17:03:41Z

@jsquyres its ready!

This updates fortran/use-mpi-f08 to generate most of the Fortran bindings from a script and template files. It also adds support for Fortran TS 29113 when possible, allowing for better Fortran array handling that matches the standard. The C files were imported from PR open-mpi#10302 and converted to templates to be fed into the binding script. Co-authored-by: Gilles Gouaillardet <[email protected]> Co-authored-by: Howard Pritchard <[email protected]> Signed-off-by: Jake Tronge <[email protected]>

github-actions bot added the Target: main label Jan 10, 2024

hppritcha requested a review from jsquyres January 10, 2024 21:18

jtronge force-pushed the bigcount branch from 1a9a108 to 37ada30 Compare January 11, 2024 15:29

jsquyres mentioned this pull request Jan 30, 2024

mpi_f08 configure tests failing for NAG #9795

Open

jsquyres reviewed Feb 7, 2024

View reviewed changes

jtronge mentioned this pull request Feb 7, 2024

Standard ABI generation code and library refactor. #12033

Draft

hppritcha reviewed Feb 8, 2024

View reviewed changes

hppritcha reviewed Feb 12, 2024

View reviewed changes

hppritcha mentioned this pull request Feb 14, 2024

COLL framework needs work to support MPI Bigcount #12336

Closed

jtronge force-pushed the bigcount branch from 2ae81f2 to cad1e4e Compare April 1, 2024 20:14

jtronge force-pushed the bigcount branch from 66909b4 to e00da70 Compare April 22, 2024 16:44

jsquyres mentioned this pull request Jul 13, 2024

Fortran: first try if type(*), dimension(..) works as-is #11591

Draft

jtronge force-pushed the bigcount branch from 48b2f53 to f49c835 Compare July 26, 2024 17:07

jtronge force-pushed the bigcount branch 2 times, most recently from 0590dd1 to c7c06e2 Compare August 6, 2024 13:21

jtronge force-pushed the bigcount branch 3 times, most recently from fa6f257 to 2811c1f Compare August 13, 2024 15:06

wenduwan added the Target: v6.0.x label Aug 13, 2024

github-actions bot removed the Target: v6.0.x label Aug 13, 2024

jtronge force-pushed the bigcount branch 2 times, most recently from 60ea322 to ab496c4 Compare August 15, 2024 21:31

jtronge force-pushed the bigcount branch from bd592ca to e29b037 Compare September 24, 2024 20:59

jtronge force-pushed the bigcount branch 2 times, most recently from b86d7de to c3e5885 Compare October 14, 2024 12:59

hppritcha added the MPI-4.0 label Oct 14, 2024

hppritcha mentioned this pull request Oct 14, 2024

big count: embiggen the pml bsend attach/detach #12860

Merged

jtronge force-pushed the bigcount branch from c3e5885 to 881e2f2 Compare October 15, 2024 12:13

jtronge force-pushed the bigcount branch from 3706021 to 89dc982 Compare October 18, 2024 22:03

jtronge marked this pull request as ready for review October 18, 2024 22:11

hppritcha requested a review from jsquyres October 19, 2024 17:03

hppritcha self-requested a review October 19, 2024 17:04

hppritcha approved these changes Oct 19, 2024

View reviewed changes

hppritcha mentioned this pull request Oct 23, 2024

fortran/use-mpi-f08: add support for Fortran 2018 ISO_Fortran_binding.h try number two #10302

Closed

jtronge force-pushed the bigcount branch from 38c4708 to 1926928 Compare October 29, 2024 12:36

		@@ -41,6 +41,36 @@ endif

		headers = bindings.h

		prototype_sources = \

		@@ -60,7 +90,7 @@ endif
		# List of all C files that have profile versions
		#

Generate bigcount interfaces for Fortran and C #12226

Are you sure you want to change the base?

Generate bigcount interfaces for Fortran and C #12226

Conversation

jtronge commented Jan 10, 2024 • edited Loading

jsquyres left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jtronge Feb 7, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jsquyres commented Feb 7, 2024

jtronge commented Feb 7, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jtronge commented Feb 14, 2024

github-actions bot commented Apr 1, 2024

github-actions bot commented Apr 1, 2024

github-actions bot commented Apr 1, 2024

github-actions bot commented Apr 3, 2024

github-actions bot commented Apr 16, 2024

github-actions bot commented Apr 17, 2024

github-actions bot commented Apr 17, 2024

github-actions bot commented Apr 18, 2024

github-actions bot commented Apr 22, 2024

github-actions bot commented Apr 22, 2024

github-actions bot commented Apr 23, 2024

github-actions bot commented Apr 24, 2024

github-actions bot commented Apr 24, 2024

hppritcha commented Aug 21, 2024

hppritcha commented Sep 29, 2024

hppritcha commented Oct 2, 2024

hppritcha commented Oct 2, 2024

hppritcha commented Oct 19, 2024

jtronge commented Jan 10, 2024 •

edited

Loading

jtronge Feb 7, 2024 •

edited

Loading