[flang-dev] Flang Biweekly Sync - Notes from the November 27th, 2019 call
Gary Klimowicz via flang-dev
flang-dev at lists.llvm.org
Fri Nov 29 18:02:24 PST 2019
Flang team,
Here are the notes from the November 27th, 2019 Flang biweekly call.
Please be sure to let us know if you have any questions and agenda topics for the next call.
The next Flang biweekly call will be Wednesday, December 11th, 2019 at 8:30 AM Pacific Time.
The next Flang Technical Community call will be Monday, December 2nd, 8:30 AM Pacific Time. Appentra will be discussing experiences with the LLVM flang compiler for tooling.
Thank you.
Gary Klimowicz
Agenda
* SC 2019 recap
* Flang Community Technical Call - Appentra
* F18 development update
* Flang development update
Highlights
* Meeting logistics information is now at the end of this email.
* Supercomputing 2019 recap
* The LLVM in HPC BoF session at SC 2019<https://sc19.supercomputing.org/program/birds-of-a-feather/> was Wednesday, November 20, 5:15-6:45 PM. Here’s the link: https://sc19.supercomputing.org/session/?sess=sess331.
* Hal Finkel reported that there was a lot of interest and a lot of questions about the new LLVM flang compiler. Many of the questions were about features and functionality, when they would be available, and how people can contribute.
* There is a strong desire to get the source code upstreamed into the monorepo, balanced with getting the commits into “LLVM shape”.
* There was a question from AMD about ongoing contributions from NVIDIA to original Flang OpenMP. While some minimal changes have come from NVIDIA refactoring, there are no further plans for functionality updates.
* Flang Community Technical Biweekly Call
* Meeting logistics are at the end of this email.
* The next scheduled call for December 2nd is from Appentra, with feedback on the use of f18 for tooling.
* Please let Hal Finkel know topics you would like brought up, or edit the Google doc to add any topics you would like to present.
* New Flang update
* Implemented user-defined assignment.
* Finished call semantic checking.
* Committed I/O runtime API header.
* Began support for forward references to derived types.
* Began work on logical expression lowering.
* Continued work on DO loop semantic checks.
* Continued work on FIR definition.
* Original Flang Dev news and updates
* More PGI bug fixes.
* Changes to better support Arm processors.
* #577 Procedure pointers to internal procedures.
* Fix for issue with STAT= on ALLOCATE statements.
* Fixes for procedure pointers.
* Coming soon: The LLVM license update.
Where we can use your help
* Topics for the next biweekly call and technical community call.
* Let us know if you have presentations or meetings relevant to the Flang community. We will publicize these in these notes.
Details (Updates below are in bold)
1. Flang on GitHub
· There have been 1,665 total commits (1,665 in the previous update).
· There are 490 people who have “starred” the flang repo and 71 forks.
· 201 people are following f18 in 36 forks. There have been 2,776 commits.
1. Flang Community Technical Call
* The last call was November 4th, a recap of the LLVM Developer Meeting. Most of the discussion was on upstreaming Flang into LLVM, preserving history, and treating MLIR as an external dependency for the time being.
* Future Topics and tentative dates:
* December 2nd: Feedback on tooling (Appentra)
* LLVM transition (mailing lists, git/monorepo, etc.)
* Potential topics for future calls include build times for the compiler and status updates on MLIR.
1. Flang community
* Website: www.flang-compiler.org<http://www.flang-compiler.org> redirects to http://github.com/flang-compiler/flang.
* Wiki: https://github.com/flang-compiler/flang/wiki is the Flang wiki page where the most up-to-date build instructions and news can be found.
* Issues: Bug reports for current Flang can be reported to https://github.com/flang-compiler/flang/issues.
* LLVM mailing lists: There are now two mailing lists hosted by llvm.org:
* The flang-dev mailing list has been set up. Subscribe at https://lists.llvm.org/cgi-bin/mailman/listinfo/flang-dev. These biweekly notes are now sent to this email list.
* The flang-commits mailing list has been set up. Subscribe at https://lists.llvm.org/cgi-bin/mailman/listinfo/flang-commits.
* Slack: flang-compiler.slack.com is used for informal real-time communication in the Flang community. Add yourself via the invite <https://join.slack.com/t/flang-compiler/shared_invite/enQtMjExOTEyMzQ3MjIxLTZkMDNlODc2YjgxZDY1NTQ2YWJiZWVhOGU2YWY0NDNiOWY0ZjQzNjYyZTNlZjBlNjgxY2U1ZjFjMmY5NzIwMzA> link.
* Supercomputing 2019
* Hal Finkel reported that there was a lot of interest and a lot of questions about the new LLVM flang compiler. Many of the questions were about features and functionality, and when they would be available and how people can contribute.
* There is a strong desire to get the source code upstreamed into the monorepo, balanced with getting the commits into “LLVM shape”.
* There was a question from AMD about ongoing contributions from NVIDIA to original Flang OpenMP. While some minimal changes have come from NVIDIA refactoring, there are no further plans for functionality updates.
1. What we’re up to with new LLVM Flang Fortran 2018 compiler
* On GitHub: The source code f18 is available on flang-compiler.github.com/f18<https://github.com/flang-compiler/f18/>. It’s recommended to start in the “documentation” directory.
* Status: We continue development on the new Fortran 2018 front-end. Recent work includes:
* Implemented user-defined assignment.
* Finished call semantic checking.
* Committed I/O runtime API header.
* Began support for forward references to derived types.
* Began work on logical expression lowering.
* Continued work on DO loop semantic checks.
* Continued work on FIR definition.
* We continue to follow the MLIR project for the flang intermediate representation.
* LLVM Project:
* The flang-dev email list has been set up on llvm.org. You can subscribe here: https://lists.llvm.org/cgi-bin/mailman/listinfo/flang-dev.
* The flang-commits email list has also been set up. There is no content yet, as we haven’t migrated to the LLVM monorepo.
* No progress on adding flang to the LLVM repo. Flang will migrate to the LLVM monorepo with its previous history and commits (as clang did, as MLIR plans). We would like to have commits preserve relevant history, and be a complete set of functionality (include relevant tests). This roughly translates into squashing commits that correspond to pull requests, but isn’t exactly a mechanical process.
* Hal recommended we not be pedantic about this, but try to get Flang into the monorepo by the end of the year. He also suggested that we find ways for the community to help with this process, to help get it done, and to help people become familiar with the code.
* We are not waiting for MLIR to be upstreamed to LLVM. Discussion at SC 2019 was that we would have the flang CMake look for an installed MLIR, and change how we find MLIR once it is in the monorepo.
1. What we’re up to with current Flang
* We will be making updates for the new Apache 2.0 license with LLVM exceptions. Changes here are likely to touch almost every file in Flang.
* We are in the process of incorporating changes to support LLVM 9.0.
* We will be opening up Flang to more committers soon. We would like to get the license changes into Flang before adding committers.
* We will be decoupling PGI and Flang development to shorten turnaround on pull requests and other development activity.
* Recent fixes pushed to Flang
* More PGI bug fixes
* Changes to better support Arm processors
* #577 Procedure pointers to internal procedures
* Fix for issue with STAT= on ALLOCATE statements
* Fixes for procedure pointers
* Upcoming changes to Flang (in review):
* LLVM Apache 2.0 with LLVM exceptions licensing change
* Pull request #573 – libpgmath: make it clang friendly; addresses bug #557 (reviewing)
* Pull request #695 – Remove duplicate files
* Assigned for work:
* #402 Compiler error on regular code
* #404 Internal compiler error with polymorphic subroutine parameters and nested derived types
* #426 Failures in #include of OpenMPI's mpif.h
* #461 OpenMP TeaLeaf_ref: local pointer variable not captured into parallel region
* #477 Pull request for module variable import debug test
* #498 Error while compiling libpgmath
* #602 fd_sincos_scalar.cpp: "error: ‘__builtin_complex’ was not declared in this scope"
* #604 Flang errors out on a final subroutine with parameterized derived types
* #608 gdwarf-<version> flag is not honoured
* #610 No debug generated for namelists
* #662 Regression : Segfault in taskloop code
* #720 Segfault on sourced allocation of polymorphic variable
* #780 Linking CXX shared library libpgmath.dylib fails on macOS
* #787 ICE with array operations inside NORM2
* #807 possible memory corruption in function ompaccel_copy_arraydescriptors
* Pull request #573 libpgmath: make it clang friendly; addresses bug #557
* Pull request #695 runtime: remove duplicate files
* Pull request #710 Test case for capturing procedure pointers to OpenMP parallel regions
* Pull request #719 libpgmath: windows and osx support
* Outstanding pull requests:
* #521 Add OpenBSD as a recognized OS (pending test on new OpenBSD systems internally)
* #529 runtime: PoC: generate call to LLVM intrinsic instead of calling runtime functions when possible; handling NINT for a beginning (proposed expansion of this pull request)
* #530 AArch64: disable test cases invalid on aarch64 (working on)
* #532 runtime: remove duplicate files (need to validate on all compilers)
* #546 Loosen ties to x86 (working on)
* #572 libpgmath: respect LIBPGMATH_WITH_GENERIC setting; fixes #559 (reviewing)
* #592 add support for macOS (awaiting CLA)
* #622 Add explicit braces to avoid dangling else problem (failed NVIDIA testing)
* #642 AArch64: significantly improve formatted input performance by using optimized libc functions on ARM64
* #647 Introduce buffered I/O and replace getc with buffered read
* #655 Add missing signed integer to a pointer conversion in DATA statement
* #658 Fix in preprocessing for Flexi app
* #659 Changes to support transpose intrinsic during initialization
* #660 Enable support for simd directives
* #672 Support isnan() GNU extension (Arm reviewing)
* #689 Fix for segfault in multiple entries OpenMP functions
* #693 runtime: ARM64 is a 64-bit architecture
* #694 libpgmath sanity: do not call sleep() in Linux systems
* #695 runtime: remove duplicate files
* #701 Fix for ICE in taskloop
* #702 Fix for random_seed when seeds have zeroes in LSB
* #710 Test case for capturing procedure pointers to OpenMP parallel regions
* #717 Support building libpgmath with clang 6
* #725 Check for gregset_t before using it
* #726 Remove linux references where unneeded
* #737 TRAILZ function added to the fortran compiler (resubmitted)
* #738 Fix for issue #550
* #742 Implement UNROLL directive
* #751 Fix for len intrinsic returning int*8 in some cases
* #752 Fix for ICE in atomic instruction generation
* #756 Fix ICE interf:new_symbol_and_link, symbol not found
* #772 Fix unresolved sprintf, fprintf error
* #773 Support for isnan GNU extension: Generate isnanf() & isnan() libm calls
* #776 The BSDs name 64-bit x86 as amd64
* #778 Fixing NCAR test problems with error tolerance lower than E-12
* #787 ICE with array operations inside NORM2
* #816 LLVM9 debuginfo change
* Public plan: NVIDIA has published a brief plan for upcoming Flang work on github.io at https://github.com/flang-compiler/flang/wiki/Roadmap. This has been recently updated to reflect current Flang and F18 priorities and to provide better context for the project and its long-term goals.
* Creating pull requests: Until we complete the relicensing for current Flang, NVIDIA requires contributor agreements from developers wishing to make pull requests. The Individual CLA is here: https://flang-compiler.github.io/files/Flang-ICLA-v4.pdf. The Corporate CLA is here: https://flang-compiler.github.io/files/Flang-CCLA-v2.pdf. Send completed CLAs to flang-cla at nvidia.com<mailto:flang-cla at nvidia.com>. Assuming the pull request builds and passes “make check-flang”, NVIDIA will take on the internal testing of the pull request using the full test suite available internally and provide feedback to the contributor as appropriate. Once we have the quality of the pull request established, it will be incorporated into Flang.
* OpenMP 4.5 support:
o No update
* Outstanding issues:
* Please refer to the Issues section on GitHub for flang-compiler/flang<https://github.com/flang-compiler/flang/issues> and flang-compiler/flang-driver<https://github.com/flang-compiler/clang/issues>.
* Debug Metadata:
* Debug metadata for PURE, ELEMENTAL and RECURSIVE subprograms has been added to Flang in open source.
* LLVM Debug metadata changes posted to LLVM Phabricator for review; responding to review comments.
* https://reviews.llvm.org/D54114 [CHARACTER intrinsic data type] Review for this is not complete, and we would like to get additional comments or alternate proposals.
* Packaging and Binary Releases:
* Binary release: We post binary releases at least quarterly. March binary releases of Flang for x86-64 and OpenPOWER (LLVM 7.0) have been posted to http://github.com/flang-compiler/flang/releases.
* Spack: Spack packages for Flang have been reviewed by the Spack team and pulled into Spack for flang_20180921. We are still working through issues with defining the spack package for flang. Brian Homerding is starting to work on Spack changes for the LLVM and Flang for the LLVM Git monorepo.
* Testing updates:
* No update.
* Application updates:
* No update.
1. Fortran 2008 features
* Pointer default initialization. Work continues. Support for procedure pointer default initialization in modules and submodules has been pushed to Flang. We plan to work on scalar pointer initialization next.
The following items are considered complete. Please report GitHub issues if you identify any problems.
* EXECUTE_COMMAND_LINE intrinsic.
* BLOCK construct.
* Allocation of polymorphic variables.
* G0 edit descriptor
* NORM2 intrinsic
* ERROR STOP
* Pointers to internal procedures
* SUBMODULE
* Serial implementation of F2018 DO CONCURRENT
* BACK optional argument support for MINLOC/MAXLOC
1. Call Logistics
Flang Community Biweekly Call
* General updates on the Flang project, both LLVM Flang and current Flang.
* Biweekly WebEx hosted by NVIDIA (which these notes cover).
* The meeting link for this biweekly call is https://nvmeet.webex.com/nvmeet/j.php?MTID=mb4edb8c799f69ec2dc0554acc969a162.
Flang Community Technical Biweekly Call
* Technical topics call.
* Hal Finkel hosts this call on Bluejeans<https://bluejeans.com/625064848>, Mondays 8:30 Pacific Time, on the weeks alternating with regular Flang biweekly call. The notes for all these calls can be found in Google Docs here<https://docs.google.com/document/d/1Z2U5UAtJ-Dag5wlMaLaW1KRmNgENNAYynJqLW2j2AZQ/>.
Meeting URL: https://bluejeans.com/625064848?src=join_info
Meeting ID: 625 064 848
Want to dial in from a phone? Dial one of the following numbers:
+1.408.317.9254 (US (San Jose))
+1.866.226.4650 (US Toll Free)
(see all numbers - https://www.bluejeans.com/premium-numbers)
Enter the meeting ID and passcode followed by #
Connecting from a room system?
Dial: bjn.vc or 199.48.152.152 and enter your meeting ID & passcode
-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information. Any unauthorized review, use, disclosure or distribution
is prohibited. If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/flang-dev/attachments/20191130/5d55f20f/attachment-0001.html>
More information about the flang-dev
mailing list