Merge pull request #351 from LLNL/v0.4.0-rc

v0.4.0 Release

.mailmap

@@ -1,15 +1,19 @@
 Aaron Black <black27@llnl.gov>                  Aaron C. Black <black27@llnl.gov>
 Adam J. Kunen <kunen1@llnl.gov>                 Adam J. Kunen <kunen1@sierra1626.llnl.gov>
+Aesha Parekh <aeshabparekh@gmail.com>           Aesha Parekh <29237123+aeshapar@users.noreply.github.com>
 Arlie Capps <capps2@llnl.gov>                   Arlie Capps <capps2@cab669.llnl.gov>
 Arlie Capps <capps2@llnl.gov>                   Arlie Capps <capps2@cab689.llnl.gov>
 Arlie Capps <capps2@llnl.gov>                   Arlie Capps <48997041+agcapps@users.noreply.github.com>
+Arlie Capps <capps2@llnl.gov>                   Arlie Capps <agcapps@ucdavis.edu>
 Austin Stromme <stromme1@llnl.gov>              Austin James Stromme <stromme1@llnl.gov>
 Austin Stromme <stromme1@llnl.gov>              Austin James Stromme <stromme1@cab689.llnl.gov>
 Austin Stromme <stromme1@llnl.gov>              Austin James Stromme <stromme1@cab669.llnl.gov>
 Benjamin Curtice Corbett <corbett5@llnl.gov>    Benjamin Curtice Corbett <corbett5@cab668.llnl.gov>
 Benjamin Curtice Corbett <corbett5@llnl.gov>    Ben Corbett <corbett5@llnl.gov>
 Benjamin Curtice Corbett <corbett5@llnl.gov>    Benjamin Corbett <corbett5@llnl.gov>
 Benjamin Curtice Corbett <corbett5@llnl.gov>    Ben Corbett <32752943+corbett5@users.noreply.github.com>
+Brian Manh Hien Han <han12@llnl.gov>            Brian Han <han12@llnl.gov>          
+Brian Manh Hien Han <han12@llnl.gov>            Brian Manh Hien Han <han12@rzansel19.coral.llnl.gov>
 Chris White <white238@llnl.gov>                 Christopher A. White <white238@llnl.gov>
 Cyrus D. Harrison <harrison37@llnl.gov>         Cyrus Harrison <cyrush@llnl.gov>
 Cyrus D. Harrison <harrison37@llnl.gov>         Cyrus Harrison <harrison37@llnl.gov>
@@ -18,11 +22,15 @@ Evan Taylor Desantola <desantola1@llnl.gov>     Evan Taylor DeSantola <desantola
 George Zagaris <zagaris2@llnl.gov>              George Zagaris <george.zagaris@gmail.com>
 Kenneth Weiss <weiss27@llnl.gov>                Kenneth Weiss <kweiss@llnl.gov>
 Kenneth Weiss <weiss27@llnl.gov>                Kenny Weiss <kennyweiss@users.noreply.github.com>
+Kenneth Weiss <weiss27@llnl.gov>                Kenny Weiss <kenny@kennyweiss.com>
+Keith Edward Healy <healy22@llnl.gov>           keithhealy <50376825+keithhealy@users.noreply.github.com>
+Keith Edward Healy <healy22@llnl.gov>           Keith Edward Healy <healy22@borax5.llnl.gov>
 Lee A. Taylor <taylor16@llnl.gov>               Lee A. Taylor <taylor@cab688.llnl.gov>
 Lee A. Taylor <taylor16@llnl.gov>               Lee A. Taylor <taylor@cab688.llnl.gov>
 Lee A. Taylor <taylor16@llnl.gov>               Lee Taylor <ltaylor@llnl.gov>
 Lee A. Taylor <taylor16@llnl.gov>               Lee Taylor <taylor16@llnl.gov>
-Matt Larsen <larsen30@llnl.gov>		        mclarsen <larsen30@llnl.gov>
+Lee A. Taylor <taylor16@llnl.gov>               Lee Taylor <31938638+ltaylor16@users.noreply.github.com>
+Matt Larsen <larsen30@llnl.gov>		            mclarsen <larsen30@llnl.gov>
 Noah S. Elliott <elliott22@llnl.gov>            Noah Elliott <elliott22@llnl.gov>
 Punita P. Sinha <sinha2@llnl.gov>               Punita P. Sinha <sinha2@cab687.llnl.gov>
 Punita P. Sinha <sinha2@llnl.gov>               Punita P. Sinha <sinha2@cab669.llnl.gov>
@@ -44,5 +52,6 @@ Axom Shared User <axom-dev@llnl.gov>            Aaron Black <atk@cab687.llnl.gov
 Axom Shared User <axom-dev@llnl.gov>            Richard D. Hornung <atk@rzalastor2.llnl.gov>
 Axom Shared User <axom-dev@llnl.gov>            Richard D. Hornung <atk@cab687.llnl.gov>
 
-uncrustify-robot <axom-dev@llnl.gov>            uncrustify-robot <asctoolkit-dev@llnl.gov>
+format-robot <axom-dev@llnl.gov>                uncrustify-robot <axom-dev@llnl.gov>
+format-robot <axom-dev@llnl.gov>                uncrustify-robot <asctoolkit-dev@llnl.gov>
 shroud-robot <axom-dev@llnl.gov>                shroud-robot <asctoolkit-dev@llnl.gov>

README.md

@@ -90,6 +90,10 @@ PackageName: fmt
 PackageHomePage: https://github.com/fmtlib/fmt  
 PackageLicenseDeclared: BSD-2-Clause  
 
+PackageName: sol  
+PackageHomePage: https://github.com/ThePhD/sol2  
+PackageLicenseDeclared: MIT License
+
 PackageName: sparsehash  
 PackageHomePage: https://github.com/sparsehash/sparsehash  
 PackageLicenseDeclared: BSD-3-Clause  

RELEASE

@@ -1,6 +1,6 @@
 *******************************************************************************
 
-Axom: ................................, version 0.3.3
+Axom: ................................, version 0.4.0
 
 Copyright (c) 2017-2020, Lawrence Livermore National Security, LLC. 
 Produced at the Lawrence Livermore National Laboratory.

RELEASE-NOTES.md

@@ -7,6 +7,133 @@ The format of this file is based on [Keep a Changelog](http://keepachangelog.com
 
 The Axom project release numbers follow [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
+## [Version 0.4.0] - Release date 2020-09-22
+
+### Added
+- Exposed the tolerance parameter `EPS` that is used to determine intersections between
+  triangles in `primal:intersect()` as an optional final parameter.
+- Added BVH spatial index option to the `mesh_tester` utility for calculating
+  triangle-triangle intersection.
+- Added `axom::execution_space< ExecSpace >::onDevice()` to check if execution
+  space is on device.
+- Added Axom macro `AXOM_SUPPRESS_HD_WARN` to silence host device compiler
+  warnings.
+- Added option to quest's `SignedDistance` class and C API to toggle whether
+  the distance query computes the sign.
+- Added a `batched` option to quest's signed distance query example application.
+  This computes all distance queries on an array of points using a single call to `computeDistance`.
+  The query uses OpenMP threading, when available.
+- Added new component, Inlet, to assist in retrieving and storing data from
+  an input deck.
+- Added the ability to specify an [Umpire] allocator ID to use with the
+  BVH. This allows the application to use a device allocator for the BVH and 
+  avoid use of Unified Memory (UM) on the GPU, which can hinder perfomrmance, 
+  or use a pool allocator to mitigate the latencies associated with allocation/deallocation.
+  The allocator ID is specified as an optional argument to the BVH constructor.
+- Added new CMake option, `AXOM_ENABLE_ANNOTATIONS`, to enable/disable code 
+  annotations in Axom. Default is OFF.
+- Added Axom annotation macros. The macros can be used to annotate functions,
+  using the `AXOM_PERF_MARK_FUNCTION` macro, or at a more fine grain level,
+  different sections of code can be annotated by wrapping them within an
+  `AXOM_PERF_MARK_SECTION` block. As a first cut, this works with NVTX tools.
+  However, the hooks are in place to add support for Caliper in the future. 
+- Added a simple interface to NVTX that allows an application to set the color
+  and category for NVTX ranges corresponding to annotated code in Axom. The
+  application can now call `axom::nvtx:set_color()` and 
+  `axom::nvtx::set_category()` to set the corresponding parameters respectively. 
+  This facilitates in the performance evaluation by allowing developers to easily 
+  filter out calls by category or visually by setting a different color to use
+  in GUI tools, such as, NVVP and NSight.
+- Added a portable floating_point_limits traits class, to return min(), max(), lowest() 
+  and epsilon() of a `float` or `double` type. The functionality is equivalent to that provided by 
+  std::numeric_limits, but, the code is host/device decorated accordingly such that it
+  can also be called on the device.
+- Added initial support for ray queries using the BVH. The caller may now supply a set of rays to 
+  a BVH and the BVH will return a set of candidate BVH bins that intersect each ray.
+- Added initial support for bounding box queries using the BVH. The caller may
+  now supply a set of bounding boxes to a BVH and the BVH will return a set of 
+  candidate BVH bins that intersect each bounding box.
+- Added an `axom-config.cmake` file to axom's installation to streamline incorporating axom
+  into user applications. See `<axom-install>/examples/axom` for example usages.
+- Added [Sol] as a built-in TPL for fast and simple `C++` and `Lua` binding.
+  Sol is automatically enabled when `LUA_DIR` is found. 
+  The version of Sol used in this release is `v2.20.6`, which requires `C++14`.
+
+### Removed
+- Removed the `AXOM_ENABLE_CUB` option, since `CUB` is no lonher used directly in
+  Axom code. Instead, we use `RAJA::stable_sort` with RAJA-v0.12.1 and fallback
+  to `std::stable_sort` with older versions of RAJA and when the code is built
+  without RAJA.
+
+### Deprecated
+
+### Changed
+- Updated Axom to support RAJA-v0.12.1 and Umpire-v4.01, but the code remains
+  backwards compatible with previous versions of RAJA and Umpire.
+- Transitioned Axom's code formatting tool from `Uncrustify` to [clang-format].
+  Axom's clang-format rules depend on clang 10.
+- Modified the command line interface for `mesh_tester` utility. Interface
+  now uses a *-m, --method* option to select the spatial index, and *-p, policy*
+  option now accepts a string or integer value.
+- Renamed the `AXOM_USE_MPI3`option to `AXOM_ENABLE_MPI3` for consistency.
+- Modified the API for the BVH to accomodate different query types. The queries are now
+  more explicitly called `BVH::findPoints()` and `BVH::findRays()`.
+- Modified the API of Axom's memory management routines to not leak usage of Umpire. Instead of 
+  passing an `umpire::Allocator` object to specify an allocator, we now use the corresponding
+  integer ID associated with the allocator.
+- All names in the C API now preserve the case of the C++ function.
+  ex. `SIDRE_datastore_new` is now `SIDRE_DataStore_new`.
+- Fortran API in slic module. `axom::slic::message` Level enums are changed
+  from  *enum-name_enumerator* to *namespace_enumerator*.
+  ex. `level_error` is now `message_error`.
+- Fortran derived-type constructors are now generic functions named afer the derived type.
+  `datastore_new` is now `SidreDataStore`
+  `iomanager_new` is now `IOManager`
+
+### Fixed
+- Fixed a bug in `primal::intersect(Segment, BoundingBox)` and added regression tests.
+- Spin's octrees can now be used with 64-bit indexes. This allows octrees 
+  with up to 64 levels of resolution when using a 64-bit index type.
+- Resolved issue with `AXOM_USE_64BIT_INDEXTYPE` configurations. Axom can once again
+  be configured with 64-bit index types.
+- Fixed a triangle-triangle intersection case in primal that produced inconsistent results
+  depending on the order of the triangle's vertices.
+- Fixed issue in the parallel construction of the BVH on GPUs, due to incoherent
+  L1 cache that could result in some data corruption in the BVH. The code now
+  calls ``__threadfence_system()`` after the parent is computed and stored back
+  to global memory to ensure that the *write*  is visible to all threads. 
+- Fixed issue in Mint that would cause the clang@9.0.0 compiler to segfault. The
+  `mint_cell_types.cpp` test was causing a segfault in the compiler. The main
+  issue triggering this compiler bug was the use of `constexpr` when defining the
+  static `cell_info` array of structs. The code has been modified to use `const`
+  instead.
+- Fixed issue in Quest's Signed Distance query that would prevent consecutive
+  calls to Quest when MPI-3 shared memory is enabled due to not properly 
+  nullifying internal pointers when finalize is called.
+- Fixed issue where the BVH would dispatch to the CPU sort() routine when the
+  specified execution policy was CUDA_EXEC async. Now, when the execution policy
+  is CUDA_EXEC the code would correctly dispatch to the GPU sort, using CUB
+  (when CUB is enabled), regardless of whether it's synchronous or asynchronous.
+- Fixed issue with missing the bvh_traverse.hpp from the install prefix, which was preventing
+  applications from using the BVH when pointing to an Axom install prefix.
+- Fixed usage of cuda kernel policies in Mint. Raja v0.11.0 changed the way max threads
+  launch bounds is calculated. Consequently, a large number of threads was being launched
+  leading to max registry count violation when linking. We are now using fixed kernel size
+  of 256 threads (16x16 in 2D and 8x8x4 in 3D).
+- Third-party libraries can now build on the Windows platform through uberenv using vcpkg
+  ("zero-to-axom support on Windows")
+
+### Known Bugs
+- Encountered a compiler bug on IBM LC platforms when using the IBM XL C/C++
+  compiler. The issue is manifested in the `generate_aabbs_and_centroids` method
+  in the `spin_bvh.cpp` unit test. It seems that the compiler does not handle
+  the lambda capture of the arrays correctly which leads to a segfault. A
+  workaround for the IBM XL compiler is provided.
+- There is a known bug in MVAPICH that prevents consecutive creation/deletion
+  of MPI windows. This was encountered on LC platforms when enabling shared
+  memory in the Signed Distance Query. See the corresponding 
+  [Github Issue](https://github.com/LLNL/axom/issues/257) for details.
+
 ## [Version 0.3.3] - Release date 2020-01-31
 
 ### Added
@@ -18,7 +145,7 @@ The Axom project release numbers follow [Semantic Versioning](http://semver.org/
 - Added [CLI11](https://github.com/CLIUtils/CLI11) command line parser as a built-in third party library.
 
 ### Removed
-
+  
 ### Deprecated
 
 ### Changed
@@ -282,7 +409,8 @@ The Axom project release numbers follow [Semantic Versioning](http://semver.org/
 ### Known Bugs
 -
 
-[Unreleased]:    https://github.com/LLNL/axom/compare/v0.3.3...develop
+[Unreleased]:    https://github.com/LLNL/axom/compare/v0.4.0...develop
+[Version 0.4.0]: https://github.com/LLNL/axom/compare/v0.3.3...v0.4.0
 [Version 0.3.3]: https://github.com/LLNL/axom/compare/v0.3.2...v0.3.3
 [Version 0.3.2]: https://github.com/LLNL/axom/compare/v0.3.1...v0.3.2
 [Version 0.3.1]: https://github.com/LLNL/axom/compare/v0.3.0...v0.3.1
@@ -291,4 +419,6 @@ The Axom project release numbers follow [Semantic Versioning](http://semver.org/
 
 [Scalable Checkpoint Restart (SCR)]: https://computation.llnl.gov/projects/scalable-checkpoint-restart-for-mpi
 [SU2 Mesh file format]: https://su2code.github.io/docs/Mesh-File/
-
+[Umpire]: https://github.com/LLNL/Umpire
+[clang-format]: https://releases.llvm.org/10.0.0/tools/clang/docs/ClangFormatStyleOptions.html
+[Sol]: https://github.com/ThePhD/sol2

azure-pipelines.yml

@@ -8,56 +8,80 @@
 variables:
   DO_BUILD: 'yes'
   DO_TEST: 'yes'
+  DO_CLEAN: 'no'
 
 strategy:
   matrix:
-    linux_gcc6:
-      VM_ImageName: 'ubuntu-16.04'
-      Compiler_ImageName: 'axom/compilers:gcc-6'
-      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
-      COMPILER: 'g++'
-      TEST_TARGET: 'linux_gcc6'
+# Temporarily disabled due to running out of space
+#    linux_gcc6:
+#      VM_ImageName: 'ubuntu-16.04'
+#      Compiler_ImageName: 'axom/tpls:gcc-6'
+#      CMAKE_EXTRA_FLAGS: ''
+#      COMPILER: 'g++'
+#      TEST_TARGET: 'linux_gcc6'
+#      HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-gcc@6.1.0'
     linux_gcc7:
       VM_ImageName: 'ubuntu-16.04'
-      Compiler_ImageName: 'axom/compilers:gcc-7'
-      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
+      Compiler_ImageName: 'axom/tpls:gcc-7'
+      CMAKE_EXTRA_FLAGS: ''
       COMPILER: 'g++'
       TEST_TARGET: 'linux_gcc7'
+      HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-gcc@7.3.0'
     linux_gcc8:
       VM_ImageName: 'ubuntu-16.04'
-      Compiler_ImageName: 'axom/compilers:gcc-8'
-      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
+      Compiler_ImageName: 'axom/tpls:gcc-8'
+      CMAKE_EXTRA_FLAGS: ''
       COMPILER: 'g++'
       TEST_TARGET: 'linux_gcc8'
-    linux_clang4:
+      HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-gcc@8.1.0'
+    linux_gcc8_64bit:
       VM_ImageName: 'ubuntu-16.04'
-      Compiler_ImageName: 'axom/compilers:clang-4'
-      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
-      COMPILER: 'clang++'
-      TEST_TARGET: 'linux_clang4'
+      Compiler_ImageName: 'axom/tpls:gcc-8'
+      CMAKE_EXTRA_FLAGS: '-DAXOM_USE_64BIT_INDEXTYPE:BOOL=ON'
+      COMPILER: 'g++'
+      TEST_TARGET: 'linux_gcc8_64bit'
+      HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-gcc@8.1.0'
+    linux_clang4:
+       VM_ImageName: 'ubuntu-16.04'
+       Compiler_ImageName: 'axom/tpls:clang-4'
+       CMAKE_EXTRA_FLAGS: ''
+       COMPILER: 'clang++'
+       TEST_TARGET: 'linux_clang4'
+       HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-clang@4.0.0'
     linux_clang6:
-      VM_ImageName: 'ubuntu-16.04'
-      Compiler_ImageName: 'axom/compilers:clang-6'
-      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
-      COMPILER: 'clang++'
-      TEST_TARGET: 'linux_clang6'
+       VM_ImageName: 'ubuntu-16.04'
+       Compiler_ImageName: 'axom/tpls:clang-6'
+       CMAKE_EXTRA_FLAGS: ''
+       COMPILER: 'clang++'
+       TEST_TARGET: 'linux_clang6'
+       HOST_CONFIG: 'docker-linux-ubuntu16.04-x86_64-clang@6.0.0'
     osx_gcc:
-      VM_ImageName: 'macos-10.13'
+      VM_ImageName: 'macos-10.14'
       CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
       TEST_TARGET: 'osx_gcc'
+    windows:
+      VM_ImageName: 'windows-2019'
+      CMAKE_EXTRA_FLAGS: '-DAXOM_ENABLE_SIDRE=Off'
+      TEST_TARGET: 'win_vs'
 
 pool:
   vmImage: $(VM_ImageName)
 
 steps:
 - checkout: self
-  clean: boolean
+  clean: true
   submodules: recursive
 - task: CMake@1
   inputs:
     workingDir: 'build'
     cmakeArgs: '$(CMAKE_EXTRA_FLAGS) ../src'
-  condition: eq( variables['Agent.OS'], 'Darwin')
+  condition: or( eq( variables['Agent.OS'], 'Windows_NT'), eq( variables['Agent.OS'], 'Darwin'))
+- task: VSBuild@1
+  inputs:
+    solution: 'build/*.sln'
+    vsVersion: 'latest'
+  condition: eq( variables['Agent.OS'], 'Windows_NT')
+  displayName: 'Visual Studio Build'
 - script: |
     cd build
     make
@@ -69,12 +93,15 @@ steps:
   displayName: '$(TEST_TARGET) Test'
   condition: eq( variables['Agent.OS'], 'Darwin')
 - script:  |
-    docker run --rm --user='root' -v `pwd`:/home/axom $(Compiler_ImageName) chown -R axom /home/axom
-    docker run --rm  -v `pwd`:/home/axom -e TEST_TARGET -e COMPILER -e DO_BUILD -e DO_TEST -e CMAKE_EXTRA_FLAGS $(Compiler_ImageName) ./scripts/azure-pipelines/linux-build_and_test.sh
+    echo " -e $TEST_TARGET -e $COMPILER -e $DO_BUILD -e $DO_TEST -e $CMAKE_EXTRA_FLAGS $(Compiler_ImageName) ./scripts/azure-pipelines/linux-build_and_test.sh"
+    docker run --rm --user='root' -v `pwd`:/home/axom/axom $(Compiler_ImageName) chown -R axom /home/axom
+    docker run --rm -v `pwd`:/home/axom/axom -e TEST_TARGET -e COMPILER -e DO_BUILD -e DO_TEST -e DO_CLEAN -e HOST_CONFIG -e CMAKE_EXTRA_FLAGS $(Compiler_ImageName) ./axom/scripts/azure-pipelines/linux-build_and_test.sh
+
   condition: eq( variables['Agent.OS'], 'Linux')
   displayName: '$(TEST_TARGET) Build & Test'
 - task: PublishTestResults@2
   inputs:
     testResultsFormat: 'cTest'
     testResultsFiles: '**/Test.xml'
     testRunTitle: '$(TEST_TARGET) Tests'
+    failTaskOnFailedTests: true