Merge branch 'lammps:develop' into alphataubio-kokkos-fixes

.github/workflows/check-vla.yml

@@ -27,9 +27,9 @@ jobs:
 
     - name: Install extra packages
       run: |
+        sudo apt-get update
         sudo apt-get install -y ccache \
                                 libeigen3-dev \
-                                libgsl-dev \
                                 libcurl4-openssl-dev \
                                 mold \
                                 mpi-default-bin \

.github/workflows/compile-msvc.yml

@@ -1,4 +1,4 @@
-# GitHub action to build LAMMPS on Windows with Visual C++
+# GitHub action to test LAMMPS on Windows with Visual C++
 name: "Windows Unit Tests"
 
 on:
@@ -11,6 +11,10 @@ on:
 
   workflow_dispatch:
 
+concurrency:
+  group: ${{ github.event_name }}-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{github.event_name == 'pull_request'}}
+
 jobs:
   build:
     name: Windows Compilation Test

.github/workflows/full-regression.yml

@@ -0,0 +1,109 @@
+# GitHub action to build LAMMPS on Linux and run regression tests
+name: "Full Regression Test"
+
+on:
+  push:
+    branches:
+      - develop
+
+  workflow_dispatch:
+
+jobs:
+  build:
+    name: Build LAMMPS
+    # restrict to official LAMMPS repository
+    if: ${{ github.repository == 'lammps/lammps' }}
+    runs-on: ubuntu-latest
+    env:
+      CCACHE_DIR: ${{ github.workspace }}/.ccache
+    strategy:
+      max-parallel: 8
+      matrix:
+        idx: [ 0, 1, 2, 3, 4, 5, 6, 7 ]
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 2
+        show-progress: false
+
+    - name: Install extra packages
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y ccache ninja-build libeigen3-dev \
+                     libcurl4-openssl-dev python3-dev \
+                     mpi-default-bin mpi-default-dev
+
+    - name: Create Build Environment
+      run: mkdir build
+
+    - name: Set up ccache
+      uses: actions/cache@v4
+      with:
+        path: ${{ env.CCACHE_DIR }}
+        key: linux-full-ccache-${{ github.sha }}
+        restore-keys: linux-full-ccache-
+
+    - name: Building LAMMPS via CMake
+      shell: bash
+      run: |
+        ccache -z
+        python3 -m venv linuxenv
+        source linuxenv/bin/activate
+        python3 -m pip install --upgrade pip
+        python3 -m pip install numpy pyyaml junit_xml
+        cmake -S cmake -B build \
+              -C cmake/presets/gcc.cmake \
+              -C cmake/presets/most.cmake \
+              -D CMAKE_CXX_COMPILER_LAUNCHER=ccache \
+              -D CMAKE_C_COMPILER_LAUNCHER=ccache \
+              -D BUILD_SHARED_LIBS=off \
+              -D DOWNLOAD_POTENTIALS=off \
+              -D PKG_MANIFOLD=on \
+              -D PKG_ML-PACE=on \
+              -D PKG_ML-RANN=on \
+              -D PKG_RHEO=on \
+              -D PKG_PTM=on \
+              -D PKG_PYTHON=on \
+              -D PKG_QTB=on \
+              -D PKG_SMTBQ=on \
+              -G Ninja
+        cmake --build build
+        ccache -s
+
+    - name: Run Full Regression Tests
+      shell: bash
+      run: |
+        source linuxenv/bin/activate
+        python3 tools/regression-tests/run_tests.py \
+               --lmp-bin=build/lmp \
+               --config-file=tools/regression-tests/config_serial.yaml \
+               --examples-top-level=examples --analyze --num-workers=8
+
+        python3 tools/regression-tests/run_tests.py \
+               --lmp-bin=build/lmp \
+               --config-file=tools/regression-tests/config_serial.yaml \
+               --list-input=input-list-${{ matrix.idx }}.txt \
+               --output-file=output-${{ matrix.idx }}.xml \
+               --progress-file=progress-${{ matrix.idx }}.yaml \
+               --log-file=run-${{ matrix.idx }}.log
+
+        tar -cvf full-regression-test-${{ matrix.idx }}.tar run-${{ matrix.idx }}.log progress-${{ matrix.idx }}.yaml output-${{ matrix.idx }}.xml
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: full-regression-test-artifact-${{ matrix.idx }}
+        path: full-regression-test-${{ matrix.idx }}.tar
+
+  merge:
+    runs-on: ubuntu-latest
+    needs: build
+    steps:
+      - name: Merge Artifacts
+        uses: actions/upload-artifact/merge@v4
+        with:
+          name: merged-full-regresssion-artifact
+          pattern: full-regression-test-artifact-*
+

.github/workflows/quick-regression.yml

@@ -0,0 +1,118 @@
+# GitHub action to build LAMMPS on Linux and run selected regression tests
+name: "Quick Regression Test"
+
+on:
+  pull_request:
+    branches:
+      - develop
+
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.event_name }}-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{github.event_name == 'pull_request'}}
+
+jobs:
+  build:
+    name: Build LAMMPS
+    # restrict to official LAMMPS repository
+    if: ${{ github.repository == 'lammps/lammps' }}
+    runs-on: ubuntu-latest
+    env:
+      CCACHE_DIR: ${{ github.workspace }}/.ccache
+    strategy:
+      max-parallel: 4
+      matrix:
+        idx: [ 0, 1, 2, 3 ]
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+        show-progress: false
+
+    - name: Install extra packages
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y ccache ninja-build libeigen3-dev \
+                     libcurl4-openssl-dev python3-dev \
+                     mpi-default-bin mpi-default-dev
+
+    - name: Create Build Environment
+      run: mkdir build
+
+    - name: Set up ccache
+      uses: actions/cache@v4
+      with:
+        path: ${{ env.CCACHE_DIR }}
+        key: linux-quick-ccache-${{ github.sha }}
+        restore-keys: linux-quick-ccache-
+
+    - name: Building LAMMPS via CMake
+      shell: bash
+      run: |
+        ccache -z
+        python3 -m venv linuxenv
+        source linuxenv/bin/activate
+        python3 -m pip install --upgrade pip
+        python3 -m pip install numpy pyyaml junit_xml
+        cmake -S cmake -B build \
+              -C cmake/presets/gcc.cmake \
+              -C cmake/presets/most.cmake \
+              -D CMAKE_CXX_COMPILER_LAUNCHER=ccache \
+              -D CMAKE_C_COMPILER_LAUNCHER=ccache \
+              -D BUILD_SHARED_LIBS=off \
+              -D DOWNLOAD_POTENTIALS=off \
+              -D PKG_MANIFOLD=on \
+              -D PKG_ML-PACE=on \
+              -D PKG_ML-RANN=on \
+              -D PKG_RHEO=on \
+              -D PKG_PTM=on \
+              -D PKG_PYTHON=on \
+              -D PKG_QTB=on \
+              -D PKG_SMTBQ=on \
+              -G Ninja
+        cmake --build build
+        ccache -s
+
+    - name: Run Regression Tests for Modified Styles
+      shell: bash
+      run: |
+        source linuxenv/bin/activate
+        python3 tools/regression-tests/run_tests.py \
+               --lmp-bin=build/lmp \
+               --config-file=tools/regression-tests/config_quick.yaml \
+               --examples-top-level=examples \
+               --quick-reference=tools/regression-tests/reference.yaml \
+               --quick --quick-branch=origin/develop --quick-max=100 --num-workers=4
+
+        if [ -f input-list-${{ matrix.idx }}.txt ]
+        then \
+           python3 tools/regression-tests/run_tests.py \
+               --lmp-bin=build/lmp \
+               --config-file=tools/regression-tests/config_quick.yaml \
+               --list-input=input-list-${{ matrix.idx }}.txt \
+               --output-file=output-${{ matrix.idx }}.xml \
+               --progress-file=progress-${{ matrix.idx }}.yaml \
+               --log-file=run-${{ matrix.idx }}.log
+        fi
+
+        tar -cvf quick-regression-test-${{ matrix.idx }}.tar run-${{ matrix.idx }}.log progress-${{ matrix.idx }}.yaml output-${{ matrix.idx }}.xml
+
+    - name: Upload artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: quick-regression-test-artifact-${{ matrix.idx }}
+        path: quick-regression-test-${{ matrix.idx }}.tar
+
+  merge:
+    runs-on: ubuntu-latest
+    needs: build
+    steps:
+      - name: Merge Artifacts
+        uses: actions/upload-artifact/merge@v4
+        with:
+          name: merged-quick-regresssion-artifact
+          pattern: quick-regression-test-artifact-*
+

.github/workflows/style-check.yml

@@ -0,0 +1,37 @@
+# GitHub action to run checks from tools/coding_standard
+name: "Check for Programming Style Conformance"
+
+on:
+  push:
+    branches:
+      - develop
+  pull_request:
+    branches:
+      - develop
+
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.event_name }}-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{github.event_name == 'pull_request'}}
+
+jobs:
+  build:
+    name: Programming Style Conformance
+    if: ${{ github.repository == 'lammps/lammps' }}
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 1
+
+    - name: Run Tests
+      working-directory: src
+      shell: bash
+      run: |
+         make check-whitespace
+         make check-permissions
+         make check-homepage
+         make check-errordocs

.github/workflows/unittest-linux.yml

@@ -11,6 +11,10 @@ on:
 
   workflow_dispatch:
 
+concurrency:
+  group: ${{ github.event_name }}-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{github.event_name == 'pull_request'}}
+
 jobs:
   build:
     name: Linux Unit Test
@@ -27,9 +31,9 @@ jobs:
 
     - name: Install extra packages
       run: |
+        sudo apt-get update
         sudo apt-get install -y ccache \
                                 libeigen3-dev \
-                                libgsl-dev \
                                 libcurl4-openssl-dev \
                                 mold \
                                 ninja-build \

.github/workflows/unittest-macos.yml

@@ -11,6 +11,10 @@ on:
 
   workflow_dispatch:
 
+concurrency:
+  group: ${{ github.event_name }}-${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{github.event_name == 'pull_request'}}
+
 jobs:
   build:
     name: MacOS Unit Test

cmake/CMakeLists.txt

@@ -497,7 +497,7 @@ if((CMAKE_CXX_COMPILER_ID STREQUAL "Intel") AND (CMAKE_CXX_STANDARD GREATER_EQUA
           PROPERTIES COMPILE_OPTIONS "-std=c++14")
 endif()
 
-if(PKG_ATC OR PKG_AWPMD OR PKG_ML-QUIP OR PKG_ML-POD OR PKG_ELECTRODE OR BUILD_TOOLS)
+if(PKG_ATC OR PKG_AWPMD OR PKG_ML-QUIP OR PKG_ML-POD OR PKG_ELECTRODE OR PKG_RHEO OR BUILD_TOOLS)
   enable_language(C)
   if (NOT USE_INTERNAL_LINALG)
     find_package(LAPACK)
@@ -572,7 +572,7 @@ else()
 endif()
 
 foreach(PKG_WITH_INCL KSPACE PYTHON ML-IAP VORONOI COLVARS ML-HDNNP MDI MOLFILE NETCDF
-        PLUMED QMMM ML-QUIP SCAFACOS MACHDYN VTK KIM COMPRESS ML-PACE LEPTON RHEO EXTRA-COMMAND)
+        PLUMED QMMM ML-QUIP SCAFACOS MACHDYN VTK KIM COMPRESS ML-PACE LEPTON EXTRA-COMMAND)
   if(PKG_${PKG_WITH_INCL})
     include(Packages/${PKG_WITH_INCL})
   endif()

cmake/Modules/Documentation.cmake

@@ -4,6 +4,8 @@
 option(BUILD_DOC "Build LAMMPS HTML documentation" OFF)
 
 if(BUILD_DOC)
+  option(BUILD_DOC_VENV "Build LAMMPS documentation virtual environment" ON)
+  mark_as_advanced(BUILD_DOC_VENV)
   # Current Sphinx versions require at least Python 3.8
   # use default (or custom) Python executable, if version is sufficient
   if(Python_VERSION VERSION_GREATER_EQUAL 3.8)
@@ -18,14 +20,6 @@ if(BUILD_DOC)
   find_package(Doxygen 1.8.10 REQUIRED)
   file(GLOB DOC_SOURCES CONFIGURE_DEPENDS ${LAMMPS_DOC_DIR}/src/[^.]*.rst)
 
-  add_custom_command(
-    OUTPUT docenv
-    COMMAND ${VIRTUALENV} docenv
-  )
-
-  set(DOCENV_BINARY_DIR ${CMAKE_BINARY_DIR}/docenv/bin)
-  set(DOCENV_REQUIREMENTS_FILE ${LAMMPS_DOC_DIR}/utils/requirements.txt)
-
   set(SPHINX_CONFIG_DIR ${LAMMPS_DOC_DIR}/utils/sphinx-config)
   set(SPHINX_CONFIG_FILE_TEMPLATE ${SPHINX_CONFIG_DIR}/conf.py.in)
   set(SPHINX_STATIC_DIR  ${SPHINX_CONFIG_DIR}/_static)
@@ -44,6 +38,15 @@ if(BUILD_DOC)
   # configure paths in conf.py, since relative paths change when file is copied
   configure_file(${SPHINX_CONFIG_FILE_TEMPLATE} ${DOC_BUILD_CONFIG_FILE})
 
+  if(BUILD_DOC_VENV)
+    add_custom_command(
+      OUTPUT docenv
+      COMMAND ${VIRTUALENV} docenv
+    )
+
+    set(DOCENV_BINARY_DIR ${CMAKE_BINARY_DIR}/docenv/bin)
+    set(DOCENV_REQUIREMENTS_FILE ${LAMMPS_DOC_DIR}/utils/requirements.txt)
+
     add_custom_command(
       OUTPUT ${DOC_BUILD_DIR}/requirements.txt
       DEPENDS docenv ${DOCENV_REQUIREMENTS_FILE}
@@ -53,6 +56,15 @@ if(BUILD_DOC)
       COMMAND ${DOCENV_BINARY_DIR}/pip $ENV{PIP_OPTIONS} install -r ${DOC_BUILD_DIR}/requirements.txt --upgrade
     )
 
+    set(DOCENV_DEPS docenv ${DOC_BUILD_DIR}/requirements.txt)
+    if(NOT TARGET Sphinx::sphinx-build)
+      add_executable(Sphinx::sphinx-build IMPORTED GLOBAL)
+      set_target_properties(Sphinx::sphinx-build PROPERTIES IMPORTED_LOCATION "${DOCENV_BINARY_DIR}/sphinx-build")
+    endif()
+  else()
+    find_package(Sphinx)
+  endif()
+
   set(MATHJAX_URL "https://github.com/mathjax/MathJax/archive/3.1.3.tar.gz" CACHE STRING "URL for MathJax tarball")
   set(MATHJAX_MD5 "b81661c6e6ba06278e6ae37b30b0c492" CACHE STRING "MD5 checksum of MathJax tarball")
   mark_as_advanced(MATHJAX_URL)
@@ -97,8 +109,8 @@ if(BUILD_DOC)
   endif()
   add_custom_command(
     OUTPUT html
-    DEPENDS ${DOC_SOURCES} docenv ${DOC_BUILD_DIR}/requirements.txt ${DOXYGEN_XML_DIR}/index.xml ${BUILD_DOC_CONFIG_FILE}
+    DEPENDS ${DOC_SOURCES} ${DOCENV_DEPS} ${DOXYGEN_XML_DIR}/index.xml ${BUILD_DOC_CONFIG_FILE}
-    COMMAND ${DOCENV_BINARY_DIR}/sphinx-build ${SPHINX_EXTRA_OPTS} -b html -c ${DOC_BUILD_DIR} -d ${DOC_BUILD_DIR}/doctrees ${LAMMPS_DOC_DIR}/src ${DOC_BUILD_DIR}/html
+    COMMAND Sphinx::sphinx-build ${SPHINX_EXTRA_OPTS} -b html -c ${DOC_BUILD_DIR} -d ${DOC_BUILD_DIR}/doctrees ${LAMMPS_DOC_DIR}/src ${DOC_BUILD_DIR}/html
     COMMAND ${CMAKE_COMMAND} -E create_symlink Manual.html ${DOC_BUILD_DIR}/html/index.html
     COMMAND ${CMAKE_COMMAND} -E copy_directory ${LAMMPS_DOC_DIR}/src/PDF ${DOC_BUILD_DIR}/html/PDF
     COMMAND ${CMAKE_COMMAND} -E remove -f ${DOXYGEN_XML_DIR}/run.stamp

cmake/Modules/FindSphinx.cmake

@@ -0,0 +1,29 @@
+# Find sphinx-build
+find_program(Sphinx_EXECUTABLE NAMES sphinx-build
+	                       PATH_SUFFIXES bin
+			       DOC "Sphinx documenation build executable")
+mark_as_advanced(Sphinx_EXECUTABLE)
+
+if(Sphinx_EXECUTABLE)
+  execute_process(COMMAND ${Sphinx_EXECUTABLE} --version
+                  OUTPUT_VARIABLE sphinx_version
+                  OUTPUT_STRIP_TRAILING_WHITESPACE
+                  RESULT_VARIABLE _sphinx_version_result)
+
+  if(_sphinx_version_result)
+    message(WARNING "Unable to determine sphinx-build verison: ${_sphinx_version_result}")
+  else()
+    string(REGEX REPLACE "sphinx-build ([0-9.]+).*"
+                         "\\1"
+                         Sphinx_VERSION
+                         "${sphinx_version}")
+  endif()
+
+  if(NOT TARGET Sphinx::sphinx-build)
+    add_executable(Sphinx::sphinx-build IMPORTED GLOBAL)
+    set_target_properties(Sphinx::sphinx-build PROPERTIES IMPORTED_LOCATION "${Sphinx_EXECUTABLE}")
+  endif()
+endif()
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(Sphinx REQUIRED_VARS Sphinx_EXECUTABLE VERSION_VAR Sphinx_VERSION)

cmake/Modules/Packages/KOKKOS.cmake

@@ -8,8 +8,24 @@ endif()
 ########################################################################
 # consistency checks and Kokkos options/settings required by LAMMPS
 if(Kokkos_ENABLE_CUDA)
-  message(STATUS "KOKKOS: Enabling CUDA LAMBDA function support")
+  option(Kokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC "CUDA asynchronous malloc support" OFF)
-  set(Kokkos_ENABLE_CUDA_LAMBDA ON CACHE BOOL "" FORCE)
+  mark_as_advanced(Kokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC)
+  if(Kokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC)
+    message(STATUS "KOKKOS: CUDA malloc async support enabled")
+  else()
+    message(STATUS "KOKKOS: CUDA malloc async support disabled")
+  endif()
+endif()
+if(Kokkos_ENABLE_HIP)
+  option(Kokkos_ENABLE_HIP_MULTIPLE_KERNEL_INSTANTIATIONS "Enable multiple kernel instantiations with HIP" ON)
+  mark_as_advanced(Kokkos_ENABLE_HIP_MULTIPLE_KERNEL_INSTANTIATIONS)
+  option(Kokkos_ENABLE_ROCTHRUST "Use RoCThrust library" ON)
+  mark_as_advanced(Kokkos_ENABLE_ROCTHRUST)
+
+  if(Kokkos_ARCH_AMD_GFX942 OR Kokkos_ARCH_AMD_GFX940)
+    option(Kokkos_ENABLE_IMPL_HIP_UNIFIED_MEMORY "Enable unified memory with HIP" ON)
+    mark_as_advanced(Kokkos_ENABLE_IMPL_HIP_UNIFIED_MEMORY)
+  endif()
 endif()
 # Adding OpenMP compiler flags without the checks done for
 # BUILD_OMP can result in compile failures. Enforce consistency.
@@ -18,6 +34,15 @@ if(Kokkos_ENABLE_OPENMP)
     message(FATAL_ERROR "Must enable BUILD_OMP with Kokkos_ENABLE_OPENMP")
   endif()
 endif()
+
+if(Kokkos_ENABLE_SERIAL)
+  if(NOT (Kokkos_ENABLE_OPENMP OR Kokkos_ENABLE_THREADS OR
+    Kokkos_ENABLE_CUDA OR Kokkos_ENABLE_HIP OR Kokkos_ENABLE_SYCL
+    OR Kokkos_ENABLE_OPENMPTARGET))
+  option(Kokkos_ENABLE_ATOMICS_BYPASS "Disable atomics for Kokkos Serial Backend" ON)
+  mark_as_advanced(Kokkos_ENABLE_ATOMICS_BYPASS)
+  endif()
+endif()
 ########################################################################
 
 option(EXTERNAL_KOKKOS "Build against external kokkos library" OFF)
@@ -45,8 +70,8 @@ if(DOWNLOAD_KOKKOS)
   list(APPEND KOKKOS_LIB_BUILD_ARGS "-DCMAKE_CXX_EXTENSIONS=${CMAKE_CXX_EXTENSIONS}")
   list(APPEND KOKKOS_LIB_BUILD_ARGS "-DCMAKE_TOOLCHAIN_FILE=${CMAKE_TOOLCHAIN_FILE}")
   include(ExternalProject)
-  set(KOKKOS_URL "https://github.com/kokkos/kokkos/archive/4.3.01.tar.gz" CACHE STRING "URL for KOKKOS tarball")
+  set(KOKKOS_URL "https://github.com/kokkos/kokkos/archive/4.4.01.tar.gz" CACHE STRING "URL for KOKKOS tarball")
-  set(KOKKOS_MD5 "243de871b3dc2cf3990c1c404032df83" CACHE STRING "MD5 checksum of KOKKOS tarball")
+  set(KOKKOS_MD5 "de6ee80d00b6212b02bfb7f1e71a8392" CACHE STRING "MD5 checksum of KOKKOS tarball")
   mark_as_advanced(KOKKOS_URL)
   mark_as_advanced(KOKKOS_MD5)
   GetFallbackURL(KOKKOS_URL KOKKOS_FALLBACK)
@@ -71,7 +96,7 @@ if(DOWNLOAD_KOKKOS)
   add_dependencies(LAMMPS::KOKKOSCORE kokkos_build)
   add_dependencies(LAMMPS::KOKKOSCONTAINERS kokkos_build)
 elseif(EXTERNAL_KOKKOS)
-  find_package(Kokkos 4.3.01 REQUIRED CONFIG)
+  find_package(Kokkos 4.4.01 REQUIRED CONFIG)
   target_link_libraries(lammps PRIVATE Kokkos::kokkos)
 else()
   set(LAMMPS_LIB_KOKKOS_SRC_DIR ${LAMMPS_LIB_SOURCE_DIR}/kokkos)

cmake/Modules/Packages/RHEO.cmake

@@ -1,2 +0,0 @@
-find_package(GSL 2.6 REQUIRED)
-target_link_libraries(lammps PRIVATE GSL::gsl)

cmake/presets/mingw-cross.cmake

@@ -67,6 +67,7 @@ set(WIN_PACKAGES
   REACTION
   REAXFF
   REPLICA
+  RHEO
   RIGID
   SHOCK
   SMTBQ

cmake/presets/most.cmake

@@ -60,6 +60,7 @@ set(ALL_PACKAGES
   REACTION
   REAXFF
   REPLICA
+  RHEO
   RIGID
   SHOCK
   SPH

cmake/presets/windows.cmake

@@ -60,6 +60,7 @@ set(WIN_PACKAGES
   REACTION
   REAXFF
   REPLICA
+  RHEO
   RIGID
   SHOCK
   SMTBQ

doc/src/Build_development.rst

@@ -138,12 +138,27 @@ during development:
 The status of this automated testing can be viewed on `https://ci.lammps.org
 <https://ci.lammps.org>`_.
 
-The scripts and inputs for integration, run, and regression testing
+The scripts and inputs for integration, run, and legacy regression
-are maintained in a
+testing are maintained in a `separate repository
-`separate repository <https://github.com/lammps/lammps-testing>`_
+<https://github.com/lammps/lammps-testing>`_ of the LAMMPS project on
-of the LAMMPS project on GitHub.  A few tests are also run as GitHub
+GitHub.  A few tests are also run as GitHub Actions and their
-Actions and their configuration files are in the ``.github/workflows/``
+configuration files are in the ``.github/workflows/`` folder of the
-folder of the LAMMPS git tree.
+LAMMPS git tree.
+
+Regression tests can also be performed locally with the :ref:`regression
+tester tool <regression>`.  The tool checks if a given LAMMPS binary run
+with selected input examples produces thermo output that is consistent
+with the provided log files.  The script can be run in one pass over all
+available input files, but it can also first create multiple lists of
+inputs or folders that can then be run with multiple workers
+concurrently to speed things up.  Another mode allows to do a quick
+check of inputs that contain commands that have changes in the current
+checkout branch relative to a git branch.  This works similar to the two
+pass mode, but will select only shorter runs and no more than 100 inputs
+that are chosen randomly.  This ensures that this test runs
+significantly faster compared to the full test run.  These test runs can
+also be performed with instrumented LAMMPS binaries (see previous
+section).
 
 The unit testing facility is integrated into the CMake build process of
 the LAMMPS source code distribution itself.  It can be enabled by

doc/src/Build_extras.rst

@@ -2251,28 +2251,38 @@ verified to work in February 2020 with Quantum Espresso versions 6.3 to
 RHEO package
 ------------
 
-To build with this package you must have the `GNU Scientific Library
+This package depends on the BPM package.
-(GSL) <https://www.gnu.org/software/gsl/>` installed in locations that
-are accessible in your environment.  The GSL library should be at least
-version 2.7.
 
 .. tabs::
 
    .. tab:: CMake build
 
-      If CMake cannot find the GSL library or include files, you can set:
-
       .. code-block:: bash
 
-         -D GSL_ROOT_DIR=path    # path to root of GSL installation
+         -D PKG_RHEO=yes               # enable the package itself
+         -D PKG_BPM=yes                # the RHEO package requires BPM
+         -D USE_INTERNAL_LINALG=value  # prefer internal LAPACK if true
+
+      Some features in the RHEO package are dependent on code in the BPM
+      package so the latter one *must* be enabled as well.
+
+      The RHEO package also requires LAPACK (and BLAS) and CMake
+      can identify their locations and pass that info to the RHEO
+      build script.  But on some systems this may cause problems when
+      linking or the dependency is not desired.  By using the setting
+      ``-D USE_INTERNAL_LINALG=yes`` when running the CMake
+      configuration, you will select compiling and linking the bundled
+      linear algebra library and work around the limitations.
 
    .. tab:: Traditional make
 
-      LAMMPS will try to auto-detect the GSL compiler and linker flags
+      The RHEO package requires LAPACK (and BLAS) which can be either
-      from the corresponding ``pkg-config`` file (``gsl.pc``), otherwise
+      a system provided library or the bundled "linalg" library. This
-      you can edit the file ``lib/rheo/Makefile.lammps``
+      is a subset of LAPACK translated to C++.  For that, one of the
-      to specify the paths and library names where indicated by comments.
+      provided ``Makefile.lammps.<config>`` files needs to be copied
-      This must be done **before** the package is installed.
+      to ``Makefile.lammps`` and edited as needed.  The default file
+      uses the bundled "linalg" library, which can be built by
+      ``make lib-linalg args='-m serial'`` in the ``src`` folder.
 
 ----------
 

doc/src/Build_settings.rst

@@ -229,8 +229,7 @@ can be used with the Intel or GNU compiler (see the ``FFT_LIB`` setting
 above).
 
 The NVIDIA Performance Libraries (NVPL) FFT library is optimized for NVIDIA
-Grace Armv9.0 architecture. You can download it from
+Grace Armv9.0 architecture. You can download it from https://docs.nvidia.com/nvpl/
-`https://docs.nvidia.com/nvpl/`_.
 
 The cuFFT and hipFFT FFT libraries are packaged with NVIDIA's CUDA and
 AMD's HIP installations, respectively. These FFT libraries require the

doc/src/Intro_authors.rst

@@ -56,7 +56,7 @@ lammps.org".  General questions about LAMMPS should be posted in the
      - SNL
      - jmgoff at sandia.gov
      - machine learned potentials, QEq solvers, Python
-   * - Megan McCarthy
+   * - Meg McCarthy
      - SNL
      - megmcca at sandia.gov
      - alloys, micro-structure, machine learned potentials
@@ -67,7 +67,7 @@ lammps.org".  General questions about LAMMPS should be posted in the
    * - `Trung Nguyen <tn_>`_
      - U Chicago
      - ndactrung at gmail.com
-     - soft matter, GPU package
+     - soft matter, GPU package, DIELECTRIC package, regression testing
 
 .. _rb:  https://rbberger.github.io/
 .. _gc:  https://enthalpiste.fr/

doc/src/Run_windows.rst

@@ -3,71 +3,70 @@ Running LAMMPS on Windows
 
 To run a serial (non-MPI) executable, follow these steps:
 
-* Get a command prompt by going to Start->Run... ,
+* Install a LAMMPS installer package from https://packages.lammps.org/windows.html
-  then typing "cmd".
+* Open the "Command Prompt" or "Terminal" app.
-* Move to the directory where you have your input script,
+* Change to the directory where you have your input script,
   (e.g. by typing: cd "Documents").
-* At the command prompt, type "lmp -in in.file", where
+* At the command prompt, type "lmp -in in.file.lmp", where
-  in.file is the name of your LAMMPS input script.
+  ``in.file.lmp`` is the name of your LAMMPS input script.
 
 Note that the serial executable includes support for multi-threading
-parallelization from the styles in the OPENMP packages.  To run with
+parallelization from the styles in the OPENMP and KOKKOS packages.
-4 threads, you can type this:
+To run with 4 threads, you can type this:
 
 .. code-block:: bash
 
-   lmp -in in.lj -pk omp 4 -sf omp
+   lmp -in in.lj.lmp -pk omp 4 -sf omp
+   lmp -in in.lj.lmp -k on t 4 -sf kk
+
+Alternately, you can also install a package with LAMMPS-GUI included and
+open the LAMMPS-GUI app (the package includes the command line version
+of LAMMPS as well) and open the input file in the GUI and run it from
+there.  For details on LAMMPS-GUI, see :doc:`Howto_lammps_gui`.
 
 ----------
 
-For the MPI executable, which allows you to run LAMMPS under Windows
+For the MS-MPI executables, which allow you to run LAMMPS under Windows
-in parallel, follow these steps.
+in parallel using MPI rather than multi-threading, follow these steps.
 
-Download and install a compatible MPI library binary package:
+Download and install the MS-MPI runtime package ``msmpisetup.exe`` from
-
+https://www.microsoft.com/en-us/download/details.aspx?id=105289 (Note
-* for 32-bit Windows: `mpich2-1.4.1p1-win-ia32.msi <https://download.lammps.org/thirdparty/mpich2-1.4.1p1-win-ia32.msi>`_
+that the ``msmpisdk.msi`` is **only** required for **compilation** of
-* for 64-bit Windows: `mpich2-1.4.1p1-win-x86-64.msi <https://download.lammps.org/thirdparty/mpich2-1.4.1p1-win-x86-64.msi>`_
+LAMMPS from source on Windows using Microsoft Visual Studio). After
-
+installation of MS-MPI perform a reboot.
-The LAMMPS Windows installer packages will automatically adjust your
-path for the default location of this MPI package. After the
-installation of the MPICH2 software, it needs to be integrated into
-the system.  For this you need to start a Command Prompt in
-*Administrator Mode* (right click on the icon and select it). Change
-into the MPICH2 installation directory, then into the subdirectory
-**bin** and execute **smpd.exe -install**\ . Exit the command window.
-
-* Get a new, regular command prompt by going to Start->Run... ,
-  then typing "cmd".
-* Move to the directory where you have your input file
-  (e.g. by typing: cd "Documents").
 
 Then you can run the executable in serial like in the example above
 or in parallel using MPI with one of the following commands:
 
 .. code-block:: bash
 
-   mpiexec -localonly 4 lmp -in in.file
+   mpiexec -localonly 4 lmp -in in.file.lmp
-   mpiexec -np 4 lmp -in in.file
+   mpiexec -np 4 lmp -in in.file.lmp
 
-where in.file is the name of your LAMMPS input script. For the latter
+where ``in.file.lmp`` is the name of your LAMMPS input script. For the
-case, you may be prompted to enter the password that you set during
+latter case, you may be prompted to enter the password that you set
-installation of the MPI library software.
+during installation of the MPI library software.
 
 In this mode, output may not immediately show up on the screen, so if
 your input script takes a long time to execute, you may need to be
 patient before the output shows up.
 
-The parallel executable can also run on a single processor by typing
+Note that the parallel executable also includes OpenMP multi-threading
-something like this:
+through both the OPENMP and the KOKKOS package, which can be combined
+with MPI using something like:
 
 .. code-block:: bash
 
-   lmp -in in.lj
+   mpiexec -localonly 2 lmp -in in.lj.lmp -pk omp 2 -sf omp
+   mpiexec -localonly 2 lmp -in in.lj.lmp -kokkos on t 2 -sf kk
 
-Note that the parallel executable also includes OpenMP
+-------------
-multi-threading, which can be combined with MPI using something like:
-
-.. code-block:: bash
-
-   mpiexec -localonly 2 lmp -in in.lj -pk omp 2 -sf omp
 
+MPI parallelization will work for *all* functionality in LAMMPS and in
+many cases the MPI parallelization is more efficient than
+multi-threading since LAMMPS was designed from ground up for MPI
+parallelization using domain decomposition.  Multi-threading is only
+available for selected styles and implemented on top of the MPI
+parallelization.  Multi-threading is most useful for systems with large
+load imbalances when using domain decomposition and a smaller number
+of threads (<= 8).

doc/src/Tools.rst

@@ -1022,7 +1022,7 @@ regression tests with a given LAMMPS binary.  The tool launches the
 LAMMPS binary with any given input script under one of the `examples`
 subdirectories, and compares the thermo output in the generated log file
 with those in the provided log file with the same number of processors
-ub the same subdirectory. If the differences between the actual and
+in the same subdirectory. If the differences between the actual and
 reference values are within specified tolerances, the test is considered
 passed.  For each test batch, that is, a set of example input scripts,
 the mpirun command, the LAMMPS command line arguments, and the

doc/src/fix_adapt.rst

@@ -319,25 +319,34 @@ all types from 1 to :math:`N`.  A leading asterisk means all types from
 :math:`N` (inclusive).  A middle asterisk means all types from m to n
 (inclusive).
 
-Currently *bond* does not support bond_style hybrid nor bond_style
+If :doc:`bond_style hybrid <bond_hybrid>` is used, *bstyle* should be a
-hybrid/overlay as bond styles. The bond styles that currently work
+sub-style name. The bond styles that currently work with fix adapt are:
-with fix_adapt are
 
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`class2 <bond_class2>`                         | r0                        | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`fene <bond_fene>`                             | k,r0                      | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
+| :doc:`fene/expand <bond_fene_expand>`               | k,r0,epsilon,sigma,shift  | type bonds |
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`fene/nm <bond_fene>`                          | k,r0                      | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`gromos <bond_gromos>`                         | k,r0                      | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`harmonic <bond_harmonic>`                     | k,r0                      | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
+| :doc:`harmonic/restrain <bond_harmonic_restrain>`   | k                         | type bonds |
++-----------------------------------------------------+---------------------------+------------+
+| :doc:`harmonic/shift <bond_harmonic_shift>`         | k,r0,r1                   | type bonds |
++-----------------------------------------------------+---------------------------+------------+
+| :doc:`harmonic/shift/cut <bond_harmonic_shift_cut>` | k,r0,r1                   | type bonds |
++-----------------------------------------------------+---------------------------+------------+
+| :doc:`mm3 <bond_mm3>`                               | k,r0                      | type bonds |
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`morse <bond_morse>`                           | r0                        | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 | :doc:`nonlinear <bond_nonlinear>`                   | epsilon,r0                | type bonds |
-+------------------------------------+------------+------------+
++-----------------------------------------------------+---------------------------+------------+
 
 ----------
 
@@ -357,15 +366,34 @@ all types from 1 to :math:`N`.  A leading asterisk means all types from
 :math:`N` (inclusive).  A middle asterisk means all types from m to n
 (inclusive).
 
-Currently *angle* does not support angle_style hybrid nor angle_style
+If :doc:`angle_style hybrid <angle_hybrid>` is used, *astyle* should be a
-hybrid/overlay as angle styles. The angle styles that currently work
+sub-style name. The angle styles that currently work with fix adapt are:
-with fix_adapt are
 
-+------------------------------------+----------+-------------+
++--------------------------------------------------------------------+-----------------+-------------+
 | :doc:`harmonic <angle_harmonic>`                                   | k,theta0        | type angles |
-+------------------------------------+----------+-------------+
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`charmm <angle_charmm>`                                       | k,theta0        | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`class2 <angle_class2>`                                       | k2,k3,k4,theta0 | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
 | :doc:`cosine <angle_cosine>`                                       | k               | type angles |
-+------------------------------------+----------+-------------+
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`cosine/periodic <angle_cosine_periodic>`                     | k,b,n           | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`cosine/squared/restricted <angle_cosine_squared_restricted>` | k,theta0        | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`dipole <angle_dipole>`                                       | k,gamma0        | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`fourier <angle_fourier>`                                     | k,c0,c1,c2      | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`fourier/simple <angle_fourier_simple>`                       | k,c,n           | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`mm3 <angle_mm3>`                                             | k,theta0        | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`quartic <angle_quartic>`                                     | k2,k3,k4,theta0 | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
+| :doc:`spica <angle_spica>`                                         | k,theta0        | type angles |
++--------------------------------------------------------------------+-----------------+-------------+
 
 Note that internally, theta0 is stored in radians, so the variable
 this fix uses to reset theta0 needs to generate values in radians.

doc/src/fix_gcmc.rst

@@ -50,8 +50,8 @@ Syntax
        *intra_energy* value = intramolecular energy (energy units)
        *tfac_insert* value = scale up/down temperature of inserted atoms (unitless)
        *overlap_cutoff* value = maximum pair distance for overlap rejection (distance units)
-       *max* value = Maximum number of molecules allowed in the system
+       *max* value = Maximum number of atoms allowed in the fix group (and region)
-       *min* value = Minimum number of molecules allowed in the system
+       *min* value = Minimum number of atoms allowed in the fix group (and region)
 
 Examples
 """"""""
@@ -380,10 +380,11 @@ an infinite positive energy to all new configurations that place any
 pair of atoms closer than the specified overlap cutoff distance.
 
 The *max* and *min* keywords allow for the restriction of the number of
-atoms in the simulation. They automatically reject all insertion or
+atoms in the fix group (and region in case the *region* keyword is
-deletion moves that would take the system beyond the set boundaries.
+used).  They automatically reject all insertion or deletion moves that
-Should the system already be beyond the boundary, only moves that bring
+would take the system beyond the set boundaries.  Should the system
-the system closer to the bounds may be accepted.
+already be beyond the boundary, only moves that bring the system closer
+to the bounds may be accepted.
 
 The *group* keyword adds all inserted atoms to the :doc:`group <group>`
 of the group-ID value. The *grouptype* keyword adds all inserted atoms

doc/src/group2ndx.rst

@@ -51,7 +51,7 @@ index file.  When specifying group IDs, only those groups will be
 written to the index file.  In order to follow the Gromacs conventions,
 the group *all* will be renamed to *System* in the index file.
 
-The *ndx2group* command will create of update group definitions from
+The *ndx2group* command will create or update group definitions from
 those stored in an index file.  Without specifying any group IDs, all
 groups except *System* will be read from the index file and the
 corresponding groups recreated.  If a group of the same name already

doc/src/read_dump.rst

@@ -115,10 +115,11 @@ to tell LAMMPS how many parallel files exist, via its specified
 
 The format of the dump file is selected through the *format* keyword.
 If specified, it must be the last keyword used, since all remaining
-arguments are passed on to the dump reader.  The *native* format is
+arguments are passed on to the dump reader.  The *native* format is for
-for native LAMMPS dump files, written with a :doc:`dump atom <dump>`
+native LAMMPS dump files, written with a :doc:`dump atom <dump>` or
-or :doc:`dump custom <dump>` command.  The *xyz* format is for generic XYZ
+:doc:`dump custom <dump>` command.  The *xyz* format is for generic XYZ
-formatted dump files.  These formats take no additional values.
+formatted dump files (see details below).  These formats take no
+additional values.
 
 The *molfile* format supports reading data through using the `VMD <vmd_>`_
 molfile plugin interface. This dump reader format is only available,
@@ -230,23 +231,39 @@ will then have a label corresponding to the fix-ID rather than "x" or
 "xs".  The *label* keyword can also be used to specify new column
 labels for fields *id* and *type*\ .
 
-For dump files in *xyz* format, only the *x*, *y*, and *z* fields are
+For dump files in *xyz* format, only the *type*, *x*, *y*, and *z*
-supported.  The dump file does not store atom IDs, so these are
+fields are supported.  There are many variants of the XYZ file format.
-assigned consecutively to the atoms as they appear in the dump file,
+LAMMPS will read the number of atoms from the first line of each frame,
-starting from 1.  Thus you should ensure that order of atoms is
+ignore the second (title) line, and then read one line for each atom in the format:
-consistent from snapshot to snapshot in the XYZ dump file.  See
+
-the :doc:`dump_modify sort <dump_modify>` command if the XYZ dump file
+.. parsed-literal::
-was written by LAMMPS.
+
+   <label> <x coordinate>  <y coordinate> <z coordinate>
+
+
+If the atom label is a numeric integer (like with XYZ files created by
+created with default settings by :doc:`dump style <dump>` *xyz*), that
+number will be used as the atom type.  If the atom label is a string,
+then a type map must be created using the :doc:`labelmap command
+<labelmap>`.  This map needs to associate each (numeric) atom type with
+a string label. The numeric atom type is stored internally.
+
+The xyz format dump file does not store atom IDs, so these are assigned
+consecutively to the atoms as they appear in the dump file, starting
+from 1.  Thus you should ensure that the order of atoms is consistent
+from snapshot to snapshot in the XYZ dump file.  See the
+:doc:`dump_modify sort <dump_modify>` command if the XYZ dump file was
+written by LAMMPS.
 
 For dump files in *molfile* format, the *x*, *y*, *z*, *vx*, *vy*, and
 *vz* fields can be specified.  However, not all molfile formats store
 velocities, or their respective plugins may not support reading of
-velocities.  The molfile dump files do not store atom IDs, so these
+velocities.  The molfile dump files do not store atom IDs, so these are
-are assigned consecutively to the atoms as they appear in the dump
+assigned consecutively to the atoms as they appear in the dump file,
-file, starting from 1.  Thus you should ensure that order of atoms are
+starting from 1.  Thus you should ensure that the order of atoms are
-consistent from snapshot to snapshot in the molfile dump file.
+consistent from snapshot to snapshot in the molfile dump file.  See the
-See the :doc:`dump_modify sort <dump_modify>` command if the dump file
+:doc:`dump_modify sort <dump_modify>` command if the dump file was
-was written by LAMMPS.
+written by LAMMPS.
 
 The *adios* format supports all fields that the *native* format supports
 except for the *q* charge field.

doc/src/region.rst

@@ -18,7 +18,7 @@ Syntax
        *delete* = no args
        *block* args = xlo xhi ylo yhi zlo zhi
          xlo,xhi,ylo,yhi,zlo,zhi = bounds of block in all dimensions (distance units)
-         xlo,xhi,ylo,yhi,zlo,zhi can be a variable
+           xlo,xhi,ylo,yhi,zlo,zhi can be a variable (see below)
        *cone* args = dim c1 c2 radlo radhi lo hi
          dim = *x* or *y* or *z* = axis of cone
          c1,c2 = coords of cone axis in other 2 dimensions (distance units)
@@ -38,6 +38,7 @@ Syntax
        *plane* args = px py pz nx ny nz
          px,py,pz = point on the plane (distance units)
          nx,ny,nz = direction normal to plane (distance units)
+           px,py,pz can be a variable (see below)
        *prism* args = xlo xhi ylo yhi zlo zhi xy xz yz
          xlo,xhi,ylo,yhi,zlo,zhi = bounds of untilted prism (distance units)
          xy = distance to tilt y in x direction (distance units)
@@ -206,9 +207,10 @@ parameters a,b,c for style *ellipsoid*, can each be specified as an
 equal-style :doc:`variable <variable>`.  Likewise, for style *sphere*
 and *ellipsoid* the x-, y-, and z- coordinates of the center of the
 sphere/ellipsoid can be specified as an equal-style variable.  And for
-style *cylinder* the two center positions c1 and c2 for the location
+style *cylinder* the two center positions c1 and c2 for the location of
-of the cylinder axes can be specified as a equal-style variable. For style *cone*
+the cylinder axes can be specified as a equal-style variable.  For style
-all properties can be defined via equal-style variables.
+*cone* all properties can be defined via equal-style variables.  For
+style *plane* the point can be defined via equal-style variables.
 
 If the value is a variable, it should be specified as v_name, where
 name is the variable name.  In this case, the variable will be

doc/utils/sphinx-config/false_positives.txt

@@ -141,6 +141,7 @@ arg
 arge
 args
 argv
+Armv
 arrhenius
 Arun
 arXiv

examples/ASPHERE/line/in.line

@@ -13,11 +13,11 @@ neigh_modify    delay 0 every 1 check yes
 neigh_modify    exclude molecule/intra all
 
 pair_style      line/lj 2.5
-pair_coeff      * * 1.0 1.0 1.0 0.25 2.5
+pair_coeff      * * 0.25 0.25 1.0 0.25 2.5
 
 fix             2 all rigid molecule langevin 2.0 2.0 1.0 492983
 
-fix             3 all deform 1 x scale 0.3 y scale 0.3
+fix             3 all deform 5 x scale 0.8 y scale 0.8
 fix             4 all enforce2d
 
 compute         10 all property/atom end1x end1y end2x end2y
@@ -26,7 +26,7 @@ compute         10 all property/atom end1x end1y end2x end2y
 #dump           2 all custom 500 dump1.line id type &
 #               c_10[1] c_10[2] c_10[3] c_10[4]
 
-timestep        0.004
+timestep        0.001
 
 compute         1 all erotate/asphere
 compute         2 all ke

examples/ASPHERE/line/in.line.srd

@@ -24,7 +24,7 @@ set             group small mass 0.01
 pair_style      lj/cut  1.5
 pair_coeff      1 1 1.0 1.0
 pair_coeff      2 2 0.0 1.0 0.0
-pair_coeff      1 2 0.0 1.0
+pair_coeff      1 2 1.0 1.0
 
 delete_atoms    overlap 1.5 small big
 
@@ -34,7 +34,7 @@ reset_timestep  0
 
 velocity        small create 1.44 87287 loop geom
 
-neighbor        0.3 multi
+neighbor        0.8 multi
 neigh_modify    delay 0 every 1 check yes
 neigh_modify    exclude molecule/intra big include big
 
@@ -46,19 +46,19 @@ neigh_modify    include big
 pair_style      line/lj 2.5
 pair_coeff      1 1 1.0  1.0 1.0 1.0 2.5
 pair_coeff      2 2 0.0  0.0 0.0 1.0 0.0
-pair_coeff      1 2 1.0 0.0 0.0 1.0 0.0
+pair_coeff      1 2 1.0  0.2 1.0 0.2 2.5
 
 # use fix SRD to push small particles out from inside big ones
 # if comment out, big particles won't see SRD particles
 
-timestep        0.001
+timestep        0.0001
 
 fix             1 big rigid molecule
 fix             2 small srd 20 big 1.0 0.25 49894 &
                 search 0.2 cubic warn 0.0001 shift yes 49829 &
                 overlap yes collision noslip
 
-fix             3 all deform 1 x scale 0.35 y scale 0.35
+fix             3 all deform 1 x scale 1.25 y scale 1.25
 fix             4 all enforce2d
 
 # diagnostics
@@ -96,12 +96,12 @@ change_box      all triclinic
 
 fix             2 small srd 20 big 1.0 0.25 49894 &
                 search 0.2 cubic warn 0.0001 shift yes 49829 &
-                overlap yes collision noslip tstat yes
+                overlap yes collision noslip #tstat yes
 
 #dump           1 all custom 500 dump2.atom.srd id type x y z ix iy iz
 #dump           2 all custom 500 dump2.line.srd id type &
 #               c_10[1] c_10[2] c_10[3] c_10[4]
 
-fix             3 all deform 1 xy erate 0.05 units box remap v
+fix             3 all deform 1 xy erate 0.0002 units box remap v
 
-run             40000
+run             30000

examples/ASPHERE/line/log.1Feb14.line.g++.8

@@ -1,213 +0,0 @@
-LAMMPS (1 Feb 2014)
-# Aspherical shear demo - 2d line box and triangle mixture, implicit solvent
-
-units		lj
-atom_style	line
-dimension	2
-
-read_data	data.line
-  orthogonal box = (-22.5539 -22.5539 -0.5) to (22.5539 22.5539 0.5)
-  4 by 2 by 1 MPI processor grid
-  reading atoms ...
-  350 atoms
-  350 lines
-
-velocity	all create 1.44 320984 loop geom
-
-neighbor	0.3 bin
-neigh_modify	delay 0 every 1 check yes
-neigh_modify	exclude molecule all
-
-pair_style	line/lj 2.5
-pair_coeff	* * 1.0 0.25
-
-fix		2 all rigid molecule langevin 2.0 2.0 1.0 492983
-100 rigid bodies with 350 atoms
-
-fix		3 all deform 1 x scale 0.3 y scale 0.3
-fix		4 all enforce2d
-
-compute		10 all property/atom end1x end1y end2x end2y
-
-#dump		1 all custom 500 dump1.atom id type x y z ix iy iz
-#dump		2 all custom 500 dump1.line id type #		c_10[1] c_10[2] c_10[3] c_10[4]
-
-timestep	0.004
-
-compute		1 all erotate/asphere
-compute		2 all ke
-compute		3 all pe
-variable	toteng equal (c_1+c_2+c_3)/atoms
-
-thermo		1000
-thermo_style	custom step temp f_2 pe ke c_1 c_2 c_3 v_toteng
-
-run		10000
-Memory usage per processor = 2.6072 Mbytes
-Step Temp 2 PotEng KinEng 1 2 3 toteng 
-       0            0    1.2780105            0           -0  0.037823677   0.50989511            0   0.54771879 
-    1000            0    1.9896906  -0.13333756           -0   0.12630626   0.72641827  -0.13333756   0.71938697 
-    2000            0    2.0408541  -0.24906647           -0   0.13199238   0.74265938  -0.24906647   0.62558529 
-    3000            0    1.9921323  -0.39759798           -0   0.11671173   0.73705927  -0.39759798   0.45617302 
-    4000            0    2.1392159  -0.36475197           -0   0.11923802    0.7975688  -0.36475197   0.55205485 
-    5000            0     2.139715  -0.52582091           -0   0.15846417   0.75855653  -0.52582091   0.39119979 
-    6000            0    2.1313904  -0.65532027           -0   0.11090422   0.80254883  -0.65532027   0.25813278 
-    7000            0    1.9416614  -0.81322598           -0   0.11722471   0.71491587  -0.81322598  0.018914608 
-    8000            0    1.9388183   -1.0581149           -0   0.10142762   0.72949452   -1.0581149  -0.22719275 
-    9000            0    2.2830265    -1.583347           -0   0.14583927   0.83260066    -1.583347  -0.60490709 
-   10000            0    4.1416666   -2.6034045           -0   0.18839177    1.5866082   -2.6034045  -0.82840455 
-Loop time of 2.84713 on 8 procs for 10000 steps with 350 atoms
-
-Pair  time (%) = 1.26607 (44.4682)
-Neigh time (%) = 0.0163046 (0.572669)
-Comm  time (%) = 0.786148 (27.612)
-Outpt time (%) = 0.000368953 (0.0129588)
-Other time (%) = 0.778241 (27.3342)
-
-Nlocal:    43.75 ave 51 max 39 min
-Histogram: 1 2 0 2 0 1 1 0 0 1
-Nghost:    170.25 ave 180 max 160 min
-Histogram: 2 1 0 0 0 0 2 0 1 2
-Neighs:    963.125 ave 1209 max 767 min
-Histogram: 2 0 2 0 1 0 1 0 1 1
-
-Total # of neighbors = 7705
-Ave neighs/atom = 22.0143
-Neighbor list builds = 987
-Dangerous builds = 0
-
-#undump          1
-#undump          2
-unfix           3
-
-change_box      all triclinic
-  triclinic box = (-6.76616 -6.76616 -0.5) to (6.76616 6.76616 0.5) with tilt (0 0 0)
-
-#dump		1 all custom 500 dump2.atom id type x y z ix iy iz
-#dump		2 all custom 500 dump2.line id type #		c_10[1] c_10[2] c_10[3] c_10[4]
-
-fix		3 all deform 1 xy erate 0.01 units box
-
-run		100000
-Memory usage per processor = 2.75978 Mbytes
-Step Temp 2 PotEng KinEng 1 2 3 toteng 
-   10000            0    4.1416666   -2.6024092           -0   0.18839177    1.5866082   -2.6024092  -0.82740923 
-   11000            0    1.8408319   -3.1031477           -0   0.12073234   0.66819561   -3.1031477   -2.3142198 
-   12000            0    2.0793172   -3.0329681           -0  0.086927592   0.80420833   -3.0329681   -2.1418322 
-   13000            0    2.2022136     -2.99776           -0   0.14309291    0.8007129     -2.99776   -2.0539542 
-   14000            0    1.9510757    -3.094649           -0   0.09482969   0.74134559    -3.094649   -2.2584737 
-   15000            0    1.9874689   -3.1431753           -0    0.1083061    0.7434663   -3.1431753   -2.2914029 
-   16000            0    1.8484778   -2.9491537           -0  0.079102883   0.71310191   -2.9491537   -2.1569489 
-   17000            0    2.1978438   -2.9675694           -0   0.11677634   0.82515673   -2.9675694   -2.0256363 
-   18000            0    2.0293397   -2.9860257           -0    0.1287845    0.7409325   -2.9860257   -2.1163087 
-   19000            0    2.0077219    -3.005622           -0   0.12697603    0.7334762    -3.005622   -2.1451698 
-   20000            0    2.1806369   -3.0622132           -0   0.11066657   0.82389212   -3.0622132   -2.1276545 
-   21000            0    1.8156509   -3.1031481           -0   0.10227614   0.67585994   -3.1031481    -2.325012 
-   22000            0    2.1028516   -3.0861182           -0  0.098877162   0.80234493   -3.0861182   -2.1848961 
-   23000            0    1.8994891   -3.0110243           -0   0.10961187   0.70445488   -3.0110243   -2.1969576 
-   24000            0    1.9305389   -3.0057136           -0   0.11735151    0.7100223   -3.0057136   -2.1783398 
-   25000            0    1.9553918   -3.0848948           -0   0.13217467   0.70585039   -3.0848948   -2.2468697 
-   26000            0    1.8903754   -2.9543658           -0  0.099925113   0.71023579   -2.9543658   -2.1442049 
-   27000            0    2.2624684   -3.2416154           -0   0.11398815   0.85564117   -3.2416154   -2.2719861 
-   28000            0    2.0335234   -3.1795174           -0   0.10291986   0.76859015   -3.1795174   -2.3080074 
-   29000            0    1.7056403   -3.1198739           -0  0.076174496   0.65481419   -3.1198739   -2.3888853 
-   30000            0    2.1203465   -3.0863113           -0   0.11355683   0.79516311   -3.0863113   -2.1775914 
-   31000            0    1.8446708   -3.2764357           -0   0.11006455   0.68050865   -3.2764357   -2.4858625 
-   32000            0    1.7947968    -3.081031           -0  0.087540776   0.68165784    -3.081031   -2.3118324 
-   33000            0    2.2576228   -3.1042222           -0   0.15666855   0.81088407   -3.1042222   -2.1366696 
-   34000            0    1.8522306   -3.0037311           -0   0.08276626   0.71104684   -3.0037311    -2.209918 
-   35000            0    2.0611686   -2.9877406           -0   0.10822003   0.77513794   -2.9877406   -2.1043826 
-   36000            0    2.0739798    -2.981184           -0   0.11784198   0.77100651    -2.981184   -2.0923355 
-   37000            0    2.2120215   -2.8961216           -0   0.13172943    0.8162798   -2.8961216   -1.9481124 
-   38000            0    2.1097357   -2.9746777           -0   0.11881736   0.78535507   -2.9746777   -2.0705053 
-   39000            0    1.8928127   -2.9482169           -0   0.10915752   0.70204792   -2.9482169   -2.1370115 
-   40000            0    1.8515483    -3.003524           -0  0.096981256   0.69653943    -3.003524   -2.2100033 
-   41000            0    1.8569065   -3.0834675           -0   0.10562739    0.6901897   -3.0834675   -2.2876504 
-   42000            0    2.2638459   -3.0577224           -0   0.10896442   0.86125524   -3.0577224   -2.0875027 
-   43000            0    2.0992151   -3.1219185           -0   0.13280434   0.76685929   -3.1219185   -2.2222548 
-   44000            0    2.3479986   -3.0702445           -0   0.14438131   0.86190379   -3.0702445   -2.0639594 
-   45000            0    2.0295235   -2.9619688           -0   0.13726365   0.73253213   -2.9619688    -2.092173 
-   46000            0    2.0597998   -3.1282569           -0   0.10139093    0.7813804   -3.1282569   -2.2454856 
-   47000            0    1.8173541   -3.1122647           -0   0.10380346   0.67506259   -3.1122647   -2.3333986 
-   48000            0    1.9650208     -3.22153           -0   0.10115952   0.74099226     -3.22153   -2.3793782 
-   49000            0    1.8836303   -3.0781944           -0   0.10561306   0.70165705   -3.0781944   -2.2709243 
-   50000            0    1.7799289   -3.1090208           -0   0.10522105   0.65760561   -3.1090208   -2.3461941 
-   51000            0    1.7270244    -2.894789           -0  0.079316066   0.66083727    -2.894789   -2.1546357 
-   52000            0    2.1036512   -3.0708266           -0   0.11762021    0.7839446   -3.0708266   -2.1692618 
-   53000            0    2.1749106    -3.043193           -0   0.11908953   0.81301499    -3.043193   -2.1110885 
-   54000            0    1.7245018   -3.1528646           -0   0.11118993   0.62788226   -3.1528646   -2.4137924 
-   55000            0    1.7599209   -3.0543405           -0  0.098996756   0.65525507   -3.0543405   -2.3000886 
-   56000            0    1.9085953   -3.1072383           -0   0.10931083   0.70865859   -3.1072383   -2.2892689 
-   57000            0    1.9682028    -3.131335           -0  0.094973379   0.74854212    -3.131335   -2.2878195 
-   58000            0    1.7813545    -3.167275           -0  0.085523136    0.6779145    -3.167275   -2.4038374 
-   59000            0    2.0324497   -3.1103534           -0   0.13343285   0.73761703   -3.1103534   -2.2393035 
-   60000            0     1.697349    -3.152831           -0  0.086035815   0.64139945    -3.152831   -2.4253957 
-   61000            0     2.061537   -3.0730484           -0   0.12753143   0.75598441   -3.0730484   -2.1895325 
-   62000            0    1.8186587   -3.1590894           -0  0.096701457   0.68272371   -3.1590894   -2.3796643 
-   63000            0    1.9692063   -3.1053354           -0  0.094864849    0.7490807   -3.1053354   -2.2613899 
-   64000            0    2.0113722   -3.0898117           -0   0.12640828   0.73560838   -3.0898117    -2.227795 
-   65000            0    1.9350085   -3.2290712           -0   0.11850297   0.71078638   -3.2290712   -2.3997819 
-   66000            0    2.0569306   -3.0489252           -0   0.12731012   0.75423156   -3.0489252   -2.1673835 
-   67000            0    2.0119071     -3.10894           -0  0.099142151   0.76310375     -3.10894   -2.2466941 
-   68000            0    2.0615321    -3.167552           -0   0.11066625   0.77284751    -3.167552   -2.2840383 
-   69000            0    1.8731077   -3.1864825           -0   0.12982742     0.672933   -3.1864825   -2.3837221 
-   70000            0     2.109088   -3.1787326           -0   0.13107619   0.77281866   -3.1787326   -2.2748377 
-   71000            0    1.9764202   -3.0731116           -0   0.12035585   0.72668137   -3.0731116   -2.2260744 
-   72000            0    1.6460591   -3.0865952           -0   0.09486718   0.61058673   -3.0865952   -2.3811413 
-   73000            0    1.9756864   -3.0431747           -0    0.1041775   0.74254526   -3.0431747    -2.196452 
-   74000            0    2.0553658   -3.1290715           -0   0.12267262   0.75819842   -3.1290715   -2.2482004 
-   75000            0     1.550997   -2.9813403           -0  0.094350599   0.57036238   -2.9813403   -2.3166273 
-   76000            0    1.9824851   -3.0057088           -0   0.11621875   0.73341774   -3.0057088   -2.1560723 
-   77000            0    1.8451505   -3.0063772           -0   0.12602143   0.66475735   -3.0063772   -2.2155984 
-   78000            0     1.934755   -3.0603773           -0   0.11800375   0.71117696   -3.0603773   -2.2311966 
-   79000            0    2.1982093    -3.095224           -0   0.12840585   0.81368383    -3.095224   -2.1531343 
-   80000            0    1.7547828   -3.0620364           -0  0.092552269   0.65949751   -3.0620364   -2.3099866 
-   81000            0    2.1064588   -3.1631358           -0   0.11415036   0.78861768   -3.1631358   -2.2603677 
-   82000            0    2.0039288   -3.1190532           -0   0.13374009   0.72508654   -3.1190532   -2.2602266 
-   83000            0    1.7749465   -3.2172309           -0   0.11017601   0.65051537   -3.2172309   -2.4565396 
-   84000            0    1.7865147   -3.1433093           -0  0.086859958   0.67878922   -3.1433093   -2.3776602 
-   85000            0    1.4541199   -3.0123872           -0  0.096301993   0.52689225   -3.0123872   -2.3891929 
-   86000            0    2.0345567   -2.9577061           -0   0.11756878    0.7543841   -2.9577061   -2.0857532 
-   87000            0    1.9655671   -3.0235755           -0   0.13068174   0.71170417   -3.0235755   -2.1811896 
-   88000            0    2.1238733   -3.0972095           -0   0.15660293   0.75362848   -3.0972095   -2.1869781 
-   89000            0      1.93413   -3.0018414           -0   0.11736651   0.71154634   -3.0018414   -2.1729285 
-   90000            0    2.0106701   -3.0778989           -0    0.1150105   0.74670527   -3.0778989   -2.2161831 
-   91000            0    1.9700224   -3.0072749           -0    0.1200151   0.72428023   -3.0072749   -2.1629796 
-   92000            0    2.3062898   -3.1245149           -0   0.16045862   0.82795128   -3.1245149    -2.136105 
-   93000            0    1.9977984   -3.1072212           -0  0.098381856   0.75781746   -3.1072212   -2.2510219 
-   94000            0    1.9684844   -3.0923519           -0   0.11228362   0.73135255   -3.0923519   -2.2487157 
-   95000            0    1.8913269   -3.1044766           -0  0.088944817   0.72162386   -3.1044766    -2.293908 
-   96000            0     2.139258   -2.9032081           -0   0.12122357    0.7956013   -2.9032081   -1.9863833 
-   97000            0    2.1549839   -3.0073823           -0   0.14177897   0.78178555   -3.0073823   -2.0838178 
-   98000            0     2.001084   -3.0411655           -0   0.16549603   0.69211139   -3.0411655   -2.1835581 
-   99000            0     2.269108   -3.0749812           -0   0.11370514   0.85876972   -3.0749812   -2.1025064 
-  100000            0    2.0270482   -3.0802101           -0  0.098599302   0.77013563   -3.0802101   -2.2114752 
-  101000            0    1.9412796   -3.1543623           -0   0.11953004   0.71244692   -3.1543623   -2.3223853 
-  102000            0    2.0146461   -3.0140006           -0   0.10152606    0.7618937   -3.0140006   -2.1505808 
-  103000            0    1.7377282   -3.1862597           -0  0.082111131   0.66262952   -3.1862597    -2.441519 
-  104000            0    1.7973897   -3.1055088           -0  0.093370304   0.67693958   -3.1055088   -2.3351989 
-  105000            0    2.2615194   -2.9636424           -0   0.12584102   0.84338157   -2.9636424   -1.9944198 
-  106000            0    1.8974182   -2.9505576           -0  0.085928679   0.72725057   -2.9505576   -2.1373783 
-  107000            0    2.0691161   -3.0804349           -0   0.12411017   0.76265389   -3.0804349   -2.1936708 
-  108000            0    2.0457472    -2.981702           -0   0.09308074   0.78366806    -2.981702   -2.1049532 
-  109000            0    1.6610604   -3.1463569           -0  0.066318676   0.64556436   -3.1463569   -2.4344738 
-  110000            0    1.9606721    -3.080013           -0   0.13164067   0.70864736    -3.080013   -2.2397249 
-Loop time of 62.2251 on 8 procs for 100000 steps with 350 atoms
-
-Pair  time (%) = 43.4946 (69.8987)
-Neigh time (%) = 0.395421 (0.635469)
-Comm  time (%) = 10.3551 (16.6414)
-Outpt time (%) = 0.00358662 (0.00576394)
-Other time (%) = 7.97644 (12.8187)
-
-Nlocal:    43.75 ave 51 max 33 min
-Histogram: 1 0 1 0 0 2 0 1 2 1
-Nghost:    168.375 ave 180 max 155 min
-Histogram: 1 1 0 0 1 0 2 2 0 1
-Neighs:    971 ave 1278 max 631 min
-Histogram: 1 1 0 1 0 1 2 1 0 1
-
-Total # of neighbors = 7768
-Ave neighs/atom = 22.1943
-Neighbor list builds = 7621
-Dangerous builds = 0

examples/ASPHERE/line/log.1Feb14.line.srd.g++.8

@@ -1,244 +0,0 @@
-LAMMPS (1 Feb 2014)
-# Aspherical shear demo - 2d line boxes, solvated by SRD particles
-
-units		lj
-atom_style	line
-atom_modify	first big
-dimension	2
-
-read_data	data.line.srd
-  orthogonal box = (-28.7968 -28.7968 -0.5) to (28.7968 28.7968 0.5)
-  4 by 2 by 1 MPI processor grid
-  reading atoms ...
-  400 atoms
-  400 lines
-
-# add small particles as hi density lattice
-
-lattice		sq 0.4
-Lattice spacing in x,y,z = 1.58114 1.58114 1.58114
-region		plane block INF INF INF INF -0.001 0.001
-lattice		sq 10.0
-Lattice spacing in x,y,z = 0.316228 0.316228 0.316228
-create_atoms	2 region plane
-Created 33489 atoms
-
-group		big type 1
-400 atoms in group big
-group		small type 2
-33489 atoms in group small
-set		group small mass 0.01
-  33489 settings made for mass
-
-# delete overlaps
-# must set 1-2 cutoff to non-zero value
-
-pair_style	lj/cut 1.5
-pair_coeff	1 1 1.0 1.0
-pair_coeff	2 2 0.0 1.0 0.0
-pair_coeff	1 2 0.0 1.0
-
-delete_atoms	overlap 1.5 small big
-Deleted 13605 atoms, new total = 20284
-
-# SRD run
-
-reset_timestep	0
-
-velocity	small create 1.44 87287 loop geom
-
-neighbor	0.3 bin
-neigh_modify	delay 0 every 1 check yes
-neigh_modify	exclude molecule big include big
-
-communicate	multi group big vel yes
-neigh_modify	include big
-
-# no pairwise interactions with small particles
-
-pair_style	line/lj 2.5
-pair_coeff	1 1 1.0 1.0
-pair_coeff	2 2 0.0 1.0 0.0
-pair_coeff	1 2 0.0 1.0 0.0
-
-# use fix SRD to push small particles out from inside big ones
-# if comment out, big particles won't see SRD particles
-
-timestep	0.001
-
-fix		1 big rigid molecule
-100 rigid bodies with 400 atoms
-fix	        2 small srd 20 big 1.0 0.25 49894 		search 0.2 cubic warn 0.0001 shift yes 49829 		overlap yes collision noslip
-
-fix		3 all deform 1 x scale 0.35 y scale 0.35
-fix		4 all enforce2d
-
-# diagnostics
-
-compute		tsmall small temp/deform
-compute		tbig big temp
-variable	pebig equal pe*atoms/count(big)
-variable	ebig equal etotal*atoms/count(big)
-
-compute		1 big erotate/asphere
-compute		2 all ke
-compute		3 all pe
-variable	toteng equal (c_1+c_2+c_3)/atoms
-
-thermo		1000
-thermo_style	custom step temp c_tsmall f_2[9] c_1 etotal 		v_pebig v_ebig press
-thermo_modify	temp tbig
-WARNING: Temperature for thermo pressure is not for group all (../thermo.cpp:439)
-
-compute		10 big property/atom end1x end1y end2x end2y
-
-#dump		1 all custom 500 dump1.atom.srd id type x y z ix iy iz
-#dump		2 all custom 500 dump1.line.srd id type #		c_10[1] c_10[2] c_10[3] c_10[4]
-
-run		10000
-WARNING: Using fix srd with box deformation but no SRD thermostat (../fix_srd.cpp:385)
-SRD info:
-  SRD/big particles = 19884 400
-  big particle diameter max/min = 1.99882 0.503306
-  SRD temperature & lamda = 1 0.2
-  SRD max distance & max velocity = 0.8 40
-  SRD grid counts: 230 230 1
-  SRD grid size: request, actual (xyz) = 0.25, 0.250407 0.250407 1
-  SRD per actual grid cell = 0.444963
-  SRD viscosity = 4.2356
-  big/SRD mass density ratio = 14.0918
-WARNING: SRD bin size for fix srd differs from user request (../fix_srd.cpp:2853)
-WARNING: Fix srd grid size > 1/4 of big particle diameter (../fix_srd.cpp:2875)
-  # of rescaled SRD velocities = 0
-  ave/max small velocity = 15.906 29.1054
-  ave/max big velocity = 0 0
-WARNING: Using compute temp/deform with inconsistent fix deform remap option (../compute_temp_deform.cpp:76)
-Memory usage per processor = 7.79007 Mbytes
-Step Temp tsmall 2[9] 1 TotEng pebig ebig Press 
-       0            0    1.4528554            0            0            0            0            0            0 
-    1000            0    1.1122612    1.1071958 0.00058011072            0            0            0   0.32625408 
-    2000            0    1.0254475    1.0231236 0.00072347646 -2.3144253e-06 -0.00011736451 -0.00011736451   0.44526308 
-    3000            0    1.0089214   0.99966408 0.00061511441 -0.00023298189 -0.011814512 -0.011814512   0.87208025 
-    4000            0    1.0066185   0.99391102 0.00079065587 -0.0021630644  -0.10968899  -0.10968899    1.3901494 
-    5000            0    1.0109003    1.0009124 0.00071588733 -0.0040023493  -0.20295913  -0.20295913    1.2736445 
-    6000            0    1.0119642    1.0105049 0.00080741361 -0.0067908165  -0.34436231  -0.34436231    2.5492857 
-    7000            0    1.0048989   0.98912274 0.00098851463 -0.010933999  -0.55446309  -0.55446309    3.5707613 
-    8000            0    1.0021597   0.99377939 0.0008405671 -0.015690841  -0.79568252  -0.79568252     6.685381 
-    9000            0    1.0197546      1.00114 0.0012645557 -0.021124602   -1.0712286   -1.0712286    18.907993 
-   10000            0    1.0401079    1.0184189 0.0036111452 0.0099395852   0.50403637   0.50403637    96.909257 
-Loop time of 3.93391 on 8 procs for 10000 steps with 20284 atoms
-
-Pair  time (%) = 0.110261 (2.80284)
-Neigh time (%) = 0.32853 (8.35124)
-Comm  time (%) = 0.199613 (5.07416)
-Outpt time (%) = 0.00108692 (0.0276295)
-Other time (%) = 3.29442 (83.7441)
-
-Nlocal:    2535.5 ave 2850 max 2226 min
-Histogram: 1 0 1 1 1 1 1 0 1 1
-Nghost:    109.25 ave 117 max 102 min
-Histogram: 1 0 2 0 1 1 1 1 0 1
-Neighs:    526 ave 665 max 407 min
-Histogram: 1 2 0 1 0 1 1 1 0 1
-
-Total # of neighbors = 4208
-Ave neighs/atom = 0.207454
-Neighbor list builds = 522
-Dangerous builds = 0
-
-#undump          1
-#undump          2
-unfix		3
-
-change_box      all triclinic
-  triclinic box = (-10.0789 -10.0789 -0.5) to (10.0789 10.0789 0.5) with tilt (0 0 0)
-
-fix	        2 small srd 20 big 1.0 0.25 49894 		search 0.2 cubic warn 0.0001 shift yes 49829 		overlap yes collision noslip tstat yes
-
-#dump		1 all custom 500 dump2.atom.srd id type x y z ix iy iz
-#dump		2 all custom 500 dump2.line.srd id type #		c_10[1] c_10[2] c_10[3] c_10[4]
-
-fix		3 all deform 1 xy erate 0.05 units box remap v
-
-run		40000
-SRD info:
-  SRD/big particles = 19884 400
-  big particle diameter max/min = 1.99882 0.503306
-  SRD temperature & lamda = 1 0.2
-  SRD max distance & max velocity = 0.8 40
-  SRD grid counts: 81 81 1
-  SRD grid size: request, actual (xyz) = 0.25, 0.248861 0.248861 1
-  SRD per actual grid cell = -11.3331
-  SRD viscosity = -1920.36
-  big/SRD mass density ratio = -0.546468
-WARNING: SRD bin size for fix srd differs from user request (../fix_srd.cpp:2853)
-WARNING: Fix srd grid size > 1/4 of big particle diameter (../fix_srd.cpp:2875)
-WARNING: Fix srd viscosity < 0.0 due to low SRD density (../fix_srd.cpp:2877)
-  # of rescaled SRD velocities = 1
-  ave/max small velocity = 12.724 40
-  ave/max big velocity = 1.54523 5.36901
-Memory usage per processor = 4.23847 Mbytes
-Step Temp tsmall 2[9] 1 TotEng pebig ebig Press 
-   10000            0    1.0277413            0 0.0036111452  0.010135973   0.51399517   0.51399517    60.794865 
-   11000            0    1.0024742            1 0.00040482851 -0.031643325    -1.604633    -1.604633    55.531632 
-   12000            0    1.0036177            1 0.00036416993 -0.030844063   -1.5641024   -1.5641024    56.694788 
-   13000            0    1.0043067            1 0.00068862721 -0.030673787   -1.5554677   -1.5554677    56.852537 
-   14000            0    1.0042533            1 0.00050997053 -0.035967153   -1.8238943   -1.8238943    51.254109 
-   15000            0     1.003956            1 0.00048458218 -0.038060147   -1.9300301   -1.9300301    50.346943 
-   16000            0    1.0059929            1 0.0004740426 -0.037433402   -1.8982478   -1.8982478     49.82532 
-   17000            0    1.0042401            1 0.00039855238 -0.034314446   -1.7400855   -1.7400855    48.430648 
-   18000            0    1.0025952            1 0.00046207703 -0.034166102    -1.732563    -1.732563    53.883613 
-   19000            0    1.0025881            1 0.00034818943 -0.037341367   -1.8935807   -1.8935807     48.90662 
-   20000            0    1.0009932            1 0.00031244041 -0.035274728   -1.7887814   -1.7887814    47.665935 
-   21000            0    1.0025603            1 0.0005514826 -0.034350221   -1.7418997   -1.7418997    63.290704 
-   22000            0    1.0038575            1 0.00034453716  -0.03576596   -1.8136918   -1.8136918    49.805726 
-   23000            0    1.0030321            1 0.00049535709 -0.035873083    -1.819124    -1.819124    48.973244 
-   24000            0    1.0048593            1 0.00034225992  -0.03322158   -1.6846663   -1.6846663    52.557383 
-   25000            0    1.0024423            1 0.00059157362 -0.030152319   -1.5290241   -1.5290241    56.733821 
-   26000            0    1.0022703            1 0.00047512976 -0.029563064    -1.499143    -1.499143    56.271943 
-   27000            0    1.0024537            1 0.00054158319  -0.02957039   -1.4995145   -1.4995145    55.246787 
-   28000            0    1.0023325            1 0.00051129428 -0.035115211   -1.7806924   -1.7806924    49.718172 
-   29000            0     1.002865            1 0.00056728135 -0.035484703   -1.7994293   -1.7994293    48.387401 
-   30000            0    1.0022693            1 0.00040284402 -0.033844303   -1.7162446   -1.7162446    50.486676 
-   31000            0    1.0037106            1 0.00056291948 -0.036529699    -1.852421    -1.852421    52.011803 
-   32000            0    1.0022542            1 0.00041877447 -0.036903458   -1.8713743   -1.8713743    52.712289 
-   33000            0    1.0021364            1 0.00040871451 -0.033616728   -1.7047043   -1.7047043    55.199593 
-   34000            0     1.003779            1 0.00040667148 -0.031745412   -1.6098098   -1.6098098    51.055261 
-   35000            0    1.0015547            1 0.00042509365 -0.033603064   -1.7040114   -1.7040114    53.887389 
-   36000            0    1.0024849            1 0.00040589851 -0.032151258   -1.6303903   -1.6303903    55.673615 
-   37000            0    1.0038204            1 0.00040542177 -0.033643242   -1.7060488   -1.7060488    51.834694 
-   38000            0    1.0031777            1 0.00040237126 -0.034711811   -1.7602359   -1.7602359    56.275778 
-   39000            0     1.002777            1 0.00051021239 -0.034375104   -1.7431615   -1.7431615    48.461145 
-   40000            0    1.0053762            1 0.00041566465 -0.036016384   -1.8263908   -1.8263908    48.937456 
-   41000            0    1.0039078            1 0.00049726673 -0.036481888   -1.8499966   -1.8499966    49.290465 
-   42000            0    1.0033408            1 0.00045236191 -0.034254524   -1.7370469   -1.7370469    52.146603 
-   43000            0    1.0039893            1 0.00048362351 -0.032519474   -1.6490625   -1.6490625    53.320599 
-   44000            0    1.0036667            1 0.00046834006 -0.031099875   -1.5770747   -1.5770747    55.907531 
-   45000            0    1.0035765            1 0.0005195031 -0.029559708   -1.4989728   -1.4989728    56.191165 
-   46000            0     1.004514            1 0.00050134348 -0.028852909    -1.463131    -1.463131    57.547696 
-   47000            0     1.003864            1 0.00038963389 -0.031339229   -1.5892123   -1.5892123    55.775764 
-   48000            0     1.003145            1 0.00035436452 -0.032390682   -1.6425315   -1.6425315    55.233354 
-   49000            0    1.0042615            1 0.00053161075 -0.029960468   -1.5192953   -1.5192953    54.611685 
-   50000            0    1.0037096            1 0.00031191389  -0.02942011   -1.4918938   -1.4918938    58.653762 
-Loop time of 27.1275 on 8 procs for 40000 steps with 20284 atoms
-
-Pair  time (%) = 1.54277 (5.68709)
-Neigh time (%) = 1.26864 (4.67658)
-Comm  time (%) = 1.19871 (4.4188)
-Outpt time (%) = 0.00421953 (0.0155544)
-Other time (%) = 23.1132 (85.202)
-
-Nlocal:    2535.5 ave 2670 max 2406 min
-Histogram: 1 1 1 0 0 2 1 1 0 1
-Nghost:    114.5 ave 124 max 103 min
-Histogram: 1 0 1 0 2 1 0 1 0 2
-Neighs:    521.5 ave 692 max 442 min
-Histogram: 3 1 1 1 0 0 0 1 0 1
-
-Total # of neighbors = 4172
-Ave neighs/atom = 0.205679
-Neighbor list builds = 2002
-Dangerous builds = 0
-
-Please see the log.cite file for references relevant to this simulation
-

examples/ASPHERE/line/log.21Sep24.line.g++.1

@@ -0,0 +1,189 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-296-g061c80104a-modified)
+  using 1 OpenMP thread(s) per MPI task
+# Aspherical shear demo - 2d line box and triangle mixture, implicit solvent
+
+units           lj
+atom_style      line
+dimension       2
+
+read_data       data.line
+Reading data file ...
+  orthogonal box = (-22.553882 -22.553882 -0.5) to (22.553882 22.553882 0.5)
+  1 by 1 by 1 MPI processor grid
+  reading atoms ...
+  350 atoms
+  350 lines
+  read_data CPU = 0.003 seconds
+
+velocity        all create 1.44 320984 loop geom
+
+neighbor        0.3 bin
+neigh_modify    delay 0 every 1 check yes
+neigh_modify    exclude molecule/intra all
+
+pair_style      line/lj 2.5
+pair_coeff      * * 0.25 0.25 1.0 0.25 2.5
+
+fix             2 all rigid molecule langevin 2.0 2.0 1.0 492983
+  100 rigid bodies with 350 atoms
+
+fix             3 all deform 5 x scale 0.8 y scale 0.8
+fix             4 all enforce2d
+
+compute         10 all property/atom end1x end1y end2x end2y
+
+#dump           1 all custom 500 dump1.atom id type x y z ix iy iz
+#dump           2 all custom 500 dump1.line id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+timestep        0.001
+
+compute         1 all erotate/asphere
+compute         2 all ke
+compute         3 all pe
+variable        toteng equal (c_1+c_2+c_3)/atoms
+
+compute_modify  thermo_temp extra/dof -350
+
+thermo          1000
+thermo_style    custom step f_2 pe ke c_1 c_2 c_3 v_toteng
+
+run             10000
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Your simulation uses code contributions which should be cited:
+
+- Type Label Framework: https://doi.org/10.1021/acs.jpcb.3c08419
+
+@Article{Gissinger24,
+ author = {Jacob R. Gissinger, Ilia Nikiforov, Yaser Afshar, Brendon Waters, Moon-ki Choi, Daniel S. Karls, Alexander Stukowski, Wonpil Im, Hendrik Heinz, Axel Kohlmeyer, and Ellad B. Tadmor},
+ title = {Type Label Framework for Bonded Force Fields in LAMMPS},
+ journal = {J. Phys. Chem. B},
+ year =    2024,
+ volume =  128,
+ number =  13,
+ pages =   {3282–-3297}
+}
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 2.8
+  ghost atom cutoff = 2.8
+  binsize = 1.4, bins = 33 33 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/2d
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 4.742 | 4.742 | 4.742 Mbytes
+   Step          f_2           PotEng         KinEng          c_1            c_2            c_3          v_toteng   
+         0   1.1872976      0              0.46543528     0.03617212     0.46543528     0              0.5016074    
+      1000   1.9084412     -0.001043719    0.71003395     0.089891202    0.71003395    -0.001043719    0.79888143   
+      2000   2.31668       -0.020711665    0.83611544     0.13062287     0.83611544    -0.020711665    0.94602664   
+      3000   2.3094506     -0.062018072    0.80584123     0.15326943     0.80584123    -0.062018072    0.8970926    
+      4000   2.4383442     -0.053659995    0.86864073     0.14696993     0.86864073    -0.053659995    0.96195067   
+      5000   2.5885917     -0.15612821     0.90351683     0.17156633     0.90351683    -0.15612821     0.91895495   
+      6000   2.1187379     -0.072041135    0.76933527     0.11557939     0.76933527    -0.072041135    0.81287353   
+      7000   2.1407592     -0.057727294    0.75154343     0.13827066     0.75154343    -0.057727294    0.8320868    
+      8000   2.0661844     -0.097529608    0.71093839     0.14547433     0.71093839    -0.097529608    0.75888311   
+      9000   2.1574445     -0.052890857    0.78826145     0.11363136     0.78826145    -0.052890857    0.84900195   
+     10000   2.473097      -0.092376489    0.85756337     0.16861278     0.85756337    -0.092376489    0.93379965   
+Loop time of 1.16704 on 1 procs for 10000 steps with 350 atoms
+
+Performance: 740334.372 tau/day, 8568.685 timesteps/s, 2.999 Matom-step/s
+99.7% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.80757    | 0.80757    | 0.80757    |   0.0 | 69.20
+Neigh   | 0.0090046  | 0.0090046  | 0.0090046  |   0.0 |  0.77
+Comm    | 0.011808   | 0.011808   | 0.011808   |   0.0 |  1.01
+Output  | 0.000194   | 0.000194   | 0.000194   |   0.0 |  0.02
+Modify  | 0.3317     | 0.3317     | 0.3317     |   0.0 | 28.42
+Other   |            | 0.006768   |            |       |  0.58
+
+Nlocal:            350 ave         350 max         350 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:            101 ave         101 max         101 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:            853 ave         853 max         853 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 853
+Ave neighs/atom = 2.4371429
+Neighbor list builds = 274
+Dangerous builds = 0
+
+#undump          1
+#undump          2
+unfix           3
+
+change_box      all triclinic
+Changing box ...
+  triclinic box = (-18.043106 -18.043106 -0.5) to (18.043106 18.043106 0.5) with tilt (0 0 0)
+
+#dump           1 all custom 500 dump2.atom id type x y z ix iy iz
+#dump           2 all custom 500 dump2.line id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+fix             3 all deform 1 xy erate 0.01 units box
+
+run             10000
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 2.8
+  ghost atom cutoff = 2.8
+  binsize = 1.4, bins = 26 26 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton/tri
+      stencil: half/bin/2d/tri
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 4.756 | 4.756 | 4.756 Mbytes
+   Step          f_2           PotEng         KinEng          c_1            c_2            c_3          v_toteng   
+     10000   2.473097      -0.092380532    0.85753671     0.16863499     0.85753671    -0.092380532    0.93379117   
+     11000   2.5185915     -0.1015737      0.84230876     0.197573       0.84230876    -0.1015737      0.93830806   
+     12000   2.5971728     -0.13087195     0.90608898     0.17248755     0.90608898    -0.13087195     0.94770458   
+     13000   2.5065714     -0.14666584     0.83804307     0.19683483     0.83804307    -0.14666584     0.88821206   
+     14000   2.1672357     -0.10720878     0.79713452     0.10973398     0.79713452    -0.10720878     0.79965972   
+     15000   2.3444438     -0.14780285     0.81642139     0.15695019     0.81642139    -0.14780285     0.82556873   
+     16000   2.2672973     -0.12590007     0.76672668     0.17081014     0.76672668    -0.12590007     0.81163675   
+     17000   2.3999651     -0.15464102     0.85708567     0.14289234     0.85708567    -0.15464102     0.84533699   
+     18000   2.2024356     -0.14077779     0.76727873     0.14718521     0.76727873    -0.14077779     0.77368615   
+     19000   2.5411572     -0.092014576    0.85750101     0.19297197     0.85750101    -0.092014576    0.95845841   
+     20000   2.2113691     -0.11944862     0.77295445     0.14564597     0.77295445    -0.11944862     0.79915181   
+Loop time of 1.76388 on 1 procs for 10000 steps with 350 atoms
+
+Performance: 489827.896 tau/day, 5669.304 timesteps/s, 1.984 Matom-step/s
+99.6% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 1.3668     | 1.3668     | 1.3668     |   0.0 | 77.49
+Neigh   | 0.018849   | 0.018849   | 0.018849   |   0.0 |  1.07
+Comm    | 0.013338   | 0.013338   | 0.013338   |   0.0 |  0.76
+Output  | 0.00020434 | 0.00020434 | 0.00020434 |   0.0 |  0.01
+Modify  | 0.35698    | 0.35698    | 0.35698    |   0.0 | 20.24
+Other   |            | 0.007667   |            |       |  0.43
+
+Nlocal:            350 ave         350 max         350 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:            119 ave         119 max         119 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:            997 ave         997 max         997 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 997
+Ave neighs/atom = 2.8485714
+Neighbor list builds = 283
+Dangerous builds = 0
+Total wall time: 0:00:02

examples/ASPHERE/line/log.21Sep24.line.g++.4

@@ -0,0 +1,189 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-296-g061c80104a-modified)
+  using 1 OpenMP thread(s) per MPI task
+# Aspherical shear demo - 2d line box and triangle mixture, implicit solvent
+
+units           lj
+atom_style      line
+dimension       2
+
+read_data       data.line
+Reading data file ...
+  orthogonal box = (-22.553882 -22.553882 -0.5) to (22.553882 22.553882 0.5)
+  2 by 2 by 1 MPI processor grid
+  reading atoms ...
+  350 atoms
+  350 lines
+  read_data CPU = 0.003 seconds
+
+velocity        all create 1.44 320984 loop geom
+
+neighbor        0.3 bin
+neigh_modify    delay 0 every 1 check yes
+neigh_modify    exclude molecule/intra all
+
+pair_style      line/lj 2.5
+pair_coeff      * * 0.25 0.25 1.0 0.25 2.5
+
+fix             2 all rigid molecule langevin 2.0 2.0 1.0 492983
+  100 rigid bodies with 350 atoms
+
+fix             3 all deform 5 x scale 0.8 y scale 0.8
+fix             4 all enforce2d
+
+compute         10 all property/atom end1x end1y end2x end2y
+
+#dump           1 all custom 500 dump1.atom id type x y z ix iy iz
+#dump           2 all custom 500 dump1.line id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+timestep        0.001
+
+compute         1 all erotate/asphere
+compute         2 all ke
+compute         3 all pe
+variable        toteng equal (c_1+c_2+c_3)/atoms
+
+compute_modify  thermo_temp extra/dof -350
+
+thermo          1000
+thermo_style    custom step f_2 pe ke c_1 c_2 c_3 v_toteng
+
+run             10000
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Your simulation uses code contributions which should be cited:
+
+- Type Label Framework: https://doi.org/10.1021/acs.jpcb.3c08419
+
+@Article{Gissinger24,
+ author = {Jacob R. Gissinger, Ilia Nikiforov, Yaser Afshar, Brendon Waters, Moon-ki Choi, Daniel S. Karls, Alexander Stukowski, Wonpil Im, Hendrik Heinz, Axel Kohlmeyer, and Ellad B. Tadmor},
+ title = {Type Label Framework for Bonded Force Fields in LAMMPS},
+ journal = {J. Phys. Chem. B},
+ year =    2024,
+ volume =  128,
+ number =  13,
+ pages =   {3282–-3297}
+}
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 2.8
+  ghost atom cutoff = 2.8
+  binsize = 1.4, bins = 33 33 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/2d
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 4.736 | 4.736 | 4.736 Mbytes
+   Step          f_2           PotEng         KinEng          c_1            c_2            c_3          v_toteng   
+         0   1.1872976      0              0.46543528     0.03617212     0.46543528     0              0.5016074    
+      1000   1.9084412     -0.001043719    0.71003395     0.089891202    0.71003395    -0.001043719    0.79888143   
+      2000   2.31668       -0.020711665    0.83611544     0.13062287     0.83611544    -0.020711665    0.94602664   
+      3000   2.3094506     -0.062018072    0.80584123     0.15326943     0.80584123    -0.062018072    0.8970926    
+      4000   2.4383442     -0.053659995    0.86864073     0.14696993     0.86864073    -0.053659995    0.96195067   
+      5000   2.5885917     -0.15612821     0.90351683     0.17156633     0.90351683    -0.15612821     0.91895494   
+      6000   2.118738      -0.072041141    0.76933528     0.1155794      0.76933528    -0.072041141    0.81287354   
+      7000   2.1407609     -0.057727453    0.75154404     0.13827078     0.75154404    -0.057727453    0.83208737   
+      8000   2.066348      -0.097639137    0.71096217     0.14551295     0.71096217    -0.097639137    0.75883597   
+      9000   2.1969096     -0.050133795    0.8028133      0.11559948     0.8028133     -0.050133795    0.86827899   
+     10000   2.3932442     -0.098008752    0.83753779     0.15678193     0.83753779    -0.098008752    0.89631097   
+Loop time of 0.596624 on 4 procs for 10000 steps with 350 atoms
+
+Performance: 1448149.372 tau/day, 16760.988 timesteps/s, 5.866 Matom-step/s
+98.2% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.17419    | 0.21073    | 0.27006    |   7.8 | 35.32
+Neigh   | 0.0015899  | 0.0018079  | 0.0020932  |   0.5 |  0.30
+Comm    | 0.041688   | 0.101      | 0.13805    |  11.3 | 16.93
+Output  | 0.00011945 | 0.0001262  | 0.000143   |   0.0 |  0.02
+Modify  | 0.26906    | 0.27183    | 0.27467    |   0.4 | 45.56
+Other   |            | 0.01113    |            |       |  1.87
+
+Nlocal:           87.5 ave          94 max          80 min
+Histogram: 1 1 0 0 0 0 0 0 0 2
+Nghost:           54.5 ave          67 max          42 min
+Histogram: 1 0 1 0 0 0 0 1 0 1
+Neighs:         212.75 ave         286 max         122 min
+Histogram: 1 0 0 1 0 0 0 0 1 1
+
+Total # of neighbors = 851
+Ave neighs/atom = 2.4314286
+Neighbor list builds = 273
+Dangerous builds = 0
+
+#undump          1
+#undump          2
+unfix           3
+
+change_box      all triclinic
+Changing box ...
+  triclinic box = (-18.043106 -18.043106 -0.5) to (18.043106 18.043106 0.5) with tilt (0 0 0)
+
+#dump           1 all custom 500 dump2.atom id type x y z ix iy iz
+#dump           2 all custom 500 dump2.line id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+fix             3 all deform 1 xy erate 0.01 units box
+
+run             10000
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 2.8
+  ghost atom cutoff = 2.8
+  binsize = 1.4, bins = 26 26 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton/tri
+      stencil: half/bin/2d/tri
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 4.752 | 4.752 | 4.752 Mbytes
+   Step          f_2           PotEng         KinEng          c_1            c_2            c_3          v_toteng   
+     10000   2.3932442     -0.098029446    0.83751359     0.1568021      0.83751359    -0.098029446    0.89628624   
+     11000   2.4541893     -0.15538223     0.82053681     0.19271549     0.82053681    -0.15538223     0.85787007   
+     12000   2.4124449     -0.14024177     0.84559598     0.15692416     0.84559598    -0.14024177     0.86227837   
+     13000   2.2095814     -0.10822636     0.73930104     0.173052       0.73930104    -0.10822636     0.80412668   
+     14000   2.0946831     -0.087303541    0.77743494     0.10023865     0.77743494    -0.087303541    0.79037005   
+     15000   2.0911016     -0.11524594     0.74044758     0.12978235     0.74044758    -0.11524594     0.75498398   
+     16000   1.9736428     -0.17259563     0.67852978     0.13943094     0.67852978    -0.17259563     0.64536509   
+     17000   2.3284367     -0.17673537     0.77018991     0.1897596      0.77018991    -0.17673537     0.78321414   
+     18000   2.3178564     -0.16634746     0.81488323     0.1487365      0.81488323    -0.16634746     0.79727227   
+     19000   2.7497251     -0.18254513     1.0041125      0.14528424     1.0041125     -0.18254513     0.96685165   
+     20000   2.3897059     -0.12664177     0.8390411      0.15426546     0.8390411     -0.12664177     0.86666479   
+Loop time of 0.79519 on 4 procs for 10000 steps with 350 atoms
+
+Performance: 1086533.223 tau/day, 12575.616 timesteps/s, 4.401 Matom-step/s
+98.2% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.26601    | 0.35051    | 0.443      |  12.3 | 44.08
+Neigh   | 0.0029891  | 0.0036725  | 0.0040631  |   0.7 |  0.46
+Comm    | 0.049744   | 0.14268    | 0.22734    |  19.3 | 17.94
+Output  | 0.00013783 | 0.00014503 | 0.00016304 |   0.0 |  0.02
+Modify  | 0.27963    | 0.28446    | 0.29049    |   0.8 | 35.77
+Other   |            | 0.01372    |            |       |  1.73
+
+Nlocal:           87.5 ave         101 max          75 min
+Histogram: 1 0 1 0 0 0 1 0 0 1
+Nghost:           61.5 ave          67 max          50 min
+Histogram: 1 0 0 0 0 0 0 0 2 1
+Neighs:         244.75 ave         268 max         198 min
+Histogram: 1 0 0 0 0 0 0 1 0 2
+
+Total # of neighbors = 979
+Ave neighs/atom = 2.7971429
+Neighbor list builds = 289
+Dangerous builds = 0
+Total wall time: 0:00:01

examples/ASPHERE/line/log.21Sep24.line.srd.g++.1

@@ -0,0 +1,363 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-296-g061c80104a-modified)
+  using 1 OpenMP thread(s) per MPI task
+# Aspherical shear demo - 2d line boxes, solvated by SRD particles
+
+units           lj
+atom_style      line
+atom_modify     first big
+dimension       2
+
+read_data       data.line.srd
+Reading data file ...
+  orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  1 by 1 by 1 MPI processor grid
+  reading atoms ...
+  400 atoms
+  400 lines
+  read_data CPU = 0.003 seconds
+
+# add small particles as hi density lattice
+
+lattice         sq 0.4
+Lattice spacing in x,y,z = 1.5811388 1.5811388 1.5811388
+region          plane block INF INF INF INF -0.001 0.001
+lattice         sq 10.0
+Lattice spacing in x,y,z = 0.31622777 0.31622777 0.31622777
+create_atoms    2 region plane
+Created 33489 atoms
+  using lattice units in orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  create_atoms CPU = 0.003 seconds
+
+group           big type 1
+400 atoms in group big
+group           small type 2
+33489 atoms in group small
+set             group small mass 0.01
+Setting atom values ...
+  33489 settings made for mass
+
+# delete overlaps
+# must set 1-2 cutoff to non-zero value
+
+pair_style      lj/cut  1.5
+pair_coeff      1 1 1.0 1.0
+pair_coeff      2 2 0.0 1.0 0.0
+pair_coeff      1 2 1.0 1.0
+
+delete_atoms    overlap 1.5 small big
+System init for delete_atoms ...
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 1.8
+  ghost atom cutoff = 1.8
+  binsize = 0.9, bins = 64 64 2
+  2 neighbor lists, perpetual/occasional/extra = 1 1 0
+  (1) command delete_atoms, occasional
+      attributes: full, newton on
+      pair build: full/bin/atomonly
+      stencil: full/bin/2d
+      bin: standard
+  (2) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/2d
+      bin: standard
+WARNING: Delete_atoms cutoff > minimum neighbor cutoff (src/delete_atoms.cpp:313)
+Deleted 13605 atoms, new total = 20284
+
+# SRD run
+
+reset_timestep  0
+
+velocity        small create 1.44 87287 loop geom
+
+neighbor        0.8 multi
+neigh_modify    delay 0 every 1 check yes
+neigh_modify    exclude molecule/intra big include big
+
+comm_modify     mode multi group big vel yes
+neigh_modify    include big
+
+# no pairwise interactions with small particles
+
+pair_style      line/lj 2.5
+pair_coeff      1 1 1.0  1.0 1.0 1.0 2.5
+pair_coeff      2 2 0.0  0.0 0.0 1.0 0.0
+pair_coeff      1 2 1.0  0.2 1.0 0.2 2.5
+
+# use fix SRD to push small particles out from inside big ones
+# if comment out, big particles won't see SRD particles
+
+timestep        0.0001
+
+fix             1 big rigid molecule
+  100 rigid bodies with 400 atoms
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip
+
+fix             3 all deform 1 x scale 1.25 y scale 1.25
+fix             4 all enforce2d
+
+# diagnostics
+
+compute         tsmall small temp/deform
+compute         tbig big temp
+variable        pebig equal pe*atoms/count(big)
+variable        ebig equal etotal*atoms/count(big)
+
+compute_modify  tbig extra/dof -350
+
+compute         1 big erotate/asphere
+compute         2 all ke
+compute         3 all pe
+variable        toteng equal (c_1+c_2+c_3)/atoms
+
+thermo          1000
+thermo_style    custom step c_tsmall f_2[9] c_1 etotal                 v_pebig v_ebig press
+thermo_modify   temp tbig
+WARNING: Temperature for thermo pressure is not for group all (src/thermo.cpp:533)
+
+compute         10 big property/atom end1x end1y end2x end2y
+
+#dump           1 all custom 500 dump1.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump1.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+run             10000
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Your simulation uses code contributions which should be cited:
+
+- Type Label Framework: https://doi.org/10.1021/acs.jpcb.3c08419
+
+@Article{Gissinger24,
+ author = {Jacob R. Gissinger, Ilia Nikiforov, Yaser Afshar, Brendon Waters, Moon-ki Choi, Daniel S. Karls, Alexander Stukowski, Wonpil Im, Hendrik Heinz, Axel Kohlmeyer, and Ellad B. Tadmor},
+ title = {Type Label Framework for Bonded Force Fields in LAMMPS},
+ journal = {J. Phys. Chem. B},
+ year =    2024,
+ volume =  128,
+ number =  13,
+ pages =   {3282–-3297}
+}
+
+- neighbor multi command: doi:10.1016/j.cpc.2008.03.005, doi:10.1007/s40571-020-00361-2
+
+@Article{Intveld08,
+ author =  {in 't Veld, P. J. and S. J.~Plimpton and G. S. Grest},
+ title =   {Accurate and Efficient Methods for Modeling Colloidal
+            Mixtures in an Explicit Solvent using Molecular Dynamics},
+ journal = {Comput.\ Phys.\ Commut.},
+ year =    2008,
+ volume =  179,
+ pages =   {320--329}
+}
+
+@article{Shire2020,
+ author = {Shire, Tom and Hanley, Kevin J. and Stratford, Kevin},
+ title = {{DEM} Simulations of Polydisperse Media: Efficient Contact
+          Detection Applied to Investigate the Quasi-Static Limit},
+ journal = {Computational Particle Mechanics},
+ year = {2020}
+@article{Monti2022,
+ author = {Monti, Joseph M. and Clemmer, Joel T. and Srivastava, 
+           Ishan and Silbert, Leonardo E. and Grest, Gary S. 
+           and Lechman, Jeremy B.},
+ title = {Large-scale frictionless jamming with power-law particle 
+          size distributions},
+ journal = {Phys. Rev. E},
+ volume = {106}
+ issue = {3}
+ year = {2022}
+}
+
+- fix srd command: doi:10.1063/1.3419070
+
+@Article{Petersen10,
+ author = {M. K. Petersen and J. B. Lechman and S. J. Plimpton and
+ G. S. Grest and in 't Veld, P. J. and P. R. Schunk},
+ title =   {Mesoscale Hydrodynamics via Stochastic Rotation
+    Dynamics: Comparison with {L}ennard-{J}ones Fluid},
+ journal = {J.~Chem.\ Phys.},
+ year =    2010,
+ volume =  132,
+ pages =   174106
+}
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using compute temp/deform with inconsistent fix deform remap option (src/compute_temp_deform.cpp:71)
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.02
+  SRD max distance & max velocity = 0.08 40
+  SRD grid counts: 230 230 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.25040659 0.25040659 1
+  SRD per actual grid cell = 0.45503978
+  SRD viscosity = 0.92062623
+  big/SRD mass density ratio = 12.298053
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 0
+  ave/max small velocity = 15.906001 29.105426
+  ave/max big velocity = 0 0
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 35 35 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton
+      stencil: half/multi/2d
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 43.18 | 43.18 | 43.18 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+         0   1.6084386      0              0              0              0              0              0            
+      1000   1.4566787      1.2042825      0.00015900089  0.0011955837   0              0.060628048    0.75322967   
+      2000   1.3561018      1.2092936      0.00026739313  0.0020783306   0              0.10539215     0.53247025   
+      3000   1.3327447      1.1696777      0.0003215116   0.0024935719   0              0.12644903     0.32024668   
+      4000   1.3031273      1.1385596      0.00037337107  0.0031214473   0              0.15828859     0.64066244   
+      5000   1.28204        1.0983763      0.00047966362  0.0033295434   0              0.16884114     0.32103666   
+      6000   1.2654396      1.0969562      0.00049917373  0.0037382387   0              0.18956609     0.16211614   
+      7000   1.2401503      1.1232046      0.00045137168  0.0037522201   0              0.19027508     0.014071672  
+      8000   1.2349025      1.059008       0.00056648801  0.0039664742   0              0.20113991     0.24483817   
+      9000   1.2272907      1.0371006      0.00056271544  0.0037847074   0              0.19192251     0.032968056  
+     10000   1.2153817      1.0209113      0.00053291164  0.0037197691   0              0.18862949     0.3237698    
+Loop time of 6.0346 on 1 procs for 10000 steps with 20284 atoms
+
+Performance: 14317.424 tau/day, 1657.109 timesteps/s, 33.613 Matom-step/s
+96.2% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.10636    | 0.10636    | 0.10636    |   0.0 |  1.76
+Neigh   | 0.01905    | 0.01905    | 0.01905    |   0.0 |  0.32
+Comm    | 0.041223   | 0.041223   | 0.041223   |   0.0 |  0.68
+Output  | 0.0017015  | 0.0017015  | 0.0017015  |   0.0 |  0.03
+Modify  | 5.8165     | 5.8165     | 5.8165     |   0.0 | 96.39
+Other   |            | 0.04976    |            |       |  0.82
+
+Nlocal:          20284 ave       20284 max       20284 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:             84 ave          84 max          84 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:              0 ave           0 max           0 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 0
+Ave neighs/atom = 0
+Neighbor list builds = 500
+Dangerous builds = 0
+
+#undump          1
+#undump          2
+unfix           3
+
+change_box      all triclinic
+Changing box ...
+  triclinic box = (-35.995947 -35.995947 -0.5) to (35.995947 35.995947 0.5) with tilt (0 0 0)
+
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip #tstat yes
+
+#dump           1 all custom 500 dump2.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump2.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+fix             3 all deform 1 xy erate 0.0002 units box remap v
+
+run             30000
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.02
+  SRD max distance & max velocity = 0.08 40
+  SRD grid counts: 288 288 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.24997185 0.24997185 1
+  SRD per actual grid cell = 0.26976265
+  SRD viscosity = 1.0312189
+  big/SRD mass density ratio = 20.672578
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 0
+  ave/max small velocity = 12.805429 40
+  ave/max big velocity = 0.50400801 2.1697856
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 44 44 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton/tri
+      stencil: half/multi/2d/tri
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 58.85 | 58.85 | 58.85 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+     10000   1.0423767      0              0.00053289774  0.0037197858   0              0.18863034     0.014558088  
+     11000   1.0333923      1.0309908      0.00058283908  0.0042336078   0              0.21468625     0.13036048   
+     12000   1.0268543      1.0290122      0.00055294944  0.0044278861   0              0.2245381      0.53275676   
+     13000   1.0257856      1.0140434      0.00051520803  0.0042888387   0              0.21748701     0.2066835    
+     14000   1.0239497      1.0321558      0.00054742053  0.004259664    0              0.21600756     0.24430097   
+     15000   1.0194475      1.0119806      0.00053619565  0.0041113765   0              0.2084879      0.18420434   
+     16000   1.016421       1.0309991      0.00049553642  0.0041338645   0              0.20962827     0.18298406   
+     17000   1.0088527      0.99723694     0.00049824173  0.0044777581   0              0.22706711     0.65557023   
+     18000   1.0073877      1.0136962      0.00058327835  0.0045355884   0              0.22999969     0.13700737   
+     19000   1.0026916      1.0253129      0.00056760151  0.0045138437   0              0.22889701     0.16443344   
+     20000   1.0024042      0.98988084     0.00057568118  0.0048644983   0              0.24667871     0.060864322  
+     21000   0.99500641     0.99075324     0.00062187543  0.0050480594   0              0.25598709     0.18428384   
+     22000   0.99286446     0.98642266     0.00061024492  0.0050497174   0              0.25607117     0.18499738   
+     23000   0.99204086     0.98869005     0.00064013575  0.0052921112   0              0.26836296     0.046987646  
+     24000   0.98768375     1.0110613      0.00065868636  0.0049285304   0              0.24992578     0.22623751   
+     25000   0.98763308     0.98752678     0.00067070463  0.0049364253   0              0.25032613     0.022819382  
+     26000   0.9844889      1.0059918      0.00075364589  0.0053381778   0              0.270699       0.089735771  
+     27000   0.98574608     0.99094039     0.00080711088  0.0054137233   0              0.27452991     0.14477885   
+     28000   0.98139419     0.97487219     0.00071242405  0.0053315389   0              0.27036234    -0.016357088  
+     29000   0.98458705     0.98600894     0.00068355573  0.0050157315   0              0.25434775     0.015749835  
+     30000   0.98440759     1.0010793      0.00067594666  0.0048121267   0              0.24402295     0.20422918   
+     31000   0.98060632     0.9825933      0.00069529947  0.0050649881  -0.00010700206  0.25684555     0.0014601267 
+     32000   0.97616624     0.94976641     0.00067682494  0.0052111319  -0.00013230279  0.2642565      0.013738873  
+     33000   0.97926122     0.95756794     0.00071774059  0.0053716513  -0.00024804252  0.27239644     0.10643895   
+     34000   0.97227771     0.94365547     0.00080885359  0.0055744144  -0.00063217879  0.28267856     0.12472959   
+     35000   0.97295679     0.98103625     0.0008130852   0.0057598371  -0.00097218501  0.29208134     0.88422099   
+     36000   0.97131919     0.97164636     0.0008345433   0.0062212675  -0.001543055    0.31548048     0.22389401   
+     37000   0.96988732     0.98589058     0.00083271625  0.0062999411  -0.0026064482   0.31947001     0.11797226   
+     38000   0.97135512     0.96486494     0.00076943974  0.0057397763  -0.0041174109   0.29106406     0.072790492  
+     39000   0.97386738     0.95218062     0.00079594844  0.0056308358  -0.0069006312   0.28553968     0.12417837   
+     40000   0.97914045     1.0033941      0.000830798    0.0055952714  -0.0086705514   0.28373621     0.1976663    
+Loop time of 15.944 on 1 procs for 30000 steps with 20284 atoms
+
+Performance: 16256.931 tau/day, 1881.589 timesteps/s, 38.166 Matom-step/s
+99.5% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.31871    | 0.31871    | 0.31871    |   0.0 |  2.00
+Neigh   | 0.082974   | 0.082974   | 0.082974   |   0.0 |  0.52
+Comm    | 0.16092    | 0.16092    | 0.16092    |   0.0 |  1.01
+Output  | 0.0047257  | 0.0047257  | 0.0047257  |   0.0 |  0.03
+Modify  | 15.189     | 15.189     | 15.189     |   0.0 | 95.27
+Other   |            | 0.1871     |            |       |  1.17
+
+Nlocal:          20284 ave       20284 max       20284 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:             76 ave          76 max          76 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:             38 ave          38 max          38 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 38
+Ave neighs/atom = 0.0018733978
+Neighbor list builds = 1500
+Dangerous builds = 0
+Total wall time: 0:00:22

examples/ASPHERE/line/log.21Sep24.line.srd.g++.4

@@ -0,0 +1,363 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-296-g061c80104a-modified)
+  using 1 OpenMP thread(s) per MPI task
+# Aspherical shear demo - 2d line boxes, solvated by SRD particles
+
+units           lj
+atom_style      line
+atom_modify     first big
+dimension       2
+
+read_data       data.line.srd
+Reading data file ...
+  orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  2 by 2 by 1 MPI processor grid
+  reading atoms ...
+  400 atoms
+  400 lines
+  read_data CPU = 0.003 seconds
+
+# add small particles as hi density lattice
+
+lattice         sq 0.4
+Lattice spacing in x,y,z = 1.5811388 1.5811388 1.5811388
+region          plane block INF INF INF INF -0.001 0.001
+lattice         sq 10.0
+Lattice spacing in x,y,z = 0.31622777 0.31622777 0.31622777
+create_atoms    2 region plane
+Created 33489 atoms
+  using lattice units in orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  create_atoms CPU = 0.001 seconds
+
+group           big type 1
+400 atoms in group big
+group           small type 2
+33489 atoms in group small
+set             group small mass 0.01
+Setting atom values ...
+  33489 settings made for mass
+
+# delete overlaps
+# must set 1-2 cutoff to non-zero value
+
+pair_style      lj/cut  1.5
+pair_coeff      1 1 1.0 1.0
+pair_coeff      2 2 0.0 1.0 0.0
+pair_coeff      1 2 1.0 1.0
+
+delete_atoms    overlap 1.5 small big
+System init for delete_atoms ...
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 1.8
+  ghost atom cutoff = 1.8
+  binsize = 0.9, bins = 64 64 2
+  2 neighbor lists, perpetual/occasional/extra = 1 1 0
+  (1) command delete_atoms, occasional
+      attributes: full, newton on
+      pair build: full/bin/atomonly
+      stencil: full/bin/2d
+      bin: standard
+  (2) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/2d
+      bin: standard
+WARNING: Delete_atoms cutoff > minimum neighbor cutoff (src/delete_atoms.cpp:313)
+Deleted 13605 atoms, new total = 20284
+
+# SRD run
+
+reset_timestep  0
+
+velocity        small create 1.44 87287 loop geom
+
+neighbor        0.8 multi
+neigh_modify    delay 0 every 1 check yes
+neigh_modify    exclude molecule/intra big include big
+
+comm_modify     mode multi group big vel yes
+neigh_modify    include big
+
+# no pairwise interactions with small particles
+
+pair_style      line/lj 2.5
+pair_coeff      1 1 1.0  1.0 1.0 1.0 2.5
+pair_coeff      2 2 0.0  0.0 0.0 1.0 0.0
+pair_coeff      1 2 1.0  0.2 1.0 0.2 2.5
+
+# use fix SRD to push small particles out from inside big ones
+# if comment out, big particles won't see SRD particles
+
+timestep        0.0001
+
+fix             1 big rigid molecule
+  100 rigid bodies with 400 atoms
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip
+
+fix             3 all deform 1 x scale 1.25 y scale 1.25
+fix             4 all enforce2d
+
+# diagnostics
+
+compute         tsmall small temp/deform
+compute         tbig big temp
+variable        pebig equal pe*atoms/count(big)
+variable        ebig equal etotal*atoms/count(big)
+
+compute_modify  tbig extra/dof -350
+
+compute         1 big erotate/asphere
+compute         2 all ke
+compute         3 all pe
+variable        toteng equal (c_1+c_2+c_3)/atoms
+
+thermo          1000
+thermo_style    custom step c_tsmall f_2[9] c_1 etotal                 v_pebig v_ebig press
+thermo_modify   temp tbig
+WARNING: Temperature for thermo pressure is not for group all (src/thermo.cpp:533)
+
+compute         10 big property/atom end1x end1y end2x end2y
+
+#dump           1 all custom 500 dump1.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump1.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+run             10000
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Your simulation uses code contributions which should be cited:
+
+- Type Label Framework: https://doi.org/10.1021/acs.jpcb.3c08419
+
+@Article{Gissinger24,
+ author = {Jacob R. Gissinger, Ilia Nikiforov, Yaser Afshar, Brendon Waters, Moon-ki Choi, Daniel S. Karls, Alexander Stukowski, Wonpil Im, Hendrik Heinz, Axel Kohlmeyer, and Ellad B. Tadmor},
+ title = {Type Label Framework for Bonded Force Fields in LAMMPS},
+ journal = {J. Phys. Chem. B},
+ year =    2024,
+ volume =  128,
+ number =  13,
+ pages =   {3282–-3297}
+}
+
+- neighbor multi command: doi:10.1016/j.cpc.2008.03.005, doi:10.1007/s40571-020-00361-2
+
+@Article{Intveld08,
+ author =  {in 't Veld, P. J. and S. J.~Plimpton and G. S. Grest},
+ title =   {Accurate and Efficient Methods for Modeling Colloidal
+            Mixtures in an Explicit Solvent using Molecular Dynamics},
+ journal = {Comput.\ Phys.\ Commut.},
+ year =    2008,
+ volume =  179,
+ pages =   {320--329}
+}
+
+@article{Shire2020,
+ author = {Shire, Tom and Hanley, Kevin J. and Stratford, Kevin},
+ title = {{DEM} Simulations of Polydisperse Media: Efficient Contact
+          Detection Applied to Investigate the Quasi-Static Limit},
+ journal = {Computational Particle Mechanics},
+ year = {2020}
+@article{Monti2022,
+ author = {Monti, Joseph M. and Clemmer, Joel T. and Srivastava, 
+           Ishan and Silbert, Leonardo E. and Grest, Gary S. 
+           and Lechman, Jeremy B.},
+ title = {Large-scale frictionless jamming with power-law particle 
+          size distributions},
+ journal = {Phys. Rev. E},
+ volume = {106}
+ issue = {3}
+ year = {2022}
+}
+
+- fix srd command: doi:10.1063/1.3419070
+
+@Article{Petersen10,
+ author = {M. K. Petersen and J. B. Lechman and S. J. Plimpton and
+ G. S. Grest and in 't Veld, P. J. and P. R. Schunk},
+ title =   {Mesoscale Hydrodynamics via Stochastic Rotation
+    Dynamics: Comparison with {L}ennard-{J}ones Fluid},
+ journal = {J.~Chem.\ Phys.},
+ year =    2010,
+ volume =  132,
+ pages =   174106
+}
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using compute temp/deform with inconsistent fix deform remap option (src/compute_temp_deform.cpp:71)
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.02
+  SRD max distance & max velocity = 0.08 40
+  SRD grid counts: 230 230 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.25040659 0.25040659 1
+  SRD per actual grid cell = 0.45503978
+  SRD viscosity = 0.92062623
+  big/SRD mass density ratio = 12.298053
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 0
+  ave/max small velocity = 15.906001 29.105426
+  ave/max big velocity = 0 0
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 35 35 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton
+      stencil: half/multi/2d
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 13.44 | 13.45 | 13.47 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+         0   1.6084386      0              0              0              0              0              0            
+      1000   1.4539924      1.2578325      0.0001679809   0.0010497614   0              0.053233399    0.69841607   
+      2000   1.3516746      1.1693535      0.00031932331  0.0021450754   0              0.10877677     0.058141137  
+      3000   1.3294093      1.1539986      0.00032558495  0.0022345521   0              0.11331414     0.085892255  
+      4000   1.3049311      1.1174444      0.00039637116  0.0026520968   0              0.13448783     0.10574422   
+      5000   1.2790124      1.1209176      0.0004519172   0.0032635257   0              0.16549339     0.59225702   
+      6000   1.2631214      1.0868777      0.00052274216  0.003527049    0              0.17885666     0.26093936   
+      7000   1.2520054      1.0780954      0.00051683183  0.0038111141   0              0.1932616     -0.0014733978 
+      8000   1.2412037      1.0587149      0.00056620143  0.0038329297   0              0.19436786     0.31576462   
+      9000   1.2242125      1.0699712      0.00065335672  0.0039948578   0              0.20257924     0.19755012   
+     10000   1.2155758      1.0279682      0.00059730828  0.004142635    0              0.21007302     0.30022953   
+Loop time of 1.92412 on 4 procs for 10000 steps with 20284 atoms
+
+Performance: 44903.607 tau/day, 5197.177 timesteps/s, 105.420 Matom-step/s
+97.6% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.026855   | 0.027529   | 0.028377   |   0.4 |  1.43
+Neigh   | 0.0073723  | 0.007433   | 0.0075693  |   0.1 |  0.39
+Comm    | 0.062837   | 0.065574   | 0.066747   |   0.6 |  3.41
+Output  | 0.00072894 | 0.00076558 | 0.00083361 |   0.0 |  0.04
+Modify  | 1.7861     | 1.7926     | 1.8        |   0.4 | 93.16
+Other   |            | 0.03026    |            |       |  1.57
+
+Nlocal:           5071 ave        5096 max        5051 min
+Histogram: 1 0 1 0 0 1 0 0 0 1
+Nghost:             44 ave          51 max          40 min
+Histogram: 2 0 0 1 0 0 0 0 0 1
+Neighs:              0 ave           0 max           0 min
+Histogram: 4 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 0
+Ave neighs/atom = 0
+Neighbor list builds = 500
+Dangerous builds = 0
+
+#undump          1
+#undump          2
+unfix           3
+
+change_box      all triclinic
+Changing box ...
+  triclinic box = (-35.995947 -35.995947 -0.5) to (35.995947 35.995947 0.5) with tilt (0 0 0)
+
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip #tstat yes
+
+#dump           1 all custom 500 dump2.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump2.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+fix             3 all deform 1 xy erate 0.0002 units box remap v
+
+run             30000
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.02
+  SRD max distance & max velocity = 0.08 40
+  SRD grid counts: 288 288 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.24997185 0.24997185 1
+  SRD per actual grid cell = 0.26976265
+  SRD viscosity = 1.0312189
+  big/SRD mass density ratio = 20.672578
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 0
+  ave/max small velocity = 12.826666 40
+  ave/max big velocity = 0.53469722 1.7415554
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 44 44 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton/tri
+      stencil: half/multi/2d/tri
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 17.6 | 17.6 | 17.61 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+     10000   1.0416233      0              0.0005972922   0.0041426543   0              0.210074       0.016213064  
+     11000   1.0366852      1.0236717      0.00066926382  0.004404743    0              0.22336452     0.097686059  
+     12000   1.028695       1.0023004      0.00065323121  0.0043971164   0              0.22297777     0.30007875   
+     13000   1.0223214      1.0296267      0.00060201395  0.0041815724   0              0.21204754     0.17307062   
+     14000   1.0210601      1.0092227      0.00057020066  0.0041936718   0              0.2126611      0.13379405   
+     15000   1.0169223      1.003359       0.00060818329  0.0046120335   0              0.23387622     0.39144341   
+     16000   1.012222       0.99937463     0.00060471571  0.004533278    0              0.22988253     0.16026051   
+     17000   1.0087993      0.98937449     0.00061843646  0.0048888197   0              0.24791205     0.095507016  
+     18000   1.0030288      0.94092316     0.00054252694  0.0046740839   0              0.2370228     -0.070838649  
+     19000   1.0010739      0.99759814     0.00054041829  0.0045481798   0              0.2306382      0.1944996    
+     20000   1.004189       1.0157285      0.00053330893  0.0044999261   0              0.22819125     0.14452619   
+     21000   0.99601133     1.003771       0.00053282188  0.0048954851   0              0.24825005     0.20196263   
+     22000   0.99445408     1.0163465      0.00059954941  0.0051668276   0              0.26200983     0.21332194   
+     23000   0.99663947     0.94920707     0.00057729723  0.0051052499   0              0.25888722     0.19282224   
+     24000   0.99500015     1.0021742      0.00064312956  0.0053430656   0              0.27094686     0.18044164   
+     25000   0.99388189     0.98446723     0.00060150964  0.0054017824   0              0.27392438     0.18844037   
+     26000   0.99632932     0.94670024     0.00052479857  0.004905147    0              0.24874        0.17961314   
+     27000   0.99112962     1.0247118      0.00053159824  0.004752248    0              0.2409865      0.12037737   
+     28000   0.98887153     0.97734068     0.00052255365  0.0050199491   0              0.25456162     0.29110866   
+     29000   0.98938025     0.99467239     0.00053095044  0.0050303107   0              0.25508706     0.59776339   
+     30000   0.99081592     0.98822122     0.00056693985  0.0052452228   0              0.26598525     0.0150843    
+     31000   0.99050903     1.0184912      0.00056742049  0.0056515686   0              0.28659104     0.07877756   
+     32000   0.98929926     0.99257634     0.00058111732  0.005740055    0              0.29107819     0.19146041   
+     33000   0.98698723     0.94872564     0.00056547335  0.0057628447   0              0.29223385     0.076755599  
+     34000   0.98967998     0.98777081     0.00056045905  0.0055645603   0              0.28217885     0.095025284  
+     35000   0.98777734     0.95746323     0.00050104756  0.0055630681  -4.7847642e-05  0.28210318     0.25667997   
+     36000   0.98661597     0.98801246     0.00047074618  0.0054500805  -5.8956193e-05  0.27637358     0.18221607   
+     37000   0.98390111     0.9837894      0.00044581144  0.0054703357  -7.4197741e-05  0.27740072     0.11959303   
+     38000   0.98092646     1.0142057      0.0004945556   0.0056372628  -8.6736668e-05  0.2858656      0.017325263  
+     39000   0.98125957     0.94219822     0.00059691168  0.0060914156  -9.9726556e-05  0.30889569     0.0081217281 
+     40000   0.98362942     0.9649582      0.00062286697  0.0063736358  -0.00010700337  0.32320707     0.293121     
+Loop time of 5.08621 on 4 procs for 30000 steps with 20284 atoms
+
+Performance: 50961.296 tau/day, 5898.298 timesteps/s, 119.641 Matom-step/s
+98.8% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.084321   | 0.084948   | 0.086233   |   0.3 |  1.67
+Neigh   | 0.027521   | 0.029247   | 0.030632   |   0.7 |  0.58
+Comm    | 0.19101    | 0.20262    | 0.2068     |   1.5 |  3.98
+Output  | 0.0019581  | 0.0020369  | 0.0022544  |   0.3 |  0.04
+Modify  | 4.681      | 4.6884     | 4.7051     |   0.4 | 92.18
+Other   |            | 0.07891    |            |       |  1.55
+
+Nlocal:           5071 ave        5197 max        4951 min
+Histogram: 1 0 0 1 0 0 1 0 0 1
+Nghost:             44 ave          50 max          41 min
+Histogram: 2 0 0 1 0 0 0 0 0 1
+Neighs:           5.75 ave          12 max           0 min
+Histogram: 1 0 0 1 0 1 0 0 0 1
+
+Total # of neighbors = 23
+Ave neighs/atom = 0.0011338986
+Neighbor list builds = 1500
+Dangerous builds = 0
+Total wall time: 0:00:07

examples/ASPHERE/line/log.21Sep24.line.srd.g++.8

@@ -0,0 +1,323 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-296-g061c80104a-modified)
+  using 1 OpenMP thread(s) per MPI task
+# Aspherical shear demo - 2d line boxes, solvated by SRD particles
+
+units           lj
+atom_style      line
+atom_modify     first big
+dimension       2
+
+read_data       data.line.srd
+Reading data file ...
+  orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  4 by 2 by 1 MPI processor grid
+  reading atoms ...
+  400 atoms
+  400 lines
+  read_data CPU = 0.002 seconds
+
+# add small particles as hi density lattice
+
+lattice         sq 0.4
+Lattice spacing in x,y,z = 1.5811388 1.5811388 1.5811388
+region          plane block INF INF INF INF -0.001 0.001
+lattice         sq 10.0
+Lattice spacing in x,y,z = 0.31622777 0.31622777 0.31622777
+create_atoms    2 region plane
+Created 33489 atoms
+  using lattice units in orthogonal box = (-28.796757 -28.796757 -0.5) to (28.796757 28.796757 0.5)
+  create_atoms CPU = 0.001 seconds
+
+group           big type 1
+400 atoms in group big
+group           small type 2
+33489 atoms in group small
+set             group small mass 0.01
+Setting atom values ...
+  33489 settings made for mass
+
+# delete overlaps
+# must set 1-2 cutoff to non-zero value
+
+pair_style      lj/cut  1.5
+pair_coeff      1 1 1.0 1.0
+pair_coeff      2 2 0.0 1.0 0.0
+pair_coeff      1 2 1.0 1.0
+
+delete_atoms    overlap 1.5 small big
+System init for delete_atoms ...
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 1.8
+  ghost atom cutoff = 1.8
+  binsize = 0.9, bins = 64 64 2
+  2 neighbor lists, perpetual/occasional/extra = 1 1 0
+  (1) command delete_atoms, occasional
+      attributes: full, newton on
+      pair build: full/bin/atomonly
+      stencil: full/bin/2d
+      bin: standard
+  (2) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/2d
+      bin: standard
+WARNING: Delete_atoms cutoff > minimum neighbor cutoff (src/delete_atoms.cpp:313)
+Deleted 13605 atoms, new total = 20284
+
+# SRD run
+
+reset_timestep  0
+
+velocity        small create 1.44 87287 loop geom
+
+neighbor        0.8 multi
+neigh_modify    delay 0 every 1 check yes
+neigh_modify    exclude molecule/intra big include big
+
+comm_modify     mode multi group big vel yes
+neigh_modify    include big
+
+# no pairwise interactions with small particles
+
+pair_style      line/lj 2.5
+pair_coeff      1 1 1.0  1.0 1.0 1.0 2.5
+pair_coeff      2 2 0.0  0.0 0.0 1.0 0.0
+pair_coeff      1 2 1.0  0.2 1.0 0.2 2.5
+
+# use fix SRD to push small particles out from inside big ones
+# if comment out, big particles won't see SRD particles
+
+timestep        0.0002
+
+fix             1 big rigid molecule
+  100 rigid bodies with 400 atoms
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip
+
+fix             3 all deform 1 x scale 1.25 y scale 1.25
+fix             4 all enforce2d
+
+# diagnostics
+
+compute         tsmall small temp/deform
+compute         tbig big temp
+variable        pebig equal pe*atoms/count(big)
+variable        ebig equal etotal*atoms/count(big)
+
+compute_modify  tbig extra/dof -350
+
+compute         1 big erotate/asphere
+compute         2 all ke
+compute         3 all pe
+variable        toteng equal (c_1+c_2+c_3)/atoms
+
+thermo          1000
+thermo_style    custom step c_tsmall f_2[9] c_1 etotal                 v_pebig v_ebig press
+thermo_modify   temp tbig
+WARNING: Temperature for thermo pressure is not for group all (src/thermo.cpp:533)
+
+compute         10 big property/atom end1x end1y end2x end2y
+
+#dump           1 all custom 500 dump1.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump1.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+run             10000
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Your simulation uses code contributions which should be cited:
+
+- Type Label Framework: https://doi.org/10.1021/acs.jpcb.3c08419
+
+@Article{Gissinger24,
+ author = {Jacob R. Gissinger, Ilia Nikiforov, Yaser Afshar, Brendon Waters, Moon-ki Choi, Daniel S. Karls, Alexander Stukowski, Wonpil Im, Hendrik Heinz, Axel Kohlmeyer, and Ellad B. Tadmor},
+ title = {Type Label Framework for Bonded Force Fields in LAMMPS},
+ journal = {J. Phys. Chem. B},
+ year =    2024,
+ volume =  128,
+ number =  13,
+ pages =   {3282–-3297}
+}
+
+- neighbor multi command: doi:10.1016/j.cpc.2008.03.005, doi:10.1007/s40571-020-00361-2
+
+@Article{Intveld08,
+ author =  {in 't Veld, P. J. and S. J.~Plimpton and G. S. Grest},
+ title =   {Accurate and Efficient Methods for Modeling Colloidal
+            Mixtures in an Explicit Solvent using Molecular Dynamics},
+ journal = {Comput.\ Phys.\ Commut.},
+ year =    2008,
+ volume =  179,
+ pages =   {320--329}
+}
+
+@article{Shire2020,
+ author = {Shire, Tom and Hanley, Kevin J. and Stratford, Kevin},
+ title = {{DEM} Simulations of Polydisperse Media: Efficient Contact
+          Detection Applied to Investigate the Quasi-Static Limit},
+ journal = {Computational Particle Mechanics},
+ year = {2020}
+@article{Monti2022,
+ author = {Monti, Joseph M. and Clemmer, Joel T. and Srivastava, 
+           Ishan and Silbert, Leonardo E. and Grest, Gary S. 
+           and Lechman, Jeremy B.},
+ title = {Large-scale frictionless jamming with power-law particle 
+          size distributions},
+ journal = {Phys. Rev. E},
+ volume = {106}
+ issue = {3}
+ year = {2022}
+}
+
+- fix srd command: doi:10.1063/1.3419070
+
+@Article{Petersen10,
+ author = {M. K. Petersen and J. B. Lechman and S. J. Plimpton and
+ G. S. Grest and in 't Veld, P. J. and P. R. Schunk},
+ title =   {Mesoscale Hydrodynamics via Stochastic Rotation
+    Dynamics: Comparison with {L}ennard-{J}ones Fluid},
+ journal = {J.~Chem.\ Phys.},
+ year =    2010,
+ volume =  132,
+ pages =   174106
+}
+
+CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
+
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using compute temp/deform with inconsistent fix deform remap option (src/compute_temp_deform.cpp:71)
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.04
+  SRD max distance & max velocity = 0.16 40
+  SRD grid counts: 230 230 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.25040659 0.25040659 1
+  SRD per actual grid cell = 0.45503978
+  SRD viscosity = 1.0732692
+  big/SRD mass density ratio = 12.298053
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 0
+  ave/max small velocity = 15.906001 29.105426
+  ave/max big velocity = 0 0
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 35 35 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton
+      stencil: half/multi/2d
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 9.752 | 9.762 | 9.781 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+         0   1.4809886      0              0              0              0              0              0            
+      1000   1.2265081      1.1522909      0.00027866069  0.0022427232   0              0.1137285      0.58262976   
+      2000   1.1757141      1.1251323      0.00040597152  0.003287261    0              0.16669701     0.27109853   
+      3000   1.126304       1.0646585      0.00050542958  0.0040748897   0              0.20663766     0.36959653   
+      4000   1.1140297      1.0761435      0.00060430927  0.0052863645   0              0.26807154     0.35099205   
+      5000   1.0823307      1.0256814      0.00066557681  0.0051213284   0              0.25970256     0.28491631   
+      6000   1.0656188      1.0387091      0.00066454105  0.005213537    0              0.26437846     0.2150581    
+      7000   1.0600108      1.0099931      0.0007238719   0.005470459    0              0.27740698     0.21846568   
+      8000   1.0464374      1.0049819      0.00061068366  0.0053702582   0              0.27232579     0.1326171    
+      9000   1.0381097      1.0126049      0.00057318728  0.0055976321   0              0.28385592     0.16797796   
+     10000   1.0312982      1.0018962      0.00050597007  0.0051153126   0              0.2593975      0.15281043   
+Loop time of 1.2997 on 8 procs for 10000 steps with 20284 atoms
+
+Performance: 132953.355 tau/day, 7694.060 timesteps/s, 156.066 Matom-step/s
+96.7% CPU use with 8 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.01406    | 0.014495   | 0.01532    |   0.3 |  1.12
+Neigh   | 0.0051916  | 0.0053424  | 0.0054386  |   0.1 |  0.41
+Comm    | 0.062053   | 0.065305   | 0.069014   |   0.9 |  5.02
+Output  | 0.00062289 | 0.00065254 | 0.00077243 |   0.0 |  0.05
+Modify  | 1.169      | 1.178      | 1.1905     |   0.6 | 90.64
+Other   |            | 0.03589    |            |       |  2.76
+
+Nlocal:         2535.5 ave        2637 max        2476 min
+Histogram: 1 3 1 1 0 0 0 0 1 1
+Nghost:         31.125 ave          42 max          25 min
+Histogram: 3 0 1 2 0 0 0 0 1 1
+Neighs:              0 ave           0 max           0 min
+Histogram: 8 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 0
+Ave neighs/atom = 0
+Neighbor list builds = 500
+Dangerous builds = 0
+
+#undump          1
+#undump          2
+unfix           3
+
+change_box      all triclinic
+Changing box ...
+  triclinic box = (-35.995947 -35.995947 -0.5) to (35.995947 35.995947 0.5) with tilt (0 0 0)
+
+fix             2 small srd 20 big 1.0 0.25 49894                 search 0.2 cubic warn 0.0001 shift yes 49829                 overlap yes collision noslip #tstat yes
+
+#dump           1 all custom 500 dump2.atom.srd id type x y z ix iy iz
+#dump           2 all custom 500 dump2.line.srd id type #               c_10[1] c_10[2] c_10[3] c_10[4]
+
+fix             3 all deform 1 xy erate 0.0002 units box remap v
+
+run             40000
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+WARNING: Using fix srd with box deformation but no SRD thermostat (src/SRD/fix_srd.cpp:405)
+SRD info:
+  SRD/big particles = 19884 400
+  big particle diameter max/min = 1.9988171 0.50330564
+  SRD temperature & lamda = 1 0.04
+  SRD max distance & max velocity = 0.16 40
+  SRD grid counts: 288 288 1
+  SRD grid size: request, actual (xyz) = 0.25, 0.24997185 0.24997185 1
+  SRD per actual grid cell = 0.26976265
+  SRD viscosity = 1.5799049
+  big/SRD mass density ratio = 20.672578
+WARNING: SRD bin size for fix srd differs from user request (src/SRD/fix_srd.cpp:2809)
+WARNING: Fix srd grid size > 1/4 of big particle diameter (src/SRD/fix_srd.cpp:2830)
+  # of rescaled SRD velocities = 1
+  ave/max small velocity = 12.481632 40
+  ave/max big velocity = 0.58815233 1.5489134
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 3.3
+  ghost atom cutoff = 3.3
+  binsize = 1.65, bins = 44 44 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair line/lj, perpetual
+      attributes: half, newton on
+      pair build: half/multi/atomonly/newton/tri
+      stencil: half/multi/2d/tri
+      bin: multi
+Per MPI rank memory allocation (min/avg/max) = 11.95 | 11.95 | 11.95 Mbytes
+   Step        c_tsmall        f_2[9]          c_1           TotEng        v_pebig         v_ebig         Press     
+     10000   0.98889306     0              0.00050595647  0.0051153289   0              0.25939833     0.020019811  
+     11000   0.98995417     0.99721412     0.0005166709   0.0054320552   0              0.27545952     0.03376188   
+     12000   0.99401168     0.95502319     0.00053659416  0.0054974426   0              0.27877531     0.056856034  
+     13000   0.98790692     0.96488357     0.00061076364  0.0053196554   0              0.26975972     0.2677228    
+     14000   0.98939233     0.98809694     0.00052317779  0.0050536922   0              0.25627273     0.20381427   
+     15000   0.99046365     0.97048828     0.00056914402  0.0053274908   0              0.27015706     0.082163088  
+     16000   0.99743447     1.0111983      0.00053706165  0.0048068384   0              0.24375478     0.06429569   
+     17000   0.99622359     1.0124982      0.0005905197   0.0055966416  -7.4541903e-05  0.2838057      0.25418037   
+     18000   0.99283929     0.99581721     0.00061730878  0.0059098964  -0.00014754106  0.29969084     0.12701702   
+     19000   0.99175517     1.0115581      0.0005128714   0.0059214809  -0.00046203987  0.30027829     0.2640506    
+     20000   0.98895811     0.97731528     0.00052294314  0.0055670021  -0.0016029702   0.28230268     0.26239209   
+     21000   0.98904093     0.98120258     0.00054169978  0.0052762731  -0.0026834879   0.26755981     0.15432481   
+     22000   0.98773149     0.99217218     0.00056091602  0.0052754729  -0.0044093345   0.26751923     0.2208051    
+     23000   0.98802862     0.96564549     0.00062226338  0.0053263454  -0.008790914    0.27009898     0.18058977   
+     24000   0.98754002     1.024015       0.00057178953  0.0054083382  -0.0094859317   0.27425683     0.01598147   
+     25000   0.98656341     0.94647246     0.00066215791  0.0053555849  -0.014877058    0.27158171     0.28794628   
+     26000   0.98750471     1.0077626      0.00058955769  0.0057081639  -0.013151869    0.28946099     0.20814911   
+     27

examples/COUPLE/plugin/liblammpsplugin.c

@@ -81,6 +81,7 @@ liblammpsplugin_t *liblammpsplugin_load(const char *lib)
   ADDSYM(python_finalize);
 
   ADDSYM(error);
+  ADDSYM(expand);
 
   ADDSYM(file);
   ADDSYM(command);

examples/COUPLE/plugin/liblammpsplugin.h

@@ -126,6 +126,7 @@ struct _liblammpsplugin {
   void (*python_finalize)();
 
   void (*error)(void *, int, const char *);
+  char *(*expand)(void *, const char *);
 
   void (*file)(void *, const char *);
   char *(*command)(void *, const char *);

examples/meam/msmeam/log.27Sep2024.msmeam.g++.1

@@ -1,4 +1,4 @@
-LAMMPS (7 Feb 2024 - Development - patch_7Feb2024_update1-182-g93942f2013-modified)
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-372-g51d104975a)
 OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
   using 1 OpenMP thread(s) per MPI task
 # Test of MEAM potential for HGa
@@ -67,7 +67,7 @@ Created 1 atoms
 variable        teng equal "c_eatoms"
 compute pot_energy all pe/atom
 compute stress all stress/atom NULL
-# dump 1 all custom 1 dump.msmeam id x y z fx fy fz c_pot_energy c_stress[1] c_stress[2] c_stress[3] c_stress[4] c_stress[5] c_stress[6]
+dump 1 all custom 1 dump.msmeam id x y z fx fy fz c_pot_energy c_stress[1] c_stress[2] c_stress[3] c_stress[4] c_stress[5] c_stress[6]
 run     1
 WARNING: No fixes with time integration, atoms won't move (src/verlet.cpp:60)
 Neighbor list info ...
@@ -89,22 +89,22 @@ Neighbor list info ...
       bin: none
 Per MPI rank memory allocation (min/avg/max) = 8.587 | 8.587 | 8.587 Mbytes
    Step          Temp          TotEng         Press           Pxx            Pyy            Pzz            Pxy            Pxz            Pyz             Lx             Ly             Lz           Volume        c_eatoms   
-         0   0              15.433079      491354.7       838670.96      635393.15      0              80195.797      0              0              8              8              8              512            15.433079    
+         0   0              15.438614      491542.52      839006.02      635621.55      0              80225.587      0              0              8              8              8              512            15.438614    
-         1   0              15.433079      491354.7       838670.96      635393.15      0              80195.797      0              0              8              8              8              512            15.433079    
+         1   0              15.438614      491542.52      839006.02      635621.55      0              80225.587      0              0              8              8              8              512            15.438614    
-Loop time of 4.4446e-05 on 1 procs for 1 steps with 3 atoms
+Loop time of 0.000144827 on 1 procs for 1 steps with 3 atoms
 
-Performance: 1943.932 ns/day, 0.012 hours/ns, 22499.213 timesteps/s, 67.498 katom-step/s
+Performance: 596.574 ns/day, 0.040 hours/ns, 6904.790 timesteps/s, 20.714 katom-step/s
-31.5% CPU use with 1 MPI tasks x 1 OpenMP threads
+21.4% CPU use with 1 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 2.9908e-05 | 2.9908e-05 | 2.9908e-05 |   0.0 | 67.29
+Pair    | 9.2136e-05 | 9.2136e-05 | 9.2136e-05 |   0.0 | 63.62
 Neigh   | 0          | 0          | 0          |   0.0 |  0.00
-Comm    | 1.033e-06  | 1.033e-06  | 1.033e-06  |   0.0 |  2.32
+Comm    | 4.389e-06  | 4.389e-06  | 4.389e-06  |   0.0 |  3.03
-Output  | 9.347e-06  | 9.347e-06  | 9.347e-06  |   0.0 | 21.03
+Output  | 3.9556e-05 | 3.9556e-05 | 3.9556e-05 |   0.0 | 27.31
-Modify  | 2.02e-07   | 2.02e-07   | 2.02e-07   |   0.0 |  0.45
+Modify  | 9.92e-07   | 9.92e-07   | 9.92e-07   |   0.0 |  0.68
-Other   |            | 3.956e-06  |            |       |  8.90
+Other   |            | 7.754e-06  |            |       |  5.35
 
 Nlocal:              3 ave           3 max           3 min
 Histogram: 1 0 0 0 0 0 0 0 0 0

examples/meam/msmeam/log.27Sep2024.msmeam.g++.4

@@ -1,4 +1,4 @@
-LAMMPS (7 Feb 2024 - Development - patch_7Feb2024_update1-182-g93942f2013-modified)
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-372-g51d104975a)
 OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
   using 1 OpenMP thread(s) per MPI task
 # Test of MEAM potential for HGa
@@ -67,7 +67,7 @@ Created 1 atoms
 variable        teng equal "c_eatoms"
 compute pot_energy all pe/atom
 compute stress all stress/atom NULL
-# dump 1 all custom 1 dump.msmeam id x y z fx fy fz c_pot_energy c_stress[1] c_stress[2] c_stress[3] c_stress[4] c_stress[5] c_stress[6]
+dump 1 all custom 1 dump.msmeam id x y z fx fy fz c_pot_energy c_stress[1] c_stress[2] c_stress[3] c_stress[4] c_stress[5] c_stress[6]
 run     1
 WARNING: No fixes with time integration, atoms won't move (src/verlet.cpp:60)
 Neighbor list info ...
@@ -89,22 +89,22 @@ Neighbor list info ...
       bin: none
 Per MPI rank memory allocation (min/avg/max) = 7.965 | 8.123 | 8.594 Mbytes
    Step          Temp          TotEng         Press           Pxx            Pyy            Pzz            Pxy            Pxz            Pyz             Lx             Ly             Lz           Volume        c_eatoms   
-         0   0              15.433079      491354.7       838670.96      635393.15      0              80195.797      0              0              8              8              8              512            15.433079    
+         0   0              15.438614      491542.52      839006.02      635621.55      0              80225.587      0              0              8              8              8              512            15.438614    
-         1   0              15.433079      491354.7       838670.96      635393.15      0              80195.797      0              0              8              8              8              512            15.433079    
+         1   0              15.438614      491542.52      839006.02      635621.55      0              80225.587      0              0              8              8              8              512            15.438614    
-Loop time of 8.70645e-05 on 4 procs for 1 steps with 3 atoms
+Loop time of 0.000328503 on 4 procs for 1 steps with 3 atoms
 
-Performance: 992.368 ns/day, 0.024 hours/ns, 11485.738 timesteps/s, 34.457 katom-step/s
+Performance: 263.011 ns/day, 0.091 hours/ns, 3044.110 timesteps/s, 9.132 katom-step/s
-29.0% CPU use with 4 MPI tasks x 1 OpenMP threads
+75.3% CPU use with 4 MPI tasks x 1 OpenMP threads
 
 MPI task timing breakdown:
 Section |  min time  |  avg time  |  max time  |%varavg| %total
 ---------------------------------------------------------------
-Pair    | 4.3957e-05 | 4.67e-05   | 5.1056e-05 |   0.0 | 53.64
+Pair    | 0.0001419  | 0.0001471  | 0.00015891 |   0.0 | 44.78
 Neigh   | 0          | 0          | 0          |   0.0 |  0.00
-Comm    | 1.105e-05  | 1.3822e-05 | 1.7033e-05 |   0.0 | 15.88
+Comm    | 2.2092e-05 | 2.8424e-05 | 3.667e-05  |   0.0 |  8.65
-Output  | 1.5765e-05 | 1.9045e-05 | 2.5216e-05 |   0.0 | 21.87
+Output  | 8.6275e-05 | 0.00010558 | 0.0001422  |   0.0 | 32.14
-Modify  | 2.58e-07   | 3.465e-07  | 3.81e-07   |   0.0 |  0.40
+Modify  | 1.093e-06  | 2.4148e-06 | 5.651e-06  |   0.0 |  0.74
-Other   |            | 7.151e-06  |            |       |  8.21
+Other   |            | 4.498e-05  |            |       | 13.69
 
 Nlocal:           0.75 ave           3 max           0 min
 Histogram: 3 0 0 0 0 0 0 0 0 1

examples/nemd/in.nemd

@@ -15,8 +15,8 @@ velocity	all create 1.44 87287 loop geom
 region          slice block 4 6 INF INF INF INF
 set             region slice type 2
 
-pair_style	lj/cut 2.5
+pair_style      lj/cut 4.0
-pair_coeff	* * 1.0 1.0 1.0
+pair_coeff      * * 1.0 1.0
 
 neighbor        0.3 bin
 neigh_modify    delay 0 every 1
@@ -24,14 +24,14 @@ neigh_modify	delay 0 every 1
 fix             1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
 fix             2 all deform 1 xy erate 0.01 remap v
 
-#dump		1 all custom 5000 dump.nemd id type x y z
+#dump           1 all custom 500 dump.nemd id type x y z
 
-#dump		2 all image 1000 image.*.jpg type type adiam 1.2
+#dump           2 all image 100 image.*.jpg type type adiam 1.2
 #dump_modify    2 pad 5
 
-#dump		3 all movie 1000 movie.mpg type type adiam 1.2
+#dump           3 all movie 100 movie.mpg type type adiam 1.2
 #dump_modify    3 pad 5
 
-thermo		1000
+thermo          50
 
-run		50000
+run             1000

examples/nemd/log.27Nov18.nemd.g++.1

@@ -1,137 +0,0 @@
-LAMMPS (27 Nov 2018)
-  using 1 OpenMP thread(s) per MPI task
-# 2d NEMD simulation
-
-units		lj
-atom_style	atomic
-dimension	2
-
-lattice		sq2 0.8442
-Lattice spacing in x,y,z = 1.53919 1.53919 1.53919
-region		box prism 0 10 0 8 -0.5 0.5 0 0 0
-create_box	2 box
-Created triclinic box = (0 0 -0.769595) to (15.3919 12.3135 0.769595) with tilt (0 0 0)
-  1 by 1 by 1 MPI processor grid
-create_atoms	1 box
-Created 160 atoms
-  Time spent = 0.000332355 secs
-mass		* 1.0
-
-velocity	all create 1.44 87287 loop geom
-
-region		slice block 4 6 INF INF INF INF
-set		region slice type 2
-  40 settings made for type
-
-pair_style	lj/cut 2.5
-pair_coeff	* * 1.0 1.0 1.0
-
-neighbor	0.3 bin
-neigh_modify	delay 0 every 1
-
-fix		1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
-fix		2 all deform 1 xy erate 0.01 remap v
-
-#dump		1 all custom 5000 dump.nemd id type x y z
-
-#dump		2 all image 1000 image.*.jpg type type adiam 1.2
-#dump_modify	2 pad 5
-
-#dump		3 all movie 1000 movie.mpg type type adiam 1.2
-#dump_modify	3 pad 5
-
-thermo		1000
-
-run		50000
-Neighbor list info ...
-  update every 1 steps, delay 0 steps, check yes
-  max neighbors/atom: 2000, page size: 100000
-  master list distance cutoff = 1.3
-  ghost atom cutoff = 1.3
-  binsize = 0.65, bins = 24 19 3
-  1 neighbor lists, perpetual/occasional/extra = 1 0 0
-  (1) pair lj/cut, perpetual
-      attributes: half, newton on
-      pair build: half/bin/newton/tri
-      stencil: half/bin/2d/newton/tri
-      bin: standard
-Per MPI rank memory allocation (min/avg/max) = 3.065 | 3.065 | 3.065 Mbytes
-Step Temp E_pair E_mol TotEng Press Volume 
-       0         1.44            0            0        1.431    1.2080502    189.52855 
-    1000    1.1326992   0.25863754            0    1.3842573    6.0588079    189.52855 
-    2000   0.99104643   0.37634349            0    1.3611959    7.8993387    189.52855 
-    3000    1.0749743   0.21908728            0    1.2873429    6.2659517    189.52855 
-    4000    1.0986742   0.27147022            0    1.3632777    5.8778262    189.52855 
-    5000     1.071838   0.23413372            0    1.2992728    5.9120887    189.52855 
-    6000    1.0013194   0.26923671            0    1.2642979    6.2802759    189.52855 
-    7000   0.94110685    0.3224557            0    1.2576806    6.1864166    189.52855 
-    8000   0.97391513   0.28793383            0     1.255762    6.5071893    189.52855 
-    9000   0.95346063   0.31050593            0    1.2580074    6.3321512    189.52855 
-   10000   0.96236447   0.26298203            0    1.2193317    6.4083918    189.52855 
-   11000    0.9511149   0.27571527            0    1.2208857    6.0949768    189.52855 
-   12000    1.0186935   0.18134918            0    1.1936758    5.1269128    189.52855 
-   13000   0.96350682   0.23171507            0       1.1892    5.7367267    189.52855 
-   14000   0.94740402   0.27357945            0    1.2150622    6.0156532    189.52855 
-   15000   0.87951545   0.27745111            0    1.1514696     6.297405    189.52855 
-   16000   0.93216196   0.27020559            0    1.1965415    6.6188833    189.52855 
-   17000   0.94109936   0.24756193            0    1.1827794    5.8993088    189.52855 
-   18000   0.97325239   0.27996398            0    1.2471335    6.1486561    189.52855 
-   19000    1.0494686   0.27132686            0    1.3142363    6.6757065    189.52855 
-   20000    1.0391862   0.25195457            0    1.2846459     6.143235    189.52855 
-   21000   0.96407137   0.27359166            0    1.2316376    5.9577116    189.52855 
-   22000   0.97954534   0.31920255            0    1.2926257    6.5320163    189.52855 
-   23000   0.97585473   0.24154424            0    1.2112999    6.0839179    189.52855 
-   24000    1.0522109    0.1646952            0    1.2103298    5.0388687    189.52855 
-   25000   0.93707172   0.25655806            0    1.1877731     5.819887    189.52855 
-   26000   0.89798775   0.26629627            0    1.1586716    6.0393558    189.52855 
-   27000   0.93259926   0.24542428            0    1.1721948    5.3560986    189.52855 
-   28000    0.8428223   0.20784302            0    1.0453977     4.956911    189.52855 
-   29000   0.81653505   0.21924932            0     1.030681     5.271501    189.52855 
-   30000   0.90157811   0.15070734            0    1.0466506     4.476142    189.52855 
-   31000   0.86580039   0.21115151            0    1.0715407    5.0056915    189.52855 
-   32000   0.89768096   0.28377249            0    1.1758429    5.8449711    189.52855 
-   33000    1.0504011   0.29009694            0     1.333933    6.1319155    189.52855 
-   34000    1.2009765   0.19137934            0    1.3848498    4.9643885    189.52855 
-   35000     1.208705   0.27071222            0    1.4718628    6.2162389    189.52855 
-   36000    1.2211309   0.28389521            0     1.497394    6.5090715    189.52855 
-   37000    1.1384381   0.42795547            0    1.5592783    8.5129272    189.52855 
-   38000    1.2198334   0.34335732            0    1.5555668    7.2940883    189.52855 
-   39000    1.1562045   0.35783089            0    1.5068091     7.340999    189.52855 
-   40000    1.2145924   0.28410558            0    1.4911068     6.234986    189.52855 
-   41000    1.1240878   0.34663237            0    1.4636946    7.1720193    189.52855 
-   42000    1.2491422   0.26815889            0     1.509494    6.1390803    189.52855 
-   43000    1.1387564   0.33755832            0    1.4691975    7.0577597    189.52855 
-   44000    1.0031598    0.4081807            0    1.4050708    8.2732113    189.52855 
-   45000    1.0166213   0.29131017            0    1.3015776    6.1907807    189.52855 
-   46000   0.96251302   0.31483519            0    1.2713325    6.6987235    189.52855 
-   47000   0.89809294   0.30909884            0    1.2015787    6.3997583    189.52855 
-   48000   0.86736217   0.31917648            0    1.1811176    7.1584774    189.52855 
-   49000   0.91979053   0.21099403            0    1.1250359    5.4968259    189.52855 
-   50000   0.87079959   0.24059333            0    1.1059504    5.6039305    189.52855 
-Loop time of 1.54353 on 1 procs for 50000 steps with 160 atoms
-
-Performance: 13993916.675 tau/day, 32393.326 timesteps/s
-99.7% CPU use with 1 MPI tasks x 1 OpenMP threads
-
-MPI task timing breakdown:
-Section |  min time  |  avg time  |  max time  |%varavg| %total
----------------------------------------------------------------
-Pair    | 0.20172    | 0.20172    | 0.20172    |   0.0 | 13.07
-Neigh   | 0.16634    | 0.16634    | 0.16634    |   0.0 | 10.78
-Comm    | 0.068928   | 0.068928   | 0.068928   |   0.0 |  4.47
-Output  | 0.00059891 | 0.00059891 | 0.00059891 |   0.0 |  0.04
-Modify  | 1.0123     | 1.0123     | 1.0123     |   0.0 | 65.59
-Other   |            | 0.09361    |            |       |  6.06
-
-Nlocal:    160 ave 160 max 160 min
-Histogram: 1 0 0 0 0 0 0 0 0 0
-Nghost:    73 ave 73 max 73 min
-Histogram: 1 0 0 0 0 0 0 0 0 0
-Neighs:    353 ave 353 max 353 min
-Histogram: 1 0 0 0 0 0 0 0 0 0
-
-Total # of neighbors = 353
-Ave neighs/atom = 2.20625
-Neighbor list builds = 5273
-Dangerous builds = 0
-Total wall time: 0:00:01

examples/nemd/log.27Nov18.nemd.g++.4

@@ -1,137 +0,0 @@
-LAMMPS (27 Nov 2018)
-  using 1 OpenMP thread(s) per MPI task
-# 2d NEMD simulation
-
-units		lj
-atom_style	atomic
-dimension	2
-
-lattice		sq2 0.8442
-Lattice spacing in x,y,z = 1.53919 1.53919 1.53919
-region		box prism 0 10 0 8 -0.5 0.5 0 0 0
-create_box	2 box
-Created triclinic box = (0 0 -0.769595) to (15.3919 12.3135 0.769595) with tilt (0 0 0)
-  2 by 2 by 1 MPI processor grid
-create_atoms	1 box
-Created 160 atoms
-  Time spent = 0.000308275 secs
-mass		* 1.0
-
-velocity	all create 1.44 87287 loop geom
-
-region		slice block 4 6 INF INF INF INF
-set		region slice type 2
-  40 settings made for type
-
-pair_style	lj/cut 2.5
-pair_coeff	* * 1.0 1.0 1.0
-
-neighbor	0.3 bin
-neigh_modify	delay 0 every 1
-
-fix		1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
-fix		2 all deform 1 xy erate 0.01 remap v
-
-#dump		1 all custom 5000 dump.nemd id type x y z
-
-#dump		2 all image 1000 image.*.jpg type type adiam 1.2
-#dump_modify	2 pad 5
-
-#dump		3 all movie 1000 movie.mpg type type adiam 1.2
-#dump_modify	3 pad 5
-
-thermo		1000
-
-run		50000
-Neighbor list info ...
-  update every 1 steps, delay 0 steps, check yes
-  max neighbors/atom: 2000, page size: 100000
-  master list distance cutoff = 1.3
-  ghost atom cutoff = 1.3
-  binsize = 0.65, bins = 24 19 3
-  1 neighbor lists, perpetual/occasional/extra = 1 0 0
-  (1) pair lj/cut, perpetual
-      attributes: half, newton on
-      pair build: half/bin/newton/tri
-      stencil: half/bin/2d/newton/tri
-      bin: standard
-Per MPI rank memory allocation (min/avg/max) = 3.062 | 3.062 | 3.062 Mbytes
-Step Temp E_pair E_mol TotEng Press Volume 
-       0         1.44            0            0        1.431    1.2080502    189.52855 
-    1000    1.1682693   0.24486562            0    1.4058332    5.8092954    189.52855 
-    2000    1.0928734   0.27609364            0    1.3621366    6.2237017    189.52855 
-    3000      1.09088   0.24816112            0    1.3322231    5.7001547    189.52855 
-    4000    1.0110684   0.29868377            0     1.303433    7.3312319    189.52855 
-    5000   0.91033678   0.28330698            0    1.1879542    6.1840352    189.52855 
-    6000   0.93416074   0.22661127            0    1.1549335    5.3619735    189.52855 
-    7000   0.93305734   0.19203739            0    1.1192631    5.2497547    189.52855 
-    8000   0.88944438   0.19421381            0    1.0780992    4.9733446    189.52855 
-    9000   0.86949257   0.21207681            0    1.0761351    5.4687076    189.52855 
-   10000   0.80088203   0.24071142            0    1.0365879     5.334545    189.52855 
-   11000   0.88899727   0.19972767            0    1.0831687    4.8832207    189.52855 
-   12000   0.93045817   0.17883252            0    1.1034753    4.9081709    189.52855 
-   13000    0.9724196   0.19089684            0    1.1572388    5.3460903    189.52855 
-   14000   0.93902186   0.25513773            0    1.1882907    6.3338337    189.52855 
-   15000   0.91879903   0.31605547            0     1.229112    6.2085671    189.52855 
-   16000    0.9860058   0.26863362            0    1.2484769     6.514688    189.52855 
-   17000    1.0354756   0.23445357            0    1.2634574    6.1519296    189.52855 
-   18000    1.0244774   0.27511827            0    1.2931927    6.2230002    189.52855 
-   19000    1.1581216   0.21558936            0    1.3664727    5.5458237    189.52855 
-   20000    1.0552168   0.29344488            0    1.3420666    6.4880315    189.52855 
-   21000   0.97925435   0.31583414            0    1.2889681    6.7584093    189.52855 
-   22000    1.0112494   0.26246834            0    1.2673974    5.4112008    189.52855 
-   23000    1.0463332   0.26049752            0    1.3002911    6.1359606    189.52855 
-   24000    1.1130319   0.19848564            0    1.3045611    5.7088487    189.52855 
-   25000    1.0355662   0.28048951            0    1.3095834    6.4596476    189.52855 
-   26000    1.0823932   0.21784218            0    1.2934704     5.106334    189.52855 
-   27000   0.99719525   0.32679678            0    1.3177596    6.7399277    189.52855 
-   28000    1.0665868   0.25002709            0    1.3099477    6.2732557    189.52855 
-   29000    1.0312798   0.30650087            0    1.3313351    7.0581024    189.52855 
-   30000    1.0388277   0.29812912            0    1.3304641    6.2533028    189.52855 
-   31000    1.0461658   0.21344416            0    1.2530714    5.3631154    189.52855 
-   32000    1.0233681   0.27545017            0    1.2924222    5.9612896    189.52855 
-   33000    1.1353086   0.20278244            0    1.3309953    5.7619128    189.52855 
-   34000    1.0374791   0.29661216            0     1.327607    6.5124409    189.52855 
-   35000    1.0752783   0.21684443            0    1.2854022    5.4759171    189.52855 
-   36000    1.0383445   0.27068641            0    1.3025412    6.8367218    189.52855 
-   37000   0.97341144   0.24034988            0    1.2076775    6.1335996    189.52855 
-   38000    0.9285918    0.2737544            0    1.1965425    5.8750327    189.52855 
-   39000   0.84869423   0.30079207            0     1.144182    6.8909326    189.52855 
-   40000   0.88237131   0.26049171            0    1.1373482    6.3932981    189.52855 
-   41000   0.90368591   0.21064132            0    1.1086792    5.5627232    189.52855 
-   42000   0.93436749   0.20367569            0    1.1322034    5.1420052    189.52855 
-   43000   0.91378588   0.26155533            0      1.16963     6.366756    189.52855 
-   44000   0.91673608   0.25967314            0    1.1706796    6.0846334    189.52855 
-   45000    1.0233334   0.25463562            0    1.2715732    6.0924255    189.52855 
-   46000   0.96184729   0.35422095            0    1.3100567    7.0249175    189.52855 
-   47000     1.134079   0.26196034            0    1.3889514    6.3476756    189.52855 
-   48000    1.0552136     0.303812            0    1.3524305    6.6968927    189.52855 
-   49000    1.1282184    0.2100955            0    1.3312626    5.8658659    189.52855 
-   50000    1.0493816   0.31540438            0    1.3582274    6.6348173    189.52855 
-Loop time of 1.09903 on 4 procs for 50000 steps with 160 atoms
-
-Performance: 19653623.953 tau/day, 45494.500 timesteps/s
-94.4% CPU use with 4 MPI tasks x 1 OpenMP threads
-
-MPI task timing breakdown:
-Section |  min time  |  avg time  |  max time  |%varavg| %total
----------------------------------------------------------------
-Pair    | 0.057854   | 0.05974    | 0.062726   |   0.7 |  5.44
-Neigh   | 0.047791   | 0.049863   | 0.054819   |   1.3 |  4.54
-Comm    | 0.3581     | 0.38553    | 0.39784    |   2.6 | 35.08
-Output  | 0.001116   | 0.0014414  | 0.0023859  |   1.4 |  0.13
-Modify  | 0.41102    | 0.42642    | 0.4493     |   2.3 | 38.80
-Other   |            | 0.176      |            |       | 16.02
-
-Nlocal:    40 ave 42 max 39 min
-Histogram: 2 0 0 1 0 0 0 0 0 1
-Nghost:    36.5 ave 37 max 36 min
-Histogram: 2 0 0 0 0 0 0 0 0 2
-Neighs:    87.5 ave 94 max 81 min
-Histogram: 1 1 0 0 0 0 0 0 1 1
-
-Total # of neighbors = 350
-Ave neighs/atom = 2.1875
-Neighbor list builds = 5276
-Dangerous builds = 0
-Total wall time: 0:00:01

examples/nemd/log.30Sep24.g++.1

@@ -0,0 +1,111 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-394-g75f86a68a7-modified)
+OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
+  using 1 OpenMP thread(s) per MPI task
+# 2d NEMD simulation
+
+units           lj
+atom_style      atomic
+dimension       2
+
+lattice         sq2 0.8442
+Lattice spacing in x,y,z = 1.5391903 1.5391903 1.5391903
+region          box prism 0 10 0 8 -0.5 0.5 0 0 0
+create_box      2 box
+Created triclinic box = (0 0 -0.76959516) to (15.391903 12.313523 0.76959516) with tilt (0 0 0)
+  1 by 1 by 1 MPI processor grid
+create_atoms    1 box
+Created 160 atoms
+  using lattice units in triclinic box = (0 0 -0.76959516) to (15.391903 12.313523 0.76959516) with tilt (0 0 0)
+  create_atoms CPU = 0.000 seconds
+mass            * 1.0
+
+velocity        all create 1.44 87287 loop geom
+
+region          slice block 4 6 INF INF INF INF
+set             region slice type 2
+Setting atom values ...
+  40 settings made for type
+
+pair_style      lj/cut 4.0
+pair_coeff      * * 1.0 1.0
+
+neighbor        0.3 bin
+neigh_modify    delay 0 every 1
+
+fix             1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
+fix             2 all deform 1 xy erate 0.01 remap v
+
+#dump           1 all custom 500 dump.nemd id type x y z
+
+#dump           2 all image 100 image.*.jpg type type adiam 1.2
+#dump_modify    2 pad 5
+
+#dump           3 all movie 100 movie.mpg type type adiam 1.2
+#dump_modify    3 pad 5
+
+thermo          50
+
+run             1000
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 4.3
+  ghost atom cutoff = 4.3
+  binsize = 2.15, bins = 8 6 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton/tri
+      stencil: half/bin/2d/tri
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 3.065 | 3.065 | 3.065 Mbytes
+   Step          Temp          E_pair         E_mol          TotEng         Press          Volume    
+         0   1.44          -2.6548731      0             -1.2238731      1.9354912      189.52855    
+        50   0.97904822    -2.1934929      0             -1.2205637      5.0001562      189.52855    
+       100   0.96423603    -2.1711413      0             -1.2129318      5.3040025      189.52855    
+       150   0.96430794    -2.153062       0             -1.194781       5.3988945      189.52855    
+       200   1.056585      -2.2379316      0             -1.1879503      5.0007883      189.52855    
+       250   1.0183256     -2.1921531      0             -1.1801921      5.5370076      189.52855    
+       300   0.96855552    -2.140465       0             -1.177963       5.7188412      189.52855    
+       350   1.0115567     -2.1883272      0             -1.1830927      5.4437104      189.52855    
+       400   0.93743103    -2.1230826      0             -1.1915105      5.7059909      189.52855    
+       450   1.1120368     -2.3041274      0             -1.1990408      4.646396       189.52855    
+       500   0.99611106    -2.2039016      0             -1.2140162      5.1526658      189.52855    
+       550   1.1075519     -2.3349751      0             -1.2343453      4.0671355      189.52855    
+       600   1.0550783     -2.3126484      0             -1.2641644      4.5423735      189.52855    
+       650   0.98516169    -2.2664919      0             -1.2874875      4.8365475      189.52855    
+       700   0.97899201    -2.2815136      0             -1.3086403      4.5415389      189.52855    
+       750   1.0107776     -2.3375258      0             -1.3330656      4.3655082      189.52855    
+       800   0.97711804    -2.3221241      0             -1.3511131      4.2153988      189.52855    
+       850   0.8984454     -2.258341       0             -1.3655108      4.6759265      189.52855    
+       900   0.85409237    -2.2157566      0             -1.3670024      5.0180073      189.52855    
+       950   0.90195434    -2.2500988      0             -1.3537817      4.8189466      189.52855    
+      1000   1.0047283     -2.3359434      0             -1.3374947      4.0788763      189.52855    
+Loop time of 0.0331477 on 1 procs for 1000 steps with 160 atoms
+
+Performance: 13032596.122 tau/day, 30168.047 timesteps/s, 4.827 Matom-step/s
+99.6% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.017584   | 0.017584   | 0.017584   |   0.0 | 53.05
+Neigh   | 0.0080996  | 0.0080996  | 0.0080996  |   0.0 | 24.43
+Comm    | 0.0010864  | 0.0010864  | 0.0010864  |   0.0 |  3.28
+Output  | 9.9819e-05 | 9.9819e-05 | 9.9819e-05 |   0.0 |  0.30
+Modify  | 0.0057062  | 0.0057062  | 0.0057062  |   0.0 | 17.21
+Other   |            | 0.0005715  |            |       |  1.72
+
+Nlocal:            160 ave         160 max         160 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:            271 ave         271 max         271 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:           3881 ave        3881 max        3881 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 3881
+Ave neighs/atom = 24.25625
+Neighbor list builds = 106
+Dangerous builds = 0
+Total wall time: 0:00:00

examples/nemd/log.30Sep24.g++.4

@@ -0,0 +1,111 @@
+LAMMPS (29 Aug 2024 - Development - patch_29Aug2024-394-g75f86a68a7-modified)
+OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98)
+  using 1 OpenMP thread(s) per MPI task
+# 2d NEMD simulation
+
+units           lj
+atom_style      atomic
+dimension       2
+
+lattice         sq2 0.8442
+Lattice spacing in x,y,z = 1.5391903 1.5391903 1.5391903
+region          box prism 0 10 0 8 -0.5 0.5 0 0 0
+create_box      2 box
+Created triclinic box = (0 0 -0.76959516) to (15.391903 12.313523 0.76959516) with tilt (0 0 0)
+  2 by 2 by 1 MPI processor grid
+create_atoms    1 box
+Created 160 atoms
+  using lattice units in triclinic box = (0 0 -0.76959516) to (15.391903 12.313523 0.76959516) with tilt (0 0 0)
+  create_atoms CPU = 0.001 seconds
+mass            * 1.0
+
+velocity        all create 1.44 87287 loop geom
+
+region          slice block 4 6 INF INF INF INF
+set             region slice type 2
+Setting atom values ...
+  40 settings made for type
+
+pair_style      lj/cut 4.0
+pair_coeff      * * 1.0 1.0
+
+neighbor        0.3 bin
+neigh_modify    delay 0 every 1
+
+fix             1 all nvt/sllod temp 1.0 1.0 1.0 tchain 1
+fix             2 all deform 1 xy erate 0.01 remap v
+
+#dump           1 all custom 500 dump.nemd id type x y z
+
+#dump           2 all image 100 image.*.jpg type type adiam 1.2
+#dump_modify    2 pad 5
+
+#dump           3 all movie 100 movie.mpg type type adiam 1.2
+#dump_modify    3 pad 5
+
+thermo          50
+
+run             1000
+Generated 0 of 1 mixed pair_coeff terms from geometric mixing rule
+Neighbor list info ...
+  update: every = 1 steps, delay = 0 steps, check = yes
+  max neighbors/atom: 2000, page size: 100000
+  master list distance cutoff = 4.3
+  ghost atom cutoff = 4.3
+  binsize = 2.15, bins = 8 6 1
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton/tri
+      stencil: half/bin/2d/tri
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 3.064 | 3.064 | 3.064 Mbytes
+   Step          Temp          E_pair         E_mol          TotEng         Press          Volume    
+         0   1.44          -2.6548731      0             -1.2238731      1.9354912      189.52855    
+        50   0.97904822    -2.1934929      0             -1.2205637      5.0001562      189.52855    
+       100   0.96423603    -2.1711413      0             -1.2129318      5.3040025      189.52855    
+       150   0.96430794    -2.153062       0             -1.194781       5.3988945      189.52855    
+       200   1.056585      -2.2379316      0             -1.1879503      5.0007883      189.52855    
+       250   1.0183256     -2.1921531      0             -1.1801921      5.5370076      189.52855    
+       300   0.96855552    -2.140465       0             -1.177963       5.7188412      189.52855    
+       350   1.0115567     -2.1883272      0             -1.1830927      5.4437104      189.52855    
+       400   0.93743103    -2.1230826      0             -1.1915105      5.7059909      189.52855    
+       450   1.1120368     -2.3041274      0             -1.1990408      4.646396       189.52855    
+       500   0.99611106    -2.2039016      0             -1.2140162      5.1526658      189.52855    
+       550   1.1075519     -2.3349751      0             -1.2343453      4.0671355      189.52855    
+       600   1.0550783     -2.3126484      0             -1.2641644      4.5423735      189.52855    
+       650   0.98516169    -2.2664919      0             -1.2874875      4.8365475      189.52855    
+       700   0.97899201    -2.2815136      0             -1.3086403      4.5415389      189.52855    
+       750   1.0107776     -2.3375258      0             -1.3330656      4.3655082      189.52855    
+       800   0.97711804    -2.3221241      0             -1.3511131      4.2153988      189.52855    
+       850   0.8984454     -2.258341       0             -1.3655108      4.6759265      189.52855    
+       900   0.85409237    -2.2157566      0             -1.3670024      5.0180073      189.52855    
+       950   0.90195434    -2.2500988      0             -1.3537817      4.8189466      189.52855    
+      1000   1.0047283     -2.3359434      0             -1.3374947      4.0788763      189.52855    
+Loop time of 0.0158907 on 4 procs for 1000 steps with 160 atoms
+
+Performance: 27185684.597 tau/day, 62929.825 timesteps/s, 10.069 Matom-step/s
+98.0% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.0044606  | 0.004562   | 0.0047619  |   0.2 | 28.71
+Neigh   | 0.0023154  | 0.0023979  | 0.002494   |   0.1 | 15.09
+Comm    | 0.0051743  | 0.0054807  | 0.0056638  |   0.3 | 34.49
+Output  | 7.5535e-05 | 8.9889e-05 | 0.00012988 |   0.0 |  0.57
+Modify  | 0.002223   | 0.0023624  | 0.0026372  |   0.3 | 14.87
+Other   |            | 0.0009979  |            |       |  6.28
+
+Nlocal:             40 ave          42 max          38 min
+Histogram: 1 0 1 0 0 0 0 1 0 1
+Nghost:          163.5 ave         166 max         162 min
+Histogram: 2 0 0 0 0 1 0 0 0 1
+Neighs:         970.25 ave        1016 max         942 min
+Histogram: 1 1 0 0 1 0 0 0 0 1
+
+Total # of neighbors = 3881
+Ave neighs/atom = 24.25625
+Neighbor list builds = 106
+Dangerous builds = 0
+Total wall time: 0:00:00

fortran/lammps.f90

@@ -100,6 +100,7 @@ MODULE LIBLAMMPS
   CONTAINS
     PROCEDURE :: close                  => lmp_close
     PROCEDURE :: error                  => lmp_error
+    PROCEDURE :: expand                 => lmp_expand
     PROCEDURE :: file                   => lmp_file
     PROCEDURE :: command                => lmp_command
     PROCEDURE :: commands_list          => lmp_commands_list
@@ -410,6 +411,14 @@ MODULE LIBLAMMPS
       TYPE(c_ptr), VALUE :: error_text
     END SUBROUTINE lammps_error
 
+    FUNCTION lammps_expand(handle, line) BIND(C)
+      IMPORT :: c_ptr
+      IMPLICIT NONE
+      TYPE(c_ptr), INTENT(IN), VALUE :: handle
+      TYPE(c_ptr), INTENT(IN), VALUE :: line
+      TYPE(c_ptr) :: lammps_expand
+    END FUNCTION lammps_expand
+
     SUBROUTINE lammps_file(handle, filename) BIND(C)
       IMPORT :: c_ptr
       IMPLICIT NONE
@@ -1107,10 +1116,24 @@ CONTAINS
     CALL lammps_free(str)
   END SUBROUTINE lmp_error
 
+  ! equivalent function to lammps_expand()
+  FUNCTION lmp_expand(self, line)
+    CLASS(lammps), INTENT(IN) :: self
+    CHARACTER(len=*), INTENT(IN) :: line
+    TYPE(c_ptr) :: str, res
+    CHARACTER(len=:), ALLOCATABLE :: lmp_expand
+
+    str = f2c_string(line)
+    res = lammps_expand(self%handle, str)
+    CALL lammps_free(str)
+    lmp_expand = c2f_string(res)
+    CALL lammps_free(res)
+  END FUNCTION lmp_expand
+
   ! equivalent function to lammps_file()
   SUBROUTINE lmp_file(self, filename)
     CLASS(lammps), INTENT(IN) :: self
-    CHARACTER(len=*) :: filename
+    CHARACTER(len=*), INTENT(IN) :: filename
     TYPE(c_ptr) :: str
 
     str = f2c_string(filename)
@@ -1121,7 +1144,7 @@ CONTAINS
   ! equivalent function to lammps_command()
   SUBROUTINE lmp_command(self, cmd)
     CLASS(lammps), INTENT(IN) :: self
-    CHARACTER(len=*) :: cmd
+    CHARACTER(len=*), INTENT(IN) :: cmd
     TYPE(c_ptr) :: str
 
     str = f2c_string(cmd)
@@ -1155,7 +1178,7 @@ CONTAINS
   ! equivalent function to lammps_commands_string()
   SUBROUTINE lmp_commands_string(self, str)
     CLASS(lammps), INTENT(IN) :: self
-    CHARACTER(len=*) :: str
+    CHARACTER(len=*), INTENT(IN) :: str
     TYPE(c_ptr) :: tmp
 
     tmp = f2c_string(str)
@@ -1173,7 +1196,7 @@ CONTAINS
   ! equivalent function to lammps_get_thermo
   REAL(c_double) FUNCTION lmp_get_thermo(self,name)
     CLASS(lammps), INTENT(IN) :: self
-    CHARACTER(LEN=*) :: name
+    CHARACTER(LEN=*), INTENT(IN) :: name
     TYPE(c_ptr) :: Cname
 
     Cname = f2c_string(name)
@@ -1185,7 +1208,7 @@ CONTAINS
   FUNCTION lmp_last_thermo(self,what,index) RESULT(thermo_data)
     CLASS(lammps), INTENT(IN), TARGET :: self
     CHARACTER(LEN=*), INTENT(IN) :: what
-    INTEGER :: index
+    INTEGER, INTENT(IN) :: index
     INTEGER(c_int) :: idx
     TYPE(lammps_data) :: thermo_data, type_data
     INTEGER(c_int) :: datatype

lib/gpu/lal_neighbor.cpp

@@ -586,8 +586,25 @@ void Neighbor::build_nbor_list(double **x, const int inum, const int host_inum,
     const int b2y=_block_cell_2d;
     const int g2x=static_cast<int>(ceil(static_cast<double>(_maxspecial)/b2x));
     const int g2y=static_cast<int>(ceil(static_cast<double>(nt)/b2y));
+    // the maximum number of blocks on the device is typically 65535
+    // in principle we can use a lower number to have more resource per block 32768
+    const int max_num_blocks = 65535;
+    int shift = 0;
+    if (g2y < max_num_blocks) {
       _shared->k_transpose.set_size(g2x,g2y,b2x,b2y);
-    _shared->k_transpose.run(&dev_special,&dev_special_t,&_maxspecial,&nt);
+      _shared->k_transpose.run(&dev_special,&dev_special_t,&_maxspecial,&nt,&shift);
+    } else {
+      // using a fixed number of blocks
+      int g2y_m = max_num_blocks;
+      _shared->k_transpose.set_size(g2x,g2y_m,b2x,b2y);
+      // number of chunks needed for the whole transpose
+      const int num_chunks = ceil(static_cast<double>(g2y) / g2y_m);
+      for (int i = 0; i < num_chunks; i++) {
+        _shared->k_transpose.run(&dev_special,&dev_special_t,&_maxspecial,&nt,&shift);
+        shift += g2y_m*b2y;
+      }
+    }
+
     time_transpose.stop();
   }
 

lib/gpu/lal_neighbor_gpu.cu

@@ -147,7 +147,7 @@ __kernel void kernel_calc_cell_counts(const unsigned *restrict cell_id,
 
 __kernel void transpose(__global tagint *restrict out,
                         const __global tagint *restrict in,
-                        int columns_in, int rows_in)
+                        int columns_in, int rows_in, int shift)
 {
   __local tagint block[BLOCK_CELL_2D][BLOCK_CELL_2D+1];
 
@@ -158,15 +158,15 @@ __kernel void transpose(__global tagint *restrict out,
 
   unsigned i=bi*BLOCK_CELL_2D+ti;
   unsigned j=bj*BLOCK_CELL_2D+tj;
-  if ((i<columns_in) && (j<rows_in))
+  if ((i<columns_in) && (j+shift<rows_in))
-    block[tj][ti]=in[j*columns_in+i];
+    block[tj][ti]=in[(j+shift)*columns_in+i];
 
    __syncthreads();
 
   i=bj*BLOCK_CELL_2D+ti;
   j=bi*BLOCK_CELL_2D+tj;
-  if ((i<rows_in) && (j<columns_in))
+  if ((i+shift<rows_in) && (j<columns_in))
-    out[j*rows_in+i] = block[ti][tj];
+    out[j*rows_in+i+shift] = block[ti][tj];
 }
 
 #ifndef LAL_USE_OLD_NEIGHBOR

lib/kokkos/CHANGELOG.md

@@ -1,12 +1,103 @@
 # CHANGELOG
 
+## [4.4.01](https://github.com/kokkos/kokkos/tree/4.4.01)
+[Full Changelog](https://github.com/kokkos/kokkos/compare/4.0.00...4.4.01)
+
+### Features:
+* Introduce new SequentialHostInit view allocation property [\#7229](https://github.com/kokkos/kokkos/pull/7229)
+
+### Backend and Architecture Enhancements:
+
+#### CUDA:
+* Experimental support for unified memory mode (intended for Grace-Hopper etc.) [\#6823](https://github.com/kokkos/kokkos/pull/6823)
+
+### Bug Fixes
+* OpenMP: Fix issue related to the visibility of an internal symbol with shared libraries that affected `ScatterView` in particular [\#7284](https://github.com/kokkos/kokkos/pull/7284)
+* Fix implicit copy assignment operators in few AVX2 masks being deleted [#7296](https://github.com/kokkos/kokkos/pull/7296)
+
+## [4.4.00](https://github.com/kokkos/kokkos/tree/4.4.00)
+[Full Changelog](https://github.com/kokkos/kokkos/compare/4.3.01...4.4.00)
+
+### Features:
+* Add `Kokkos::View` conversions from and to [`std::mdspan`](https://en.cppreference.com/w/cpp/container/mdspan) [\#6830](https://github.com/kokkos/kokkos/pull/6830) [\#7069](https://github.com/kokkos/kokkos/pull/7069)
+
+### Backend and Architecture Enhancements:
+
+#### CUDA:
+* `nvcc_wrapper`: Adding ability to process `--disable-warnings` flag [\#6936](https://github.com/kokkos/kokkos/issues/6936)
+* Use recommended/max team size functions in Cuda ParallelFor and Reduce constructors [\#6891](https://github.com/kokkos/kokkos/issues/6891)
+* Improve compile-times when building with `Kokkos_ENABLE_DEBUG_BOUNDS_CHECK` in Cuda [\#7013](https://github.com/kokkos/kokkos/pull/7013)
+
+#### HIP:
+* Use HIP builtin atomics [\#6882](https://github.com/kokkos/kokkos/pull/6882) [\#7000](https://github.com/kokkos/kokkos/pull/7000)
+* Enable user-specified compiler and linker flags for AMD GPUs [\#7127](https://github.com/kokkos/kokkos/pull/7127)
+
+#### SYCL:
+* Add support for Graphs [\#6912](https://github.com/kokkos/kokkos/pull/6912)
+* Fix multi-GPU support [\#6887](https://github.com/kokkos/kokkos/pull/6887)
+* Improve performance of reduction and scan operations [\#6562](https://github.com/kokkos/kokkos/pull/6562), [\#6750](https://github.com/kokkos/kokkos/pull/6750)
+* Fix lock for guarding scratch space in `TeamPolicy` `parallel_reduce` [\#6988](https://github.com/kokkos/kokkos/pull/6988)
+* Include submission command queue property information into `SYCL::print_configuration()` [\#7004](https://github.com/kokkos/kokkos/pull/7004)
+
+#### OpenACC:
+* Make `TeamPolicy` `parallel_for` execute on the correct async queue [\#7012](https://github.com/kokkos/kokkos/pull/7012)
+
+#### OpenMPTarget:
+* Honor user requested loop ordering in `MDRange` policy [\#6925](https://github.com/kokkos/kokkos/pull/6925)
+* Prevent data races by guarding the scratch space used in `parallel_scan` [\#6998](https://github.com/kokkos/kokkos/pull/6998)
+
+#### HPX:
+* Workaround issue with template argument deduction to support compilation with NVCC [\#7015](https://github.com/kokkos/kokkos/pull/7015)
+
+### General Enhancements
+* Improve performance of view copies in host parallel regions [\#6730](https://github.com/kokkos/kokkos/pull/6730)
+* Harmonize convertibility rules of `Kokkos::RandomAccessIterator` with `View`s [\#6929](https://github.com/kokkos/kokkos/pull/6929)
+* Add a check precondition non-overlapping ranges for the `adjacent_difference` algorithm in debug mode [\#6922](https://github.com/kokkos/kokkos/pull/6922)
+* Add deduction guides for `TeamPolicy` [\#7030](https://github.com/kokkos/kokkos/pull/7030)
+* SIMD: Allow flexible vector width for 32 bit types [\#6802](https://github.com/kokkos/kokkos/pull/6802)
+* Updates for `Kokkos::Array`: add `kokkos_swap(Array<T, N>)` specialization [\#6943](https://github.com/kokkos/kokkos/pull/6943), add `Kokkos::to_array` [\#6375](https://github.com/kokkos/kokkos/pull/6375),  make `Kokkos::Array` equality-comparable [\#7148](https://github.com/kokkos/kokkos/pull/7148)
+* Structured binding support for `Kokkos::complex` [\#7040](https://github.com/kokkos/kokkos/pull/7040)
+
+### Build System Changes
+* Do not require OpenMP support for languages other than CXX [\#6965](https://github.com/kokkos/kokkos/pull/6965)
+* Update Intel GPU architectures in Makefile [\#6895](https://github.com/kokkos/kokkos/pull/6895)
+* Fix use of OpenMP with Cuda or HIP as compile language [\#6972](https://github.com/kokkos/kokkos/pull/6972)
+* Define and enforce new minimum compiler versions for C++20 support [\#7128](https://github.com/kokkos/kokkos/pull/7128), [\#7123](https://github.com/kokkos/kokkos/pull/7123)
+* Add nvidia Grace CPU architecture: `Kokkos_ARCH_ARMV9_GRACE` [\#7158](https://github.com/kokkos/kokkos/pull/7158)
+* Fix Makefile.kokkos for Threads [\#6896](https://github.com/kokkos/kokkos/pull/6896)
+* Remove support for NVHPC as CUDA device compiler [\#6987](https://github.com/kokkos/kokkos/pull/6987)
+* Fix using CUDAToolkit for CMake 3.28.4 and higher [\#7062](https://github.com/kokkos/kokkos/pull/7062)
+
+### Incompatibilities (i.e. breaking changes)
+* Drop `Kokkos::Array` special treatment in `View`s [\#6906](https://github.com/kokkos/kokkos/pull/6906)
+* Drop `Experimental::RawMemoryAllocationFailure` [\#7145](https://github.com/kokkos/kokkos/pull/7145)
+
+### Deprecations
+* Remove `Experimental::LayoutTiled` class template and deprecate `is_layouttiled` trait [\#6907](https://github.com/kokkos/kokkos/pull/6907)
+* Deprecate `Kokkos::layout_iterate_type_selector` [\#7076](https://github.com/kokkos/kokkos/pull/7076)
+* Deprecate specialization of `Kokkos::pair` for a single element [\#6947](https://github.com/kokkos/kokkos/pull/6947)
+* Deprecate `deep_copy` of `UnorderedMap` of different size [\#6812](https://github.com/kokkos/kokkos/pull/6812)
+* Deprecate trailing `Proxy` template argument of `Kokkos::Array` [\#6934](https://github.com/kokkos/kokkos/pull/6934)
+* Deprecate implicit conversions of integers to `ChunkSize` [\#7151](https://github.com/kokkos/kokkos/pull/7151)
+* Deprecate implicit conversions to execution spaces [\#7156](https://github.com/kokkos/kokkos/pull/7156)
+
+### Bug Fixes
+* Do not return a copy of the input functor in `Experimental::for_each` [\#6910](https://github.com/kokkos/kokkos/pull/6910)
+* Fix `realloc` on views of non-default constructible element types [\#6993](https://github.com/kokkos/kokkos/pull/6993)
+* Fix undefined behavior in `View` initialization or fill with zeros [\#7014](https://github.com/kokkos/kokkos/pull/7014)
+* Fix `sort_by_key` on host execution spaces when building with NVCC [\#7059](https://github.com/kokkos/kokkos/pull/7059)
+* Fix using shared libraries and -fvisibility=hidden [\#7065](https://github.com/kokkos/kokkos/pull/7065)
+* Fix view reference counting when functor copy constructor throws in parallel dispatch [\#6289](https://github.com/kokkos/kokkos/pull/6289)
+* Fix `initialize(InitializationSetting)` for handling `print_configuration` setting [\#7098](https://github.com/kokkos/kokkos/pull/7098)
+* Thread safety fixes for the Serial and OpenMP backend [\#7080](https://github.com/kokkos/kokkos/pull/7080), [\#6151](https://github.com/kokkos/kokkos/pull/6151)
+
 ## [4.3.01](https://github.com/kokkos/kokkos/tree/4.3.01)
 [Full Changelog](https://github.com/kokkos/kokkos/compare/4.3.00...4.3.01)
 
 ### Backend and Architecture Enhancements:
 
 #### HIP:
-* MI300 support unified memory support [\#6877](https://github.com/kokkos/kokkos/pull/6877)
+* MI300 support unified memory [\#6877](https://github.com/kokkos/kokkos/pull/6877)
 
 ### Bug Fixes
 * Serial: Use the provided execution space instance in TeamPolicy [\#6951](https://github.com/kokkos/kokkos/pull/6951)

lib/kokkos/CITATION.cff

@@ -0,0 +1,65 @@
+cff-version: 1.2.0
+title: Kokkos
+message: >-
+  If you use this software, please cite the overview paper
+type: software
+authors:
+  - name: The Kokkos authors
+    website: https://kokkos.org/community/team/
+identifiers:
+  - type: url
+    website: https://kokkos.org/kokkos-core-wiki/citation.html
+repository-code: 'https://github.com/kokkos/kokkos'
+url: 'https://kokkos.org/'
+license: Apache-2.0
+preferred-citation:
+  type: article
+  authors:
+  - given-names: Christian R.
+    family-names: Trott
+  - given-names: Damien
+    family-names: Lebrun-Grandié
+  - given-names: Daniel
+    family-names: Arndt
+  - family-names: Ciesko
+    given-names: Jan
+  - given-names: Vinh
+    family-names: Dang
+  - family-names: Ellingwood
+    given-names: Nathan
+  - given-names: Rahulkumar
+    family-names: Gayatri
+  - given-names: Evan
+    family-names: Harvey
+  - given-names: Daisy S.
+    family-names: Hollman
+  - given-names: Dan
+    family-names: Ibanez
+  - given-names: Nevin
+    family-names: Liber
+  - given-names: Jonathan
+    family-names: Madsen
+  - given-names: Jeff
+    family-names: Miles
+  - given-names: David
+    family-names: Poliakoff
+  - given-names: Amy
+    family-names: Powell
+  - given-names: Sivasankaran
+    family-names: Rajamanickam
+  - given-names: Mikael
+    family-names: Simberg
+  - given-names: Dan
+    family-names: Sunderland
+  - given-names: Bruno
+    family-names: Turcksin
+  - given-names: Jeremiah
+    family-names: Wilke
+  doi: 10.1109/TPDS.2021.3097283
+  journal: IEEE Transactions on Parallel and Distributed Systems
+  start: 805
+  end: 817
+  title: "Kokkos 3: Programming Model Extensions for the Exascale Era"
+  volume: 33
+  issue: 4
+  year: 2022

lib/kokkos/CMakeLists.txt

@@ -150,7 +150,7 @@ ENDIF()
 
 
 set(Kokkos_VERSION_MAJOR 4)
-set(Kokkos_VERSION_MINOR 3)
+set(Kokkos_VERSION_MINOR 4)
 set(Kokkos_VERSION_PATCH 1)
 set(Kokkos_VERSION "${Kokkos_VERSION_MAJOR}.${Kokkos_VERSION_MINOR}.${Kokkos_VERSION_PATCH}")
 message(STATUS "Kokkos version: ${Kokkos_VERSION}")

lib/kokkos/Makefile.kokkos

@@ -11,7 +11,7 @@ CXXFLAGS += $(SHFLAGS)
 endif
 
 KOKKOS_VERSION_MAJOR = 4
-KOKKOS_VERSION_MINOR = 3
+KOKKOS_VERSION_MINOR = 4
 KOKKOS_VERSION_PATCH = 1
 KOKKOS_VERSION = $(shell echo $(KOKKOS_VERSION_MAJOR)*10000+$(KOKKOS_VERSION_MINOR)*100+$(KOKKOS_VERSION_PATCH) | bc)
 
@@ -21,11 +21,11 @@ KOKKOS_DEVICES ?= "OpenMP"
 # Options:
 # Intel:    KNC,KNL,SNB,HSW,BDW,SKL,SKX,ICL,ICX,SPR
 # NVIDIA:   Kepler,Kepler30,Kepler32,Kepler35,Kepler37,Maxwell,Maxwell50,Maxwell52,Maxwell53,Pascal60,Pascal61,Volta70,Volta72,Turing75,Ampere80,Ampere86,Ada89,Hopper90
-# ARM:      ARMv80,ARMv81,ARMv8-ThunderX,ARMv8-TX2,A64FX
+# ARM:      ARMv80,ARMv81,ARMv8-ThunderX,ARMv8-TX2,A64FX,ARMv9-Grace
 # IBM:      Power8,Power9
 # AMD-GPUS: AMD_GFX906,AMD_GFX908,AMD_GFX90A,AMD_GFX940,AMD_GFX942,AMD_GFX1030,AMD_GFX1100,AMD_GFX1103
 # AMD-CPUS: AMDAVX,Zen,Zen2,Zen3
-# Intel-GPUs: Gen9,Gen11,Gen12LP,DG1,XeHP,PVC
+# Intel-GPUs: Intel_Gen,Intel_Gen9,Intel_Gen11,Intel_Gen12LP,Intel_DG1,Intel_XeHP,Intel_PVC
 KOKKOS_ARCH ?= ""
 # Options: yes,no
 KOKKOS_DEBUG ?= "no"
@@ -41,7 +41,7 @@ KOKKOS_STANDALONE_CMAKE ?= "no"
 
 # Default settings specific options.
 # Options: force_uvm,use_ldg,rdc,enable_lambda,enable_constexpr,disable_malloc_async
-KOKKOS_CUDA_OPTIONS ?= "enable_lambda"
+KOKKOS_CUDA_OPTIONS ?= "disable_malloc_async"
 
 # Options: rdc
 KOKKOS_HIP_OPTIONS ?= ""
@@ -328,12 +328,43 @@ KOKKOS_INTERNAL_USE_ARCH_ICL := $(call kokkos_has_string,$(KOKKOS_ARCH),ICL)
 KOKKOS_INTERNAL_USE_ARCH_ICX := $(call kokkos_has_string,$(KOKKOS_ARCH),ICX)
 KOKKOS_INTERNAL_USE_ARCH_SPR := $(call kokkos_has_string,$(KOKKOS_ARCH),SPR)
 
-KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen)
+# Traditionally, we supported, e.g., IntelGen9 instead of Intel_Gen9. The latter
+# matches the CMake option but we also accept the former for backward-compatibility.
 KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9 := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen9)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9 := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_Gen9)
+endif
 KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN11 := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen11)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN11), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN11 := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_Gen11)
+endif
 KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN12LP := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen12LP)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN12LP), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN12LP := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_Gen12LP)
+endif
+KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen9)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9 := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_Gen9)
+endif
+KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN_SET := $(shell expr $(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN9)  \
+                                                     + $(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN11) \
+                                                     + $(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN12LP))
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN_SET), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelGen)
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN), 0)
+    KOKKOS_INTERNAL_USE_ARCH_INTEL_GEN := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_Gen)
+  endif
+endif
 KOKKOS_INTERNAL_USE_ARCH_INTEL_DG1 := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelDG1)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_DG1), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_DG1 := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_DG1)
+endif
 KOKKOS_INTERNAL_USE_ARCH_INTEL_XEHP := $(call kokkos_has_string,$(KOKKOS_ARCH),IntelXeHP)
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_INTEL_XEHP), 0)
+  KOKKOS_INTERNAL_USE_ARCH_INTEL_XEHP := $(call kokkos_has_string,$(KOKKOS_ARCH),Intel_XeHP)
+endif
+# Traditionally the architecture was called PVC instead of Intel_PVC. This
+# version makes us accept IntelPVC and Intel_PVC as well.
 KOKKOS_INTERNAL_USE_ARCH_INTEL_PVC := $(call kokkos_has_string,$(KOKKOS_ARCH),PVC)
 
 # NVIDIA based.
@@ -394,7 +425,8 @@ KOKKOS_INTERNAL_USE_ARCH_ARMV81 := $(call kokkos_has_string,$(KOKKOS_ARCH),ARMv8
 KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX := $(call kokkos_has_string,$(KOKKOS_ARCH),ARMv8-ThunderX)
 KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX2 := $(call kokkos_has_string,$(KOKKOS_ARCH),ARMv8-TX2)
 KOKKOS_INTERNAL_USE_ARCH_A64FX := $(call kokkos_has_string,$(KOKKOS_ARCH),A64FX)
-KOKKOS_INTERNAL_USE_ARCH_ARM := $(strip $(shell echo $(KOKKOS_INTERNAL_USE_ARCH_ARMV80)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV81)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX2)+$(KOKKOS_INTERNAL_USE_ARCH_A64FX) | bc))
+KOKKOS_INTERNAL_USE_ARCH_ARMV9_GRACE := $(call kokkos_has_string,$(KOKKOS_ARCH),ARMv9-Grace)
+KOKKOS_INTERNAL_USE_ARCH_ARM := $(strip $(shell echo $(KOKKOS_INTERNAL_USE_ARCH_ARMV80)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV81)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV8_THUNDERX2)+$(KOKKOS_INTERNAL_USE_ARCH_A64FX)+$(KOKKOS_INTERNAL_USE_ARCH_ARMV9_GRACE) | bc))
 
 # IBM based.
 KOKKOS_INTERNAL_USE_ARCH_POWER8 := $(call kokkos_has_string,$(KOKKOS_ARCH),Power8)
@@ -758,6 +790,14 @@ ifeq ($(KOKKOS_INTERNAL_USE_ARCH_A64FX), 1)
   endif
 endif
 
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_ARMV9_GRACE), 1)
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_ARMV9_GRACE")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_ARM_NEON")
+
+  KOKKOS_CXXFLAGS += -mcpu=neoverse-v2  -msve-vector-bits=128
+  KOKKOS_LDFLAGS += -mcpu=neoverse-v2  -msve-vector-bits=128
+endif
+
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_ZEN), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_ZEN")
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AVX2")
@@ -1216,6 +1256,8 @@ ifeq ($(KOKKOS_INTERNAL_DISABLE_BUNDLED_MDSPAN), 0)
 endif
 tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_IMPL_MDSPAN")
 
+tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_IMPL_REF_COUNT_BRANCH_UNLIKELY")
+
 KOKKOS_INTERNAL_LS_CONFIG := $(shell ls KokkosCore_config.h 2>&1)
 
 ifeq ($(KOKKOS_INTERNAL_LS_CONFIG), KokkosCore_config.h)

lib/kokkos/Makefile.targets

@@ -81,7 +81,7 @@ ifeq ($(KOKKOS_INTERNAL_USE_THREADS), 1)
 Kokkos_Threads_Instance.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Threads/Kokkos_Threads_Instance.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Threads/Kokkos_Threads_Instance.cpp
 Kokkos_Threads_Spinwait.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Threads/Kokkos_Threads_Spinwait.cpp
-	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Threads/Kokkos_Spinwait.cpp
+	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Threads/Kokkos_Threads_Spinwait.cpp
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENMP), 1)

lib/kokkos/README.md

@@ -1,4 +1,4 @@
-![Kokkos](https://avatars2.githubusercontent.com/u/10199860?s=200&v=4)
+[![Kokkos](https://avatars2.githubusercontent.com/u/10199860?s=200&v=4)](https://kokkos.org)
 
 # Kokkos: Core Libraries
 
@@ -10,43 +10,66 @@ hierarchies and multiple types of execution resources. It currently can use
 CUDA, HIP, SYCL, HPX, OpenMP and C++ threads as backend programming models with several other
 backends in development.
 
-**Kokkos Core is part of the Kokkos C++ Performance Portability Programming EcoSystem.**
+**Kokkos Core is part of the [Kokkos C++ Performance Portability Programming Ecosystem](https://kokkos.org/about/abstract/).**
 
-For the complete documentation, click below:
+Kokkos is a [Linux Foundation](https://linuxfoundation.org) project.
 
-# [kokkos.github.io/kokkos-core-wiki](https://kokkos.github.io/kokkos-core-wiki)
+## Learning about Kokkos
-
-# Learning about Kokkos
 
 To start learning about Kokkos:
 
-- [Kokkos Lectures](https://kokkos.github.io/kokkos-core-wiki/videolectures.html): they contain a mix of lecture videos and hands-on exercises covering all the important Kokkos Ecosystem capabilities.
+- [Kokkos Lectures](https://kokkos.org/kokkos-core-wiki/videolectures.html): they contain a mix of lecture videos and hands-on exercises covering all the important capabilities.
 
-- [Programming guide](https://kokkos.github.io/kokkos-core-wiki/programmingguide.html): contains in "narrative" form a technical description of the programming model, machine model, and the main building blocks like the Views and parallel dispatch.
+- [Programming guide](https://kokkos.org/kokkos-core-wiki/programmingguide.html): contains in "narrative" form a technical description of the programming model, machine model, and the main building blocks like the Views and parallel dispatch.
 
-- [API reference](https://kokkos.github.io/kokkos-core-wiki/): organized by category, i.e., [core](https://kokkos.github.io/kokkos-core-wiki/API/core-index.html), [algorithms](https://kokkos.github.io/kokkos-core-wiki/API/algorithms-index.html) and [containers](https://kokkos.github.io/kokkos-core-wiki/API/containers-index.html) or, if you prefer, in [alphabetical order](https://kokkos.github.io/kokkos-core-wiki/API/alphabetical.html).
+- [API reference](https://kokkos.org/kokkos-core-wiki/): organized by category, i.e., [core](https://kokkos.org/kokkos-core-wiki/API/core-index.html), [algorithms](https://kokkos.org/kokkos-core-wiki/API/algorithms-index.html) and [containers](https://kokkos.org/kokkos-core-wiki/API/containers-index.html) or, if you prefer, in [alphabetical order](https://kokkos.org/kokkos-core-wiki/API/alphabetical.html).
 
-- [Use cases and Examples](https://kokkos.github.io/kokkos-core-wiki/usecases.html): a series of examples ranging from how to use Kokkos with MPI to Fortran interoperability.
+- [Use cases and Examples](https://kokkos.org/kokkos-core-wiki/usecases.html): a serie of examples ranging from how to use Kokkos with MPI to Fortran interoperability.
+
+## Obtaining Kokkos
+
+The latest release of Kokkos can be obtained from the [GitHub releases page](https://github.com/kokkos/kokkos/releases/latest).
+
+The current release is [4.3.01](https://github.com/kokkos/kokkos/releases/tag/4.3.01).
+
+```bash
+curl -OJ -L https://github.com/kokkos/kokkos/archive/refs/tags/4.3.01.tar.gz
+# Or with wget
+wget https://github.com/kokkos/kokkos/archive/refs/tags/4.3.01.tar.gz
+```
+
+To clone the latest development version of Kokkos from GitHub:
+
+```bash
+git clone -b develop  https://github.com/kokkos/kokkos.git
+```
+
+### Building Kokkos
+
+To build Kokkos, you will need to have a C++ compiler that supports C++17 or later.
+All requirements including minimum and primary tested compiler versions can be found [here](https://kokkos.org/kokkos-core-wiki/requirements.html).
+
+Building and installation instructions are described [here](https://kokkos.org/kokkos-core-wiki/building.html).
+
+You can also install Kokkos using [Spack](https://spack.io/): `spack install kokkos`. [Available configuration options](https://packages.spack.io/package.html?name=kokkos) can be displayed using `spack info kokkos`.
+
+## For the complete documentation: [kokkos.org/kokkos-core-wiki/](https://kokkos.org/kokkos-core-wiki/)
+
+## Support
 
 For questions find us on Slack: https://kokkosteam.slack.com or open a GitHub issue.
 
 For non-public questions send an email to: *crtrott(at)sandia.gov*
 
-# Contributing to Kokkos
+## Contributing
 
-Please see [this page](https://kokkos.github.io/kokkos-core-wiki/contributing.html) for details on how to contribute.
+Please see [this page](https://kokkos.org/kokkos-core-wiki/contributing.html) for details on how to contribute.
 
-# Requirements, Building and Installing
+## Citing Kokkos
 
-All requirements including minimum and primary tested compiler versions can be found [here](https://kokkos.github.io/kokkos-core-wiki/requirements.html).
+Please see the [following page](https://kokkos.org/kokkos-core-wiki/citation.html).
 
-Building and installation instructions are described [here](https://kokkos.github.io/kokkos-core-wiki/building.html).
+## License
-
-# Citing Kokkos
-
-Please see the [following page](https://kokkos.github.io/kokkos-core-wiki/citation.html).
-
-# License
 
 [![License](https://img.shields.io/badge/License-Apache--2.0_WITH_LLVM--exception-blue)](https://spdx.org/licenses/LLVM-exception.html)
 

lib/kokkos/algorithms/src/sorting/impl/Kokkos_SortByKeyImpl.hpp

@@ -189,6 +189,33 @@ void applyPermutation(const ExecutionSpace& space,
       KOKKOS_LAMBDA(int i) { view(i) = view_copy(permutation(i)); });
 }
 
+// FIXME_NVCC: nvcc has trouble compiling lambdas inside a function with
+// variadic templates (sort_by_key_via_sort). Switch to using functors instead.
+template <typename Permute>
+struct IotaFunctor {
+  Permute _permute;
+  KOKKOS_FUNCTION void operator()(int i) const { _permute(i) = i; }
+};
+template <typename Keys>
+struct LessFunctor {
+  Keys _keys;
+  KOKKOS_FUNCTION bool operator()(int i, int j) const {
+    return _keys(i) < _keys(j);
+  }
+};
+
+// FIXME_NVCC+MSVC: We can't use a lambda instead of a functor which gave us
+// "For this host platform/dialect, an extended lambda cannot be defined inside
+// the 'if' or 'else' block of a constexpr if statement"
+template <typename Keys, typename Comparator>
+struct KeyComparisonFunctor {
+  Keys m_keys;
+  Comparator m_comparator;
+  KOKKOS_FUNCTION bool operator()(int i, int j) const {
+    return m_comparator(m_keys(i), m_keys(j));
+  }
+};
+
 template <class ExecutionSpace, class KeysDataType, class... KeysProperties,
           class ValuesDataType, class... ValuesProperties,
           class... MaybeComparator>
@@ -207,10 +234,9 @@ void sort_by_key_via_sort(
       n);
 
   // iota
-  Kokkos::parallel_for(
+  Kokkos::parallel_for("Kokkos::sort_by_key_via_sort::iota",
-      "Kokkos::sort_by_key_via_sort::iota",
                        Kokkos::RangePolicy<ExecutionSpace>(exec, 0, n),
-      KOKKOS_LAMBDA(int i) { permute(i) = i; });
+                       IotaFunctor<decltype(permute)>{permute});
 
   using Layout =
       typename Kokkos::View<unsigned int*, ExecutionSpace>::array_layout;
@@ -228,16 +254,15 @@ void sort_by_key_via_sort(
     Kokkos::DefaultHostExecutionSpace host_exec;
 
     if constexpr (sizeof...(MaybeComparator) == 0) {
-      Kokkos::sort(
+      Kokkos::sort(host_exec, host_permute,
-          host_exec, host_permute,
+                   LessFunctor<decltype(host_keys)>{host_keys});
-          KOKKOS_LAMBDA(int i, int j) { return host_keys(i) < host_keys(j); });
     } else {
       auto keys_comparator =
           std::get<0>(std::tuple<MaybeComparator...>(maybeComparator...));
       Kokkos::sort(
-          host_exec, host_permute, KOKKOS_LAMBDA(int i, int j) {
+          host_exec, host_permute,
-            return keys_comparator(host_keys(i), host_keys(j));
+          KeyComparisonFunctor<decltype(host_keys), decltype(keys_comparator)>{
-          });
+              host_keys, keys_comparator});
     }
     host_exec.fence("Kokkos::Impl::sort_by_key_via_sort: after host sort");
     Kokkos::deep_copy(exec, permute, host_permute);
@@ -262,16 +287,14 @@ void sort_by_key_via_sort(
     }
 #else
     if constexpr (sizeof...(MaybeComparator) == 0) {
-      Kokkos::sort(
+      Kokkos::sort(exec, permute, LessFunctor<decltype(keys)>{keys});
-          exec, permute,
-          KOKKOS_LAMBDA(int i, int j) { return keys(i) < keys(j); });
     } else {
       auto keys_comparator =
           std::get<0>(std::tuple<MaybeComparator...>(maybeComparator...));
       Kokkos::sort(
-          exec, permute, KOKKOS_LAMBDA(int i, int j) {
+          exec, permute,
-            return keys_comparator(keys(i), keys(j));
+          KeyComparisonFunctor<decltype(keys), decltype(keys_comparator)>{
-          });
+              keys, keys_comparator});
     }
 #endif
   }

lib/kokkos/algorithms/src/std_algorithms/Kokkos_ForEach.hpp

@@ -29,33 +29,31 @@ namespace Experimental {
 template <
     class ExecutionSpace, class IteratorType, class UnaryFunctorType,
     std::enable_if_t<Kokkos::is_execution_space_v<ExecutionSpace>, int> = 0>
-UnaryFunctorType for_each(const std::string& label, const ExecutionSpace& ex,
+void for_each(const std::string& label, const ExecutionSpace& ex,
-                          IteratorType first, IteratorType last,
+              IteratorType first, IteratorType last, UnaryFunctorType functor) {
-                          UnaryFunctorType functor) {
+  Impl::for_each_exespace_impl(label, ex, first, last, std::move(functor));
-  return Impl::for_each_exespace_impl(label, ex, first, last,
-                                      std::move(functor));
 }
 
 template <
     class ExecutionSpace, class IteratorType, class UnaryFunctorType,
     std::enable_if_t<Kokkos::is_execution_space_v<ExecutionSpace>, int> = 0>
-UnaryFunctorType for_each(const ExecutionSpace& ex, IteratorType first,
+void for_each(const ExecutionSpace& ex, IteratorType first, IteratorType last,
-                          IteratorType last, UnaryFunctorType functor) {
+              UnaryFunctorType functor) {
-  return Impl::for_each_exespace_impl("Kokkos::for_each_iterator_api_default",
+  Impl::for_each_exespace_impl("Kokkos::for_each_iterator_api_default", ex,
-                                      ex, first, last, std::move(functor));
+                               first, last, std::move(functor));
 }
 
 template <
     class ExecutionSpace, class DataType, class... Properties,
     class UnaryFunctorType,
     std::enable_if_t<Kokkos::is_execution_space_v<ExecutionSpace>, int> = 0>
-UnaryFunctorType for_each(const std::string& label, const ExecutionSpace& ex,
+void for_each(const std::string& label, const ExecutionSpace& ex,
               const ::Kokkos::View<DataType, Properties...>& v,
               UnaryFunctorType functor) {
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(v);
 
   namespace KE = ::Kokkos::Experimental;
-  return Impl::for_each_exespace_impl(label, ex, KE::begin(v), KE::end(v),
+  Impl::for_each_exespace_impl(label, ex, KE::begin(v), KE::end(v),
                                std::move(functor));
 }
 
@@ -63,15 +61,14 @@ template <
     class ExecutionSpace, class DataType, class... Properties,
     class UnaryFunctorType,
     std::enable_if_t<Kokkos::is_execution_space_v<ExecutionSpace>, int> = 0>
-UnaryFunctorType for_each(const ExecutionSpace& ex,
+void for_each(const ExecutionSpace& ex,
               const ::Kokkos::View<DataType, Properties...>& v,
               UnaryFunctorType functor) {
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(v);
 
   namespace KE = ::Kokkos::Experimental;
-  return Impl::for_each_exespace_impl("Kokkos::for_each_view_api_default", ex,
+  Impl::for_each_exespace_impl("Kokkos::for_each_view_api_default", ex,
-                                      KE::begin(v), KE::end(v),
+                               KE::begin(v), KE::end(v), std::move(functor));
-                                      std::move(functor));
 }
 
 //
@@ -82,23 +79,22 @@ UnaryFunctorType for_each(const ExecutionSpace& ex,
 
 template <class TeamHandleType, class IteratorType, class UnaryFunctorType,
           std::enable_if_t<Kokkos::is_team_handle_v<TeamHandleType>, int> = 0>
-KOKKOS_FUNCTION UnaryFunctorType for_each(const TeamHandleType& teamHandle,
+KOKKOS_FUNCTION void for_each(const TeamHandleType& teamHandle,
                               IteratorType first, IteratorType last,
                               UnaryFunctorType functor) {
-  return Impl::for_each_team_impl(teamHandle, first, last, std::move(functor));
+  Impl::for_each_team_impl(teamHandle, first, last, std::move(functor));
 }
 
 template <class TeamHandleType, class DataType, class... Properties,
           class UnaryFunctorType,
           std::enable_if_t<Kokkos::is_team_handle_v<TeamHandleType>, int> = 0>
-KOKKOS_FUNCTION UnaryFunctorType
+KOKKOS_FUNCTION void for_each(const TeamHandleType& teamHandle,
-for_each(const TeamHandleType& teamHandle,
                               const ::Kokkos::View<DataType, Properties...>& v,
                               UnaryFunctorType functor) {
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(v);
 
   namespace KE = ::Kokkos::Experimental;
-  return Impl::for_each_team_impl(teamHandle, KE::begin(v), KE::end(v),
+  Impl::for_each_team_impl(teamHandle, KE::begin(v), KE::end(v),
                            std::move(functor));
 }
 

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_AdjacentDifference.hpp

@@ -82,6 +82,11 @@ OutputIteratorType adjacent_difference_exespace_impl(
     return first_dest;
   }
 
+#ifdef KOKKOS_ENABLE_DEBUG
+  // check for overlapping iterators
+  Impl::expect_no_overlap(first_from, last_from, first_dest);
+#endif
+
   // run
   const auto num_elements =
       Kokkos::Experimental::distance(first_from, last_from);
@@ -114,6 +119,11 @@ KOKKOS_FUNCTION OutputIteratorType adjacent_difference_team_impl(
     return first_dest;
   }
 
+#ifdef KOKKOS_ENABLE_DEBUG
+  // check for overlapping iterators
+  Impl::expect_no_overlap(first_from, last_from, first_dest);
+#endif
+
   // run
   const auto num_elements =
       Kokkos::Experimental::distance(first_from, last_from);

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_Constraints.hpp

@@ -24,6 +24,9 @@ namespace Kokkos {
 namespace Experimental {
 namespace Impl {
 
+template <class T>
+class RandomAccessIterator;
+
 template <typename T, typename enable = void>
 struct is_admissible_to_kokkos_std_algorithms : std::false_type {};
 
@@ -58,6 +61,18 @@ using is_iterator = Kokkos::is_detected<iterator_category_t, T>;
 template <class T>
 inline constexpr bool is_iterator_v = is_iterator<T>::value;
 
+template <typename ViewType>
+struct is_kokkos_iterator : std::false_type {};
+
+template <typename ViewType>
+struct is_kokkos_iterator<RandomAccessIterator<ViewType>> {
+  static constexpr bool value =
+      is_admissible_to_kokkos_std_algorithms<ViewType>::value;
+};
+
+template <class T>
+inline constexpr bool is_kokkos_iterator_v = is_kokkos_iterator<T>::value;
+
 //
 // are_iterators
 //
@@ -215,6 +230,38 @@ KOKKOS_INLINE_FUNCTION void expect_valid_range(IteratorType first,
   (void)last;
 }
 
+//
+// Check if kokkos iterators are overlapping
+//
+template <typename IteratorType1, typename IteratorType2>
+KOKKOS_INLINE_FUNCTION void expect_no_overlap(
+    [[maybe_unused]] IteratorType1 first, [[maybe_unused]] IteratorType1 last,
+    [[maybe_unused]] IteratorType2 s_first) {
+  if constexpr (is_kokkos_iterator_v<IteratorType1> &&
+                is_kokkos_iterator_v<IteratorType2>) {
+    auto const view1 = first.view();
+    auto const view2 = s_first.view();
+
+    std::size_t stride1  = view1.stride(0);
+    std::size_t stride2  = view2.stride(0);
+    ptrdiff_t first_diff = view1.data() - view2.data();
+
+    // FIXME If strides are not identical, checks may not be made
+    // with the cost of O(1)
+    // Currently, checks are made only if strides are identical
+    // If first_diff == 0, there is already an overlap
+    if (stride1 == stride2 || first_diff == 0) {
+      [[maybe_unused]] bool is_no_overlap  = (first_diff % stride1);
+      auto* first_pointer1                 = view1.data();
+      auto* first_pointer2                 = view2.data();
+      [[maybe_unused]] auto* last_pointer1 = first_pointer1 + (last - first);
+      [[maybe_unused]] auto* last_pointer2 = first_pointer2 + (last - first);
+      KOKKOS_EXPECTS(first_pointer1 >= last_pointer2 ||
+                     last_pointer1 <= first_pointer2 || is_no_overlap);
+    }
+  }
+}
+
 }  // namespace Impl
 }  // namespace Experimental
 }  // namespace Kokkos

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_CopyIf.hpp

@@ -151,7 +151,8 @@ KOKKOS_FUNCTION OutputIterator copy_if_team_impl(
   }
 
 #if defined KOKKOS_COMPILER_INTEL ||                                  \
-    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130)
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+     !defined(KOKKOS_COMPILER_MSVC))
   __builtin_unreachable();
 #endif
 }

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_ForEachForEachN.hpp

@@ -42,8 +42,7 @@ struct StdForEachFunctor {
 };
 
 template <class HandleType, class IteratorType, class UnaryFunctorType>
-UnaryFunctorType for_each_exespace_impl(const std::string& label,
+void for_each_exespace_impl(const std::string& label, const HandleType& handle,
-                                        const HandleType& handle,
                             IteratorType first, IteratorType last,
                             UnaryFunctorType functor) {
   // checks
@@ -56,8 +55,6 @@ UnaryFunctorType for_each_exespace_impl(const std::string& label,
       label, RangePolicy<HandleType>(handle, 0, num_elements),
       StdForEachFunctor<IteratorType, UnaryFunctorType>(first, functor));
   handle.fence("Kokkos::for_each: fence after operation");
-
-  return functor;
 }
 
 template <class ExecutionSpace, class IteratorType, class SizeType,
@@ -75,7 +72,7 @@ IteratorType for_each_n_exespace_impl(const std::string& label,
   }
 
   for_each_exespace_impl(label, ex, first, last, std::move(functor));
-  // no neeed to fence since for_each_exespace_impl fences already
+  // no need to fence since for_each_exespace_impl fences already
 
   return last;
 }
@@ -84,9 +81,9 @@ IteratorType for_each_n_exespace_impl(const std::string& label,
 // team impl
 //
 template <class TeamHandleType, class IteratorType, class UnaryFunctorType>
-KOKKOS_FUNCTION UnaryFunctorType
+KOKKOS_FUNCTION void for_each_team_impl(const TeamHandleType& teamHandle,
-for_each_team_impl(const TeamHandleType& teamHandle, IteratorType first,
+                                        IteratorType first, IteratorType last,
-                   IteratorType last, UnaryFunctorType functor) {
+                                        UnaryFunctorType functor) {
   // checks
   Impl::static_assert_random_access_and_accessible(teamHandle, first);
   Impl::expect_valid_range(first, last);
@@ -96,7 +93,6 @@ for_each_team_impl(const TeamHandleType& teamHandle, IteratorType first,
       TeamThreadRange(teamHandle, 0, num_elements),
       StdForEachFunctor<IteratorType, UnaryFunctorType>(first, functor));
   teamHandle.team_barrier();
-  return functor;
 }
 
 template <class TeamHandleType, class IteratorType, class SizeType,
@@ -113,7 +109,7 @@ for_each_n_team_impl(const TeamHandleType& teamHandle, IteratorType first,
   }
 
   for_each_team_impl(teamHandle, first, last, std::move(functor));
-  // no neeed to fence since for_each_team_impl fences already
+  // no need to fence since for_each_team_impl fences already
 
   return last;
 }

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_RandomAccessIterator.hpp

@@ -59,6 +59,30 @@ class RandomAccessIterator< ::Kokkos::View<DataType, Args...> > {
                                                 ptrdiff_t current_index)
       : m_view(view), m_current_index(current_index) {}
 
+#ifndef KOKKOS_ENABLE_CXX17  // C++20 and beyond
+  template <class OtherViewType>
+  requires(std::is_constructible_v<view_type, OtherViewType>) KOKKOS_FUNCTION
+      explicit(!std::is_convertible_v<OtherViewType, view_type>)
+          RandomAccessIterator(const RandomAccessIterator<OtherViewType>& other)
+      : m_view(other.m_view), m_current_index(other.m_current_index) {}
+#else
+  template <
+      class OtherViewType,
+      std::enable_if_t<std::is_constructible_v<view_type, OtherViewType> &&
+                           !std::is_convertible_v<OtherViewType, view_type>,
+                       int> = 0>
+  KOKKOS_FUNCTION explicit RandomAccessIterator(
+      const RandomAccessIterator<OtherViewType>& other)
+      : m_view(other.m_view), m_current_index(other.m_current_index) {}
+
+  template <class OtherViewType,
+            std::enable_if_t<std::is_convertible_v<OtherViewType, view_type>,
+                             int> = 0>
+  KOKKOS_FUNCTION RandomAccessIterator(
+      const RandomAccessIterator<OtherViewType>& other)
+      : m_view(other.m_view), m_current_index(other.m_current_index) {}
+#endif
+
   KOKKOS_FUNCTION
   iterator_type& operator++() {
     ++m_current_index;
@@ -152,9 +176,16 @@ class RandomAccessIterator< ::Kokkos::View<DataType, Args...> > {
   KOKKOS_FUNCTION
   reference operator*() const { return m_view(m_current_index); }
 
+  KOKKOS_FUNCTION
+  view_type view() const { return m_view; }
+
  private:
   view_type m_view;
   ptrdiff_t m_current_index = 0;
+
+  // Needed for the converting constructor accepting another iterator
+  template <class>
+  friend class RandomAccessIterator;
 };
 
 }  // namespace Impl

lib/kokkos/algorithms/src/std_algorithms/impl/Kokkos_UniqueCopy.hpp

@@ -176,7 +176,8 @@ KOKKOS_FUNCTION OutputIterator unique_copy_team_impl(
     }
 
 #if defined KOKKOS_COMPILER_INTEL ||                                  \
-    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130)
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+     !defined(KOKKOS_COMPILER_MSVC))
     __builtin_unreachable();
 #endif
   }

lib/kokkos/algorithms/unit_tests/TestRandomAccessIterator.cpp

@@ -46,6 +46,44 @@ TEST_F(random_access_iterator_test, constructor) {
   EXPECT_TRUE(true);
 }
 
+TEST_F(random_access_iterator_test, constructiblity) {
+  auto first_d  = KE::begin(m_dynamic_view);
+  auto cfirst_d = KE::cbegin(m_dynamic_view);
+
+  static_assert(std::is_constructible_v<decltype(cfirst_d), decltype(first_d)>);
+  static_assert(
+      !std::is_constructible_v<decltype(first_d), decltype(cfirst_d)>);
+  [[maybe_unused]] decltype(cfirst_d) tmp_cfirst_d(first_d);
+
+  auto first_s  = KE::begin(m_static_view);
+  auto cfirst_s = KE::cbegin(m_static_view);
+
+  static_assert(std::is_constructible_v<decltype(cfirst_s), decltype(first_s)>);
+  static_assert(
+      !std::is_constructible_v<decltype(first_s), decltype(cfirst_s)>);
+  [[maybe_unused]] decltype(cfirst_s) tmp_cfirst_s(first_s);
+
+  auto first_st  = KE::begin(m_strided_view);
+  auto cfirst_st = KE::cbegin(m_strided_view);
+
+  static_assert(
+      std::is_constructible_v<decltype(cfirst_st), decltype(first_st)>);
+  static_assert(
+      !std::is_constructible_v<decltype(first_st), decltype(cfirst_st)>);
+  [[maybe_unused]] decltype(cfirst_st) tmp_cfirst_st(first_st);
+
+  // [FIXME] Better to have tests for the explicit specifier with an expression.
+  // As soon as View converting constructors are re-implemented with a
+  // conditional explicit, we may add those tests.
+  static_assert(std::is_constructible_v<decltype(first_s), decltype(first_d)>);
+  static_assert(std::is_constructible_v<decltype(first_st), decltype(first_d)>);
+  static_assert(std::is_constructible_v<decltype(first_d), decltype(first_s)>);
+  static_assert(std::is_constructible_v<decltype(first_st), decltype(first_s)>);
+  static_assert(std::is_constructible_v<decltype(first_d), decltype(first_st)>);
+  static_assert(std::is_constructible_v<decltype(first_s), decltype(first_st)>);
+  EXPECT_TRUE(true);
+}
+
 template <class IteratorType, class ValueType>
 void test_random_access_it_verify(IteratorType it, ValueType gold_value) {
   using view_t = Kokkos::View<typename IteratorType::value_type>;

lib/kokkos/algorithms/unit_tests/TestSortByKey.hpp

@@ -69,7 +69,7 @@ void iota(ExecutionSpace const &space, ViewType const &v,
           typename ViewType::value_type value = 0) {
   using ValueType = typename ViewType::value_type;
   Kokkos::parallel_for(
-      "ArborX::Algorithms::iota",
+      "Kokkos::Algorithms::iota",
       Kokkos::RangePolicy<ExecutionSpace>(space, 0, v.extent(0)),
       KOKKOS_LAMBDA(int i) { v(i) = value + (ValueType)i; });
 }
@@ -87,6 +87,18 @@ TEST(TEST_CATEGORY, SortByKeyEmptyView) {
       Kokkos::Experimental::sort_by_key(ExecutionSpace(), keys, values));
 }
 
+// Test #7036
+TEST(TEST_CATEGORY, SortByKeyEmptyViewHost) {
+  using ExecutionSpace = Kokkos::DefaultHostExecutionSpace;
+
+  // does not matter if we use int or something else
+  Kokkos::View<int *, ExecutionSpace> keys("keys", 0);
+  Kokkos::View<float *, ExecutionSpace> values("values", 0);
+
+  ASSERT_NO_THROW(
+      Kokkos::Experimental::sort_by_key(ExecutionSpace(), keys, values));
+}
+
 TEST(TEST_CATEGORY, SortByKey) {
   using ExecutionSpace = TEST_EXECSPACE;
   using MemorySpace    = typename ExecutionSpace::memory_space;

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsConstraints.cpp

@@ -81,5 +81,114 @@ TEST(std_algorithms, is_admissible_to_std_algorithms) {
                strided_view_3d_t>::value);
 }
 
+TEST(std_algorithms, expect_no_overlap) {
+  namespace KE     = Kokkos::Experimental;
+  using value_type = double;
+
+  static constexpr size_t extent0 = 13;
+
+  //-------------
+  // 1d views
+  //-------------
+  using static_view_1d_t = Kokkos::View<value_type[extent0]>;
+  [[maybe_unused]] static_view_1d_t static_view_1d{
+      "std-algo-test-1d-contiguous-view-static"};
+
+  using dyn_view_1d_t = Kokkos::View<value_type*>;
+  [[maybe_unused]] dyn_view_1d_t dynamic_view_1d{
+      "std-algo-test-1d-contiguous-view-dynamic", extent0};
+
+  using strided_view_1d_t = Kokkos::View<value_type*, Kokkos::LayoutStride>;
+  Kokkos::LayoutStride layout1d{extent0, 2};
+  strided_view_1d_t strided_view_1d{"std-algo-test-1d-strided-view", layout1d};
+
+// Overlapping because iterators are identical
+#if defined(KOKKOS_ENABLE_DEBUG)
+  auto first_s = KE::begin(static_view_1d);
+  auto last_s  = first_s + extent0;
+  EXPECT_DEATH({ KE::Impl::expect_no_overlap(first_s, last_s, first_s); },
+               "Kokkos contract violation:.*");
+
+  auto first_d = KE::begin(dynamic_view_1d);
+  auto last_d  = first_d + extent0;
+  EXPECT_DEATH({ KE::Impl::expect_no_overlap(first_d, last_d, first_d); },
+               "Kokkos contract violation:.*");
+
+  auto first_st = KE::begin(strided_view_1d);
+  auto last_st  = first_st + extent0;
+  EXPECT_DEATH({ KE::Impl::expect_no_overlap(first_st, last_st, first_st); },
+               "Kokkos contract violation:.*");
+#endif
+
+  // Ranges are overlapped
+  static constexpr size_t sub_extent0 = 6, offset0 = 3;
+  std::pair<size_t, size_t> range0(0, sub_extent0),
+      range1(offset0, offset0 + sub_extent0);
+#if defined(KOKKOS_ENABLE_DEBUG)
+  auto static_view_1d_0 = Kokkos::subview(static_view_1d, range0);
+  auto static_view_1d_1 = Kokkos::subview(static_view_1d, range1);
+  auto first_s0         = KE::begin(static_view_1d_0);  // [0, 6)
+  auto last_s0          = first_s0 + static_view_1d_0.extent(0);
+  auto first_s1         = KE::begin(static_view_1d_1);  // [3, 9)
+  EXPECT_DEATH({ KE::Impl::expect_no_overlap(first_s0, last_s0, first_s1); },
+               "Kokkos contract violation:.*");
+
+  auto dynamic_view_1d_0 = Kokkos::subview(dynamic_view_1d, range0);
+  auto dynamic_view_1d_1 = Kokkos::subview(dynamic_view_1d, range1);
+  auto first_d0          = KE::begin(dynamic_view_1d_0);  // [0, 6)
+  auto last_d0           = first_d0 + dynamic_view_1d_0.extent(0);
+  auto first_d1          = KE::begin(dynamic_view_1d_1);  // [3, 9)
+  EXPECT_DEATH({ KE::Impl::expect_no_overlap(first_d0, last_d0, first_d1); },
+               "Kokkos contract violation:.*");
+#endif
+
+  auto strided_view_1d_0 = Kokkos::subview(strided_view_1d, range0);
+  auto strided_view_1d_1 = Kokkos::subview(strided_view_1d, range1);
+  auto first_st0         = KE::begin(strided_view_1d_0);  // [0, 12)
+  auto last_st0          = first_st0 + strided_view_1d_0.extent(0);
+  auto first_st1         = KE::begin(strided_view_1d_1);  // [3, 15)
+  // Does not overlap since offset (=3) is not divisible by stride (=2)
+  EXPECT_NO_THROW(
+      { KE::Impl::expect_no_overlap(first_st0, last_st0, first_st1); });
+
+  // Iterating over the same range without overlapping
+  Kokkos::View<value_type[2][extent0], Kokkos::LayoutLeft> static_view_2d{
+      "std-algo-test-2d-contiguous-view-static"};
+  auto sub_static_view_1d_0 = Kokkos::subview(static_view_2d, 0, Kokkos::ALL);
+  auto sub_static_view_1d_1 = Kokkos::subview(static_view_2d, 1, Kokkos::ALL);
+  auto sub_first_s0         = KE::begin(sub_static_view_1d_0);  // 0, 2, 4, ...
+  auto sub_last_s0          = sub_first_s0 + sub_static_view_1d_0.extent(0);
+  auto sub_first_s1         = KE::begin(sub_static_view_1d_1);  // 1, 3, 5, ...
+
+  EXPECT_NO_THROW({
+    KE::Impl::expect_no_overlap(sub_first_s0, sub_last_s0, sub_first_s1);
+  });
+
+  Kokkos::View<value_type**, Kokkos::LayoutLeft> dynamic_view_2d{
+      "std-algo-test-2d-contiguous-view-dynamic", 2, extent0};
+  auto sub_dynamic_view_1d_0 = Kokkos::subview(dynamic_view_2d, 0, Kokkos::ALL);
+  auto sub_dynamic_view_1d_1 = Kokkos::subview(dynamic_view_2d, 1, Kokkos::ALL);
+  auto sub_first_d0 = KE::begin(sub_dynamic_view_1d_0);  // 0, 2, 4, ...
+  auto sub_last_d0  = sub_first_d0 + sub_dynamic_view_1d_0.extent(0);
+  auto sub_first_d1 = KE::begin(sub_dynamic_view_1d_1);  // 1, 3, 5, ...
+
+  EXPECT_NO_THROW({
+    KE::Impl::expect_no_overlap(sub_first_d0, sub_last_d0, sub_first_d1);
+  });
+
+  Kokkos::LayoutStride layout2d{2, 3, extent0, 2 * 3};
+  Kokkos::View<value_type**, Kokkos::LayoutStride> strided_view_2d{
+      "std-algo-test-2d-contiguous-view-strided", layout2d};
+  auto sub_strided_view_1d_0 = Kokkos::subview(strided_view_2d, 0, Kokkos::ALL);
+  auto sub_strided_view_1d_1 = Kokkos::subview(strided_view_2d, 1, Kokkos::ALL);
+  auto sub_first_st0 = KE::begin(sub_strided_view_1d_0);  // 0, 6, 12, ...
+  auto sub_last_st0  = sub_first_st0 + sub_strided_view_1d_0.extent(0);
+  auto sub_first_st1 = KE::begin(sub_strided_view_1d_1);  // 1, 7, 13, ...
+
+  EXPECT_NO_THROW({
+    KE::Impl::expect_no_overlap(sub_first_st0, sub_last_st0, sub_first_st1);
+  });
+}
+
 }  // namespace stdalgos
 }  // namespace Test

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamExclusiveScan.cpp

@@ -85,7 +85,7 @@ struct TestFunctorA {
         break;
       }
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
 
       case 2: {
         auto it = KE::exclusive_scan(
@@ -213,7 +213,7 @@ void test_A(std::size_t numTeams, std::size_t numCols, int apiId) {
         break;
       }
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
       case 2:
       case 3: {
         auto it = exclusive_scan(KE::cbegin(rowFrom), KE::cend(rowFrom),
@@ -242,7 +242,7 @@ template <class LayoutTag, class ValueType, class InPlaceOrVoid = void>
 void run_all_scenarios() {
   for (int numTeams : teamSizesToTest) {
     for (const auto& numCols : {0, 1, 2, 13, 101, 1444, 8153}) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
       for (int apiId : {0, 1, 2, 3}) {
 #else
       for (int apiId : {0, 1}) {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamIsSorted.cpp

@@ -52,7 +52,7 @@ struct TestFunctorA {
       Kokkos::single(Kokkos::PerTeam(member),
                      [=, *this]() { m_returnsView(myRowIndex) = result; });
     }
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
     else if (m_apiPick == 2) {
       using value_type = typename ViewType::value_type;
       result = KE::is_sorted(member, KE::cbegin(myRowView), KE::cend(myRowView),
@@ -179,7 +179,7 @@ template <class LayoutTag, class ValueType>
 void run_all_scenarios(bool makeDataSortedOnPurpose) {
   for (int numTeams : teamSizesToTest) {
     for (const auto& numCols : {0, 1, 2, 13, 101, 1444, 5153}) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
       for (int apiId : {0, 1, 2, 3}) {
 #else
       for (int apiId : {0, 1}) {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamIsSortedUntil.cpp

@@ -73,7 +73,7 @@ struct TestFunctorA {
         m_distancesView(myRowIndex) = resultDist;
       });
     }
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
     else if (m_apiPick == 2) {
       using value_type = typename ViewType::value_type;
       auto it          = KE::is_sorted_until(member, KE::cbegin(myRowView),
@@ -226,7 +226,7 @@ template <class LayoutTag, class ValueType>
 void run_all_scenarios(const std::string& name, const std::vector<int>& cols) {
   for (int numTeams : teamSizesToTest) {
     for (const auto& numCols : cols) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
       for (int apiId : {0, 1, 2, 3}) {
 #else
       for (int apiId : {0, 1}) {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamMaxElement.cpp

@@ -59,7 +59,7 @@ struct TestFunctorA {
         m_distancesView(myRowIndex) = resultDist;
       });
     }
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
     else if (m_apiPick == 2) {
       using value_type = typename ViewType::value_type;
       auto it =
@@ -170,7 +170,7 @@ void run_all_scenarios() {
 }
 
 TEST(std_algorithms_max_element_team_test, test) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
   run_all_scenarios<DynamicTag, int>();
   run_all_scenarios<StridedTwoRowsTag, double>();
   run_all_scenarios<StridedThreeRowsTag, int>();

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamMinElement.cpp

@@ -59,7 +59,7 @@ struct TestFunctorA {
         m_distancesView(myRowIndex) = resultDist;
       });
     }
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
     else if (m_apiPick == 2) {
       using value_type = typename ViewType::value_type;
       auto it =
@@ -169,7 +169,7 @@ void run_all_scenarios() {
 }
 
 TEST(std_algorithms_min_element_team_test, test) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
   run_all_scenarios<DynamicTag, int>();
   run_all_scenarios<StridedTwoRowsTag, double>();
   run_all_scenarios<StridedThreeRowsTag, int>();

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamMinMaxElement.cpp

@@ -66,7 +66,7 @@ struct TestFunctorA {
         m_distancesView(myRowIndex, 1) = resultDist2;
       });
     }
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
     else if (m_apiPick == 2) {
       using value_type = typename ViewType::value_type;
       auto itPair =
@@ -188,7 +188,7 @@ void run_all_scenarios() {
 }
 
 TEST(std_algorithms_minmax_element_team_test, test) {
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
   run_all_scenarios<DynamicTag, int>();
   run_all_scenarios<StridedTwoRowsTag, double>();
   run_all_scenarios<StridedThreeRowsTag, int>();

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamReduce.cpp

@@ -16,7 +16,7 @@
 
 #include <TestStdAlgorithmsCommon.hpp>
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
 
 namespace Test {
 namespace stdalgos {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamTransformExclusiveScan.cpp

@@ -16,7 +16,7 @@
 
 #include <TestStdAlgorithmsCommon.hpp>
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
 
 namespace Test {
 namespace stdalgos {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamTransformInclusiveScan.cpp

@@ -16,7 +16,7 @@
 
 #include <TestStdAlgorithmsCommon.hpp>
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
 
 namespace Test {
 namespace stdalgos {

lib/kokkos/algorithms/unit_tests/TestStdAlgorithmsTeamTransformReduce.cpp

@@ -16,7 +16,7 @@
 
 #include <TestStdAlgorithmsCommon.hpp>
 
-#if not defined KOKKOS_ENABLE_OPENMPTARGET
+#ifndef KOKKOS_ENABLE_OPENMPTARGET
 
 namespace Test {
 namespace stdalgos {

lib/kokkos/appveyor.yml

@@ -5,6 +5,6 @@ build_script:
 - cmd: >-
     mkdir build &&
     cd build &&
-    cmake c:\projects\source -DKokkos_ENABLE_TESTS=ON -DCMAKE_CXX_FLAGS="/W0 /EHsc" -DKokkos_ENABLE_DEPRECATED_CODE_4=ON -DKokkos_ENABLE_DEPRECATION_WARNINGS=OFF &&
+    cmake c:\projects\source -DKokkos_ENABLE_IMPL_MDSPAN=OFF -DKokkos_ENABLE_TESTS=ON -DCMAKE_CXX_FLAGS="/W0 /EHsc" -DKokkos_ENABLE_DEPRECATED_CODE_4=ON -DKokkos_ENABLE_DEPRECATION_WARNINGS=OFF &&
     cmake --build . --target install &&
     ctest -C Debug --output-on-failure

lib/kokkos/benchmarks/CMakeLists.txt

@@ -4,7 +4,7 @@ KOKKOS_ADD_BENCHMARK_DIRECTORIES(gather)
 KOKKOS_ADD_BENCHMARK_DIRECTORIES(gups)
 KOKKOS_ADD_BENCHMARK_DIRECTORIES(launch_latency)
 KOKKOS_ADD_BENCHMARK_DIRECTORIES(stream)
-
+KOKKOS_ADD_BENCHMARK_DIRECTORIES(view_copy_constructor)
 #FIXME_OPENMPTARGET - These two benchmarks cause ICE. Commenting them for now but a deeper analysis on the cause and a possible fix will follow.
 IF(NOT Kokkos_ENABLE_OPENMPTARGET)
     KOKKOS_ADD_BENCHMARK_DIRECTORIES(policy_performance)

lib/kokkos/benchmarks/view_copy_constructor/CMakeLists.txt

@@ -0,0 +1,4 @@
+KOKKOS_ADD_EXECUTABLE(
+  view_copy_constructor
+  SOURCES view_copy_constructor.cpp
+)

lib/kokkos/benchmarks/view_copy_constructor/Makefile

@@ -0,0 +1,46 @@
+KOKKOS_DEVICES=Serial
+KOKKOS_ARCH = ""
+
+
+MAKEFILE_PATH := $(subst Makefile,,$(abspath $(lastword $(MAKEFILE_LIST))))
+
+ifndef KOKKOS_PATH
+  KOKKOS_PATH = $(MAKEFILE_PATH)../..
+endif
+
+SRC = $(wildcard $(MAKEFILE_PATH)*.cpp)
+HEADERS = $(wildcard $(MAKEFILE_PATH)*.hpp)
+
+vpath %.cpp $(sort $(dir $(SRC)))
+
+default: build
+	echo "Start Build"
+
+CXX = clang++
+EXE = view_copy_constructor.exe
+
+CXXFLAGS ?= -Ofast
+override CXXFLAGS += -I$(MAKEFILE_PATH)
+
+DEPFLAGS = -M
+LINK = ${CXX}
+LINKFLAGS = -Ofast
+KOKKOS_CXX_STANDARD=c++20
+
+OBJ = $(notdir $(SRC:.cpp=.o))
+LIB =
+
+include $(KOKKOS_PATH)/Makefile.kokkos
+
+build: $(EXE)
+
+$(EXE): $(OBJ) $(KOKKOS_LINK_DEPENDS)
+	$(LINK) $(KOKKOS_LDFLAGS) $(LINKFLAGS) $(EXTRA_PATH) $(OBJ) $(KOKKOS_LIBS) $(LIB) -o $(EXE)
+
+clean: kokkos-clean
+	rm -f *.o view_copy_constructor.cuda view_copy_constructor.exe
+
+# Compilation rules
+
+%.o:%.cpp $(KOKKOS_CPP_DEPENDS) $(HEADERS)
+	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) $(EXTRA_INC) -c $< -o $(notdir $@)

lib/kokkos/benchmarks/view_copy_constructor/view_copy_constructor.cpp

@@ -0,0 +1,310 @@
+//@HEADER
+// ************************************************************************
+//
+//                        Kokkos v. 4.0
+//       Copyright (2022) National Technology & Engineering
+//               Solutions of Sandia, LLC (NTESS).
+//
+// Under the terms of Contract DE-NA0003525 with NTESS,
+// the U.S. Government retains certain rights in this software.
+//
+// Part of Kokkos, under the Apache License v2.0 with LLVM Exceptions.
+// See https://kokkos.org/LICENSE for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//@HEADER
+
+// The function "test_view_collection" exposes the copy constructor
+// and destructor overheads in Kokkos View objects
+// Please see the lines marked by "NOTE".
+
+#include <limits>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <sys/time.h>
+#include <Kokkos_Core.hpp>
+#include <iostream>
+
+// NVIEWS is the number of Kokkos View objects in our ViewCollection object
+// We have chosen a large value of 40 to make it easier to see performance
+// differences when using the likelihood attribute
+#define NVIEWS 40
+
+class ViewCollection {
+ public:
+  Kokkos::View<double*> v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, v13,
+      v14, v15, v16, v17, v18, v19, v20, v21, v22, v23, v24, v25, v26, v27, v28,
+      v29, v30, v31, v32, v33, v34, v35, v36, v37, v38, v39, v40;
+  double m_expected_sum;
+  double m_side_effect;
+  int m_N;
+
+  ViewCollection(int N)
+      : v1("v1", N),
+        v2("v2", N),
+        v3("v3", N),
+        v4("v4", N),
+        v5("v5", N),
+        v6("v6", N),
+        v7("v7", N),
+        v8("v8", N),
+        v9("v9", N),
+        v10("v10", N),
+        v11("v11", N),
+        v12("v12", N),
+        v13("v13", N),
+        v14("v14", N),
+        v15("v15", N),
+        v16("v16", N),
+        v17("v17", N),
+        v18("v18", N),
+        v19("v19", N),
+        v20("v20", N),
+        v21("v21", N),
+        v22("v22", N),
+        v23("v23", N),
+        v24("v24", N),
+        v25("v25", N),
+        v26("v26", N),
+        v27("v27", N),
+        v28("v28", N),
+        v29("v29", N),
+        v30("v30", N),
+        v31("v31", N),
+        v32("v32", N),
+        v33("v33", N),
+        v34("v34", N),
+        v35("v35", N),
+        v36("v36", N),
+        v37("v37", N),
+        v38("v38", N),
+        v39("v39", N),
+        v40("v40", N),
+        m_expected_sum(N * NVIEWS),
+        m_side_effect(0.0),
+        m_N(N) {
+    for (int i = 0; i < N; ++i) {
+      v1(i)  = 1;
+      v2(i)  = 1;
+      v3(i)  = 1;
+      v4(i)  = 1;
+      v5(i)  = 1;
+      v6(i)  = 1;
+      v7(i)  = 1;
+      v8(i)  = 1;
+      v9(i)  = 1;
+      v10(i) = 1;
+      v11(i) = 1;
+      v12(i) = 1;
+      v13(i) = 1;
+      v14(i) = 1;
+      v15(i) = 1;
+      v16(i) = 1;
+      v17(i) = 1;
+      v18(i) = 1;
+      v19(i) = 1;
+      v20(i) = 1;
+      v21(i) = 1;
+      v22(i) = 1;
+      v23(i) = 1;
+      v24(i) = 1;
+      v25(i) = 1;
+      v26(i) = 1;
+      v27(i) = 1;
+      v28(i) = 1;
+      v29(i) = 1;
+      v30(i) = 1;
+      v31(i) = 1;
+      v32(i) = 1;
+      v33(i) = 1;
+      v34(i) = 1;
+      v35(i) = 1;
+      v36(i) = 1;
+      v37(i) = 1;
+      v38(i) = 1;
+      v39(i) = 1;
+      v40(i) = 1;
+    }
+  }
+
+// The ADD_COPY_CONSTRUCTOR macro is helpful to compare time in the copy
+// constructor between compilers. We have found that the GNU compiler
+// is sometimes able to inline the default copy constructor.
+#ifdef ADD_COPY_CONSTRUCTOR
+  __attribute__((noinline)) ViewCollection(const ViewCollection& other)
+      : v1(other.v1),
+        v2(other.v2),
+        v3(other.v3),
+        v4(other.v4),
+        v5(other.v5),
+        v6(other.v6),
+        v7(other.v7),
+        v8(other.v8),
+        v9(other.v9),
+        v10(other.v10),
+        v11(other.v11),
+        v12(other.v12),
+        v13(other.v13),
+        v14(other.v14),
+        v15(other.v15),
+        v16(other.v16),
+        v17(other.v17),
+        v18(other.v18),
+        v19(other.v19),
+        v20(other.v20),
+        v21(other.v21),
+        v22(other.v22),
+        v23(other.v23),
+        v24(other.v24),
+        v25(other.v25),
+        v26(other.v26),
+        v27(other.v27),
+        v28(other.v28),
+        v29(other.v29),
+        v30(other.v30),
+        v31(other.v31),
+        v32(other.v32),
+        v33(other.v33),
+        v34(other.v34),
+        v35(other.v35),
+        v36(other.v36),
+        v37(other.v37),
+        v38(other.v38),
+        v39(other.v39),
+        v40(other.v40),
+        m_expected_sum(other.m_expected_sum),
+        m_side_effect(other.m_side_effect),
+        m_N(other.m_N) {}
+#endif
+
+  KOKKOS_INLINE_FUNCTION
+  double sum_views(int ii, bool execute_kernel) {
+    double result = 0.0;
+    if (execute_kernel) {
+      // This code is only executed when using the command line option -k
+      // The computation references all Kokkos views. This may help our
+      // effort to stop compilers from optimizing away the Kokkos views
+      for (int i = 0; i < m_N; ++i) {
+        result += v1(i) + v2(i) + v3(i) + v4(i) + v5(i) + v6(i) + v7(i) +
+                  v8(i) + v9(i) + v10(i) + v11(i) + v12(i) + v13(i) + v14(i) +
+                  v15(i) + v16(i) + v17(i) + v18(i) + v19(i) + v20(i) + v21(i) +
+                  v22(i) + v23(i) + v24(i) + v25(i) + v26(i) + v27(i) + v28(i) +
+                  v29(i) + v30(i) + v31(i) + v32(i) + v33(i) + v34(i) + v35(i) +
+                  v36(i) + v37(i) + v38(i) + v39(i) + v40(i);
+      }
+    } else {
+      result = m_expected_sum;
+    }
+    // This statement introduces a side effect that may help our effort to
+    // stop compilers from optimizing away the temporary ViewCollection object
+    m_side_effect = result * (ii + 1);
+    return result;
+  }
+};
+
+void test_view_collection_kk(int N, int num_iter, bool execute_kernel) {
+  ViewCollection view_collection(N);
+
+  Kokkos::Timer view_collection_timer;
+  double max_value = 0.0;
+  // Max Reduction boilerplate code taken from slide 53 of
+  // kokkos-tutorials/LectureSeries/KokkosTutorial_02_ViewsAndSpaces.pdf
+  Kokkos::parallel_reduce(
+      "collection-reduction", num_iter,
+      KOKKOS_LAMBDA(int i, double& valueToUpdate) {
+        // NOTE: The following lines expose the Kokkos View overheads
+        ViewCollection tmp_view_collection = view_collection;
+        double my_value = tmp_view_collection.sum_views(i, execute_kernel);
+        if (my_value > valueToUpdate) valueToUpdate = my_value;
+      },
+      Kokkos::Max<double>(max_value));
+  double view_collection_time = view_collection_timer.seconds();
+
+  bool success = std::fabs(max_value - N * NVIEWS) < 1.E-6;
+  std::cout << "View Time = " << view_collection_time << " seconds"
+            << std::endl;
+  if (success) {
+    std::cout << "Kokkos run:" << std::endl;
+    std::cout << "SUCCESS" << std::endl;
+  } else {
+    std::cout << "FAILURE" << std::endl;
+  }
+}
+
+void test_view_collection_serial(int N, int num_iter, bool execute_kernel) {
+  ViewCollection view_collection(N);
+
+  Kokkos::Timer view_collection_timer;
+  double max_value = 0.0;
+  // Max Reduction boilerplate code taken from slide 53 of
+  // kokkos-tutorials/LectureSeries/KokkosTutorial_02_ViewsAndSpaces.pdf
+  for (int i = 0; i < num_iter; ++i) {
+    // NOTE: The following lines expose the Kokkos View overheads
+    ViewCollection tmp_view_collection = view_collection;
+    double my_value = tmp_view_collection.sum_views(i, execute_kernel);
+    if (my_value > max_value) max_value = my_value;
+  }
+  double view_collection_time = view_collection_timer.seconds();
+
+  bool success = std::fabs(max_value - N * NVIEWS) < 1.E-6;
+  std::cout << "View Time 2 = " << view_collection_time << " seconds"
+            << std::endl;
+  if (success) {
+    std::cout << "Serial run:" << std::endl;
+    std::cout << "SUCCESS" << std::endl;
+  } else {
+    std::cout << "FAILURE" << std::endl;
+  }
+}
+
+int main(int argc, char* argv[]) {
+  // The benchmark is only testing reference counting for views on host.
+#if defined(KOKKOS_ENABLE_OPENMP) || defined(KOKKOS_ENABLE_SERIAL) || \
+    defined(KOKKOS_ENABLE_THREADS) || defined(KOKKOS_ENABLE_HPX)
+  int N               = 1;
+  int num_iter        = 1 << 27;
+  bool execute_kernel = false;
+
+  for (int i = 0; i < argc; i++) {
+    if ((strcmp(argv[i], "-N") == 0)) {
+      N = atoi(argv[++i]);
+      if (N < 1) {
+        std::cout << "Array extent must be >= 1" << std::endl;
+        exit(1);
+      }
+    } else if (strcmp(argv[i], "-i") == 0) {
+      num_iter = atoi(argv[++i]);
+      if (num_iter < 1) {
+        std::cout << "Number of iterations must be >= 1" << std::endl;
+        exit(1);
+      }
+    } else if (strcmp(argv[i], "-k") == 0) {
+      execute_kernel = true;
+    } else if ((strcmp(argv[i], "-h") == 0)) {
+      printf("  Options:\n");
+      printf("  -N <int>: Array extent\n");
+      printf("  -i <int>: Number of iterations\n");
+      printf("  -k:       Execute the summation kernel\n");
+      printf("  -h:       Print this message\n\n");
+      exit(1);
+    }
+  }
+
+  std::cout << "Array extent = " << N << std::endl;
+  std::cout << "Iterations = " << num_iter << std::endl;
+  std::cout << "Execute summation kernel = " << std::boolalpha << execute_kernel
+            << std::noboolalpha << std::endl;
+
+  // Test inside a Kokkos kernel.
+  Kokkos::initialize(argc, argv);
+  { test_view_collection_kk(N, num_iter, execute_kernel); }
+
+  // Test outside Kokkos kernel.
+  test_view_collection_serial(N, num_iter, execute_kernel);
+
+  Kokkos::finalize();
+#endif
+
+  return 0;
+}

lib/kokkos/bin/nvcc_wrapper

@@ -233,7 +233,7 @@ do
     cuda_args="$cuda_args $1"
     ;;
   #Handle more known nvcc args
-  --extended-lambda|--expt-extended-lambda|--expt-relaxed-constexpr|--Wno-deprecated-gpu-targets|-Wno-deprecated-gpu-targets|-allow-unsupported-compiler|--allow-unsupported-compiler)
+  --extended-lambda|--expt-extended-lambda|--expt-relaxed-constexpr|--Wno-deprecated-gpu-targets|-Wno-deprecated-gpu-targets|-allow-unsupported-compiler|--allow-unsupported-compiler|--disable-warnings)
     cuda_args="$cuda_args $1"
     ;;
   #Handle known nvcc args that have an argument

lib/kokkos/cmake/Dependencies.cmake

@@ -1,6 +1,5 @@
 TRIBITS_PACKAGE_DEFINE_DEPENDENCIES(
   LIB_OPTIONAL_TPLS Pthread CUDA HWLOC DLlib
-  TEST_OPTIONAL_TPLS CUSPARSE
   )
 
 TRIBITS_TPL_TENTATIVELY_ENABLE(DLlib)

lib/kokkos/cmake/KokkosConfigCommon.cmake.in

@@ -225,8 +225,13 @@ FUNCTION(kokkos_compilation)
     # if built w/o CUDA support, we want to basically make this a no-op
     SET(_Kokkos_ENABLE_CUDA @Kokkos_ENABLE_CUDA@)
 
+
+    IF(CMAKE_VERSION VERSION_GREATER_EQUAL 3.17)
+      SET(MAYBE_CURRENT_INSTALLATION_ROOT "${CMAKE_CURRENT_FUNCTION_LIST_DIR}/../../..")
+    ENDIF()
+
     # search relative first and then absolute
-    SET(_HINTS "${CMAKE_CURRENT_LIST_DIR}/../.." "@CMAKE_INSTALL_PREFIX@")
+    SET(_HINTS "${MAYBE_CURRENT_INSTALLATION_ROOT}" "@CMAKE_INSTALL_PREFIX@")
 
     # find kokkos_launch_compiler
     FIND_PROGRAM(Kokkos_COMPILE_LAUNCHER

lib/kokkos/cmake/KokkosCore_config.h.in

@@ -37,6 +37,7 @@
 #cmakedefine KOKKOS_ENABLE_CUDA_LAMBDA  // deprecated
 #cmakedefine KOKKOS_ENABLE_CUDA_CONSTEXPR
 #cmakedefine KOKKOS_ENABLE_IMPL_CUDA_MALLOC_ASYNC
+#cmakedefine KOKKOS_ENABLE_IMPL_CUDA_UNIFIED_MEMORY
 #cmakedefine KOKKOS_ENABLE_HIP_RELOCATABLE_DEVICE_CODE
 #cmakedefine KOKKOS_ENABLE_HIP_MULTIPLE_KERNEL_INSTANTIATIONS
 #cmakedefine KOKKOS_ENABLE_IMPL_HIP_UNIFIED_MEMORY
@@ -52,6 +53,8 @@
 #cmakedefine KOKKOS_OPT_RANGE_AGGRESSIVE_VECTORIZATION  // deprecated
 #cmakedefine KOKKOS_ENABLE_AGGRESSIVE_VECTORIZATION
 #cmakedefine KOKKOS_ENABLE_IMPL_MDSPAN
+#cmakedefine KOKKOS_ENABLE_IMPL_REF_COUNT_BRANCH_UNLIKELY
+#cmakedefine KOKKOS_ENABLE_IMPL_VIEW_OF_VIEWS_DESTRUCTOR_PRECONDITION_VIOLATION_WORKAROUND
 #cmakedefine KOKKOS_ENABLE_ATOMICS_BYPASS
 
 /* TPL Settings */
@@ -65,6 +68,7 @@
 #cmakedefine KOKKOS_ARCH_ARMV8_THUNDERX
 #cmakedefine KOKKOS_ARCH_ARMV81
 #cmakedefine KOKKOS_ARCH_ARMV8_THUNDERX2
+#cmakedefine KOKKOS_ARCH_ARMV9_GRACE
 #cmakedefine KOKKOS_ARCH_A64FX
 #cmakedefine KOKKOS_ARCH_AVX
 #cmakedefine KOKKOS_ARCH_AVX2

lib/kokkos/cmake/Modules/FindTPLCUDA.cmake

@@ -7,13 +7,18 @@ IF (NOT CUDAToolkit_ROOT)
   ENDIF()
 ENDIF()
 
-# FIXME CMake 3.28.4 creates more targets than we export
+IF(KOKKOS_CXX_HOST_COMPILER_ID STREQUAL NVHPC AND CMAKE_VERSION VERSION_LESS "3.20.1")
-IF(CMAKE_VERSION VERSION_GREATER_EQUAL "3.17.0" AND CMAKE_VERSION VERSION_LESS "3.28.4")
+  MESSAGE(FATAL_ERROR "Using NVHPC as host compiler requires at least CMake 3.20.1")
-  find_package(CUDAToolkit)
-ELSE()
-  include(${CMAKE_CURRENT_LIST_DIR}/CudaToolkit.cmake)
 ENDIF()
 
+IF(CMAKE_VERSION VERSION_GREATER_EQUAL "3.17.0")
+  find_package(CUDAToolkit REQUIRED)
+  KOKKOS_CREATE_IMPORTED_TPL(CUDA INTERFACE
+    LINK_LIBRARIES CUDA::cuda_driver CUDA::cudart
+  )
+  KOKKOS_EXPORT_CMAKE_TPL(CUDAToolkit REQUIRED)
+ELSE()
+  include(${CMAKE_CURRENT_LIST_DIR}/CudaToolkit.cmake)
 
   IF (TARGET CUDA::cudart)
     SET(FOUND_CUDART TRUE)
@@ -30,14 +35,10 @@ ELSE()
   ENDIF()
 
   include(FindPackageHandleStandardArgs)
-IF(KOKKOS_CXX_HOST_COMPILER_ID STREQUAL NVHPC)
+  FIND_PACKAGE_HANDLE_STANDARD_ARGS(TPLCUDA ${DEFAULT_MSG} FOUND_CUDART FOUND_CUDA_DRIVER)
-  SET(KOKKOS_CUDA_ERROR "Using NVHPC as host compiler requires at least CMake 3.20.1")
-ELSE()
-  SET(KOKKOS_CUDA_ERROR DEFAULT_MSG)
-ENDIF()
-FIND_PACKAGE_HANDLE_STANDARD_ARGS(TPLCUDA ${KOKKOS_CUDA_ERROR} FOUND_CUDART FOUND_CUDA_DRIVER)
   IF (FOUND_CUDA_DRIVER AND FOUND_CUDART)
     KOKKOS_CREATE_IMPORTED_TPL(CUDA INTERFACE
       LINK_LIBRARIES CUDA::cuda_driver CUDA::cudart
     )
   ENDIF()
+ENDIF()

lib/kokkos/cmake/deps/CUDA.cmake

@@ -35,7 +35,6 @@ IF(NOT _CUDA_FAILURE)
   GLOBAL_SET(TPL_CUDA_LIBRARY_DIRS)
   GLOBAL_SET(TPL_CUDA_INCLUDE_DIRS ${CUDA_TOOLKIT_INCLUDE})
   GLOBAL_SET(TPL_CUDA_LIBRARIES ${CUDA_CUDART_LIBRARY} ${CUDA_cublas_LIBRARY} ${CUDA_cufft_LIBRARY})
-  KOKKOS_CREATE_IMPORTED_TPL_LIBRARY(CUSPARSE)
 ELSE()
   SET(TPL_ENABLE_CUDA OFF)
 ENDIF()

lib/kokkos/cmake/deps/CUSPARSE.cmake

@@ -1,26 +0,0 @@
-#@HEADER
-# ************************************************************************
-#
-#                        Kokkos v. 4.0
-#       Copyright (2022) National Technology & Engineering
-#               Solutions of Sandia, LLC (NTESS).
-#
-# Under the terms of Contract DE-NA0003525 with NTESS,
-# the U.S. Government retains certain rights in this software.
-#
-# Part of Kokkos, under the Apache License v2.0 with LLVM Exceptions.
-#
-# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-#
-# ************************************************************************
-# @HEADER
-
-#include(${TRIBITS_DEPS_DIR}/CUDA.cmake)
-
-#IF (TPL_ENABLE_CUDA)
-#  GLOBAL_SET(TPL_CUSPARSE_LIBRARY_DIRS)
-#  GLOBAL_SET(TPL_CUSPARSE_INCLUDE_DIRS ${TPL_CUDA_INCLUDE_DIRS})
-#  GLOBAL_SET(TPL_CUSPARSE_LIBRARIES    ${CUDA_cusparse_LIBRARY})
-#  KOKKOS_CREATE_IMPORTED_TPL_LIBRARY(CUSPARSE)
-#ENDIF()
-

lib/kokkos/cmake/fake_tribits.cmake

@@ -118,14 +118,6 @@ FUNCTION(KOKKOS_ADD_TEST)
   ENDIF()
 ENDFUNCTION()
 
-FUNCTION(KOKKOS_ADD_ADVANCED_TEST)
-  if (KOKKOS_HAS_TRILINOS)
-    TRIBITS_ADD_ADVANCED_TEST(${ARGN})
-  else()
-    # TODO Write this
-  endif()
-ENDFUNCTION()
-
 MACRO(KOKKOS_CREATE_IMPORTED_TPL_LIBRARY TPL_NAME)
   ADD_INTERFACE_LIBRARY(TPL_LIB_${TPL_NAME})
   TARGET_LINK_LIBRARIES(TPL_LIB_${TPL_NAME} LINK_PUBLIC ${TPL_${TPL_NAME}_LIBRARIES})

lib/kokkos/cmake/kokkos_arch.cmake

@@ -28,6 +28,7 @@ KOKKOS_CHECK_DEPRECATED_OPTIONS(
 #-------------------------------------------------------------------------------
 SET(KOKKOS_ARCH_LIST)
 
+include(CheckCXXCompilerFlag)
 
 KOKKOS_DEPRECATED_LIST(ARCH ARCH)
 
@@ -49,6 +50,7 @@ DECLARE_AND_CHECK_HOST_ARCH(ARMV81            "ARMv8.1 Compatible CPU")
 DECLARE_AND_CHECK_HOST_ARCH(ARMV8_THUNDERX    "ARMv8 Cavium ThunderX CPU")
 DECLARE_AND_CHECK_HOST_ARCH(ARMV8_THUNDERX2   "ARMv8 Cavium ThunderX2 CPU")
 DECLARE_AND_CHECK_HOST_ARCH(A64FX             "ARMv8.2 with SVE Support")
+DECLARE_AND_CHECK_HOST_ARCH(ARMV9_GRACE       "ARMv9 NVIDIA Grace CPU")
 DECLARE_AND_CHECK_HOST_ARCH(SNB               "Intel Sandy/Ivy Bridge CPUs")
 DECLARE_AND_CHECK_HOST_ARCH(HSW               "Intel Haswell CPUs")
 DECLARE_AND_CHECK_HOST_ARCH(BDW               "Intel Broadwell Xeon E-class CPUs")
@@ -189,12 +191,6 @@ IF (KOKKOS_CXX_COMPILER_ID STREQUAL Clang)
   ELSEIF(CUDAToolkit_BIN_DIR)
     GLOBAL_APPEND(KOKKOS_CUDA_OPTIONS --cuda-path=${CUDAToolkit_BIN_DIR}/..)
   ENDIF()
-ELSEIF (KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC)
-  SET(CUDA_ARCH_FLAG "-gpu")
-  GLOBAL_APPEND(KOKKOS_CUDA_OPTIONS -cuda)
-  IF (KOKKOS_ENABLE_CUDA) # FIXME ideally unreachable when CUDA not enabled
-    GLOBAL_APPEND(KOKKOS_LINK_OPTIONS -cuda)
-  ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA)
   SET(CUDA_ARCH_FLAG "-arch")
 ENDIF()
@@ -209,6 +205,11 @@ ENDIF()
 
 
 #------------------------------- KOKKOS_HIP_OPTIONS ---------------------------
+KOKKOS_OPTION(IMPL_AMDGPU_FLAGS "" STRING "Set compiler flags for AMD GPUs")
+KOKKOS_OPTION(IMPL_AMDGPU_LINK "" STRING "Set linker flags for AMD GPUs")
+MARK_AS_ADVANCED(Kokkos_IMPL_AMDGPU_FLAGS)
+MARK_AS_ADVANCED(Kokkos_IMPL_AMDGPU_LINK)
+
 #clear anything that might be in the cache
 GLOBAL_SET(KOKKOS_AMDGPU_OPTIONS)
 IF(KOKKOS_ENABLE_HIP)
@@ -301,6 +302,20 @@ IF (KOKKOS_ARCH_A64FX)
   )
 ENDIF()
 
+IF (KOKKOS_ARCH_ARMV9_GRACE)
+  SET(KOKKOS_ARCH_ARM_NEON ON)
+  check_cxx_compiler_flag("-mcpu=neoverse-n2" COMPILER_SUPPORTS_NEOVERSE_N2)
+  check_cxx_compiler_flag("-msve-vector-bits=128" COMPILER_SUPPORTS_SVE_VECTOR_BITS)
+  IF (COMPILER_SUPPORTS_NEOVERSE_N2 AND COMPILER_SUPPORTS_SVE_VECTOR_BITS)
+    COMPILER_SPECIFIC_FLAGS(
+      COMPILER_ID KOKKOS_CXX_HOST_COMPILER_ID
+      DEFAULT -mcpu=neoverse-n2 -msve-vector-bits=128
+    )
+  ELSE()
+    MESSAGE(WARNING "Compiler does not support ARMv9 Grace architecture")
+  ENDIF()
+ENDIF()
+
 IF (KOKKOS_ARCH_ZEN)
   COMPILER_SPECIFIC_FLAGS(
     COMPILER_ID KOKKOS_CXX_HOST_COMPILER_ID
@@ -535,16 +550,16 @@ IF (KOKKOS_CXX_HOST_COMPILER_ID STREQUAL NVHPC)
   SET(KOKKOS_ARCH_AVX512XEON OFF)
 ENDIF()
 
+# FIXME_NVCC nvcc doesn't seem to support Arm Neon.
+IF(KOKKOS_ARCH_ARM_NEON AND KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA)
+  UNSET(KOKKOS_ARCH_ARM_NEON)
+ENDIF()
+
 IF (NOT KOKKOS_COMPILE_LANGUAGE STREQUAL CUDA)
   IF (KOKKOS_ENABLE_CUDA_RELOCATABLE_DEVICE_CODE)
       COMPILER_SPECIFIC_FLAGS(
         Clang  -fcuda-rdc
         NVIDIA --relocatable-device-code=true
-        NVHPC -gpu=rdc
-      )
-  ELSEIF(KOKKOS_ENABLE_CUDA)
-    COMPILER_SPECIFIC_FLAGS(
-      NVHPC -gpu=nordc
       )
   ENDIF()
 ENDIF()
@@ -571,7 +586,7 @@ IF (KOKKOS_ENABLE_HIP)
     COMPILER_SPECIFIC_FLAGS(
       DEFAULT -fgpu-rdc
     )
-    IF (NOT KOKKOS_CXX_COMPILER_ID STREQUAL HIPCC)
+    IF (NOT KOKKOS_CXX_COMPILER_ID STREQUAL HIPCC AND NOT KOKKOS_IMPL_AMDGPU_FLAGS)
       COMPILER_SPECIFIC_LINK_OPTIONS(
         DEFAULT --hip-link
       )
@@ -653,11 +668,6 @@ FUNCTION(CHECK_CUDA_ARCH ARCH FLAG)
       SET(KOKKOS_CUDA_ARCH_FLAG ${FLAG} PARENT_SCOPE)
       IF(KOKKOS_ENABLE_COMPILE_AS_CMAKE_LANGUAGE)
         SET(CMAKE_CUDA_ARCHITECTURES ${KOKKOS_CUDA_ARCHITECTURES} PARENT_SCOPE)
-      ELSE()
-        IF(KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC)
-          STRING(REPLACE "sm_" "cc" NVHPC_CUDA_ARCH ${FLAG})
-          GLOBAL_APPEND(KOKKOS_CUDA_OPTIONS "${CUDA_ARCH_FLAG}=${NVHPC_CUDA_ARCH}")
-          GLOBAL_APPEND(KOKKOS_LINK_OPTIONS "${CUDA_ARCH_FLAG}=${NVHPC_CUDA_ARCH}")
       ELSE()
         GLOBAL_APPEND(KOKKOS_CUDA_OPTIONS "${CUDA_ARCH_FLAG}=${FLAG}")
         IF(KOKKOS_ENABLE_CUDA_RELOCATABLE_DEVICE_CODE OR KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA)
@@ -666,7 +676,6 @@ FUNCTION(CHECK_CUDA_ARCH ARCH FLAG)
       ENDIF()
     ENDIF()
   ENDIF()
-  ENDIF()
   LIST(APPEND KOKKOS_CUDA_ARCH_FLAGS ${FLAG})
   SET(KOKKOS_CUDA_ARCH_FLAGS ${KOKKOS_CUDA_ARCH_FLAGS} PARENT_SCOPE)
   LIST(APPEND KOKKOS_CUDA_ARCH_LIST ${ARCH})
@@ -707,8 +716,10 @@ FUNCTION(CHECK_AMDGPU_ARCH ARCH FLAG)
         IF(KOKKOS_ENABLE_HIP)
           SET(KOKKOS_HIP_ARCHITECTURES ${FLAG} PARENT_SCOPE)
         ENDIF()
+        IF(NOT KOKKOS_IMPL_AMDGPU_FLAGS)
           SET(KOKKOS_AMDGPU_ARCH_FLAG ${FLAG} PARENT_SCOPE)
           GLOBAL_APPEND(KOKKOS_AMDGPU_OPTIONS "${AMDGPU_ARCH_FLAG}=${FLAG}")
+        ENDIF()
         IF(KOKKOS_ENABLE_HIP_RELOCATABLE_DEVICE_CODE)
           GLOBAL_APPEND(KOKKOS_LINK_OPTIONS "${AMDGPU_ARCH_FLAG}=${FLAG}")
         ENDIF()
@@ -724,6 +735,15 @@ FOREACH(ARCH IN LISTS SUPPORTED_AMD_ARCHS)
   CHECK_AMDGPU_ARCH(${ARCH} ${FLAG})
 ENDFOREACH()
 
+IF(KOKKOS_IMPL_AMDGPU_FLAGS)
+  IF (NOT AMDGPU_ARCH_ALREADY_SPECIFIED)
+    MESSAGE(FATAL_ERROR "When IMPL_AMDGPU_FLAGS is set the architecture autodectection is disabled. "
+      "Please explicitly set the GPU architecture.")
+  ENDIF()
+  GLOBAL_APPEND(KOKKOS_AMDGPU_OPTIONS "${KOKKOS_IMPL_AMDGPU_FLAGS}")
+  GLOBAL_APPEND(KOKKOS_LINK_OPTIONS "${KOKKOS_IMPL_AMDGPU_LINK}")
+ENDIF()
+
 MACRO(SET_AND_CHECK_AMD_ARCH ARCH FLAG)
   KOKKOS_SET_OPTION(ARCH_${ARCH} ON)
   CHECK_AMDGPU_ARCH(${ARCH} ${FLAG})
@@ -984,7 +1004,7 @@ IF (KOKKOS_ARCH_HOPPER90)
 ENDIF()
 
 #HIP detection of gpu arch
-IF(KOKKOS_ENABLE_HIP AND NOT AMDGPU_ARCH_ALREADY_SPECIFIED)
+IF(KOKKOS_ENABLE_HIP AND NOT AMDGPU_ARCH_ALREADY_SPECIFIED AND NOT KOKKOS_IMPL_AMDGPU_FLAGS)
   FIND_PROGRAM(ROCM_ENUMERATOR rocm_agent_enumerator)
   IF(NOT ROCM_ENUMERATOR)
     MESSAGE(FATAL_ERROR "Autodetection of AMD GPU architecture not possible as "

lib/kokkos/cmake/kokkos_compiler_id.cmake

@@ -42,12 +42,8 @@ IF(Kokkos_ENABLE_CUDA)
     # If launcher was found and nvcc_wrapper was not specified as
     # compiler and `CMAKE_CXX_COMPILIER_LAUNCHER` is not set, set to use launcher.
     # Will ensure CMAKE_CXX_COMPILER is replaced by nvcc_wrapper
-    IF(Kokkos_COMPILE_LAUNCHER AND NOT INTERNAL_HAVE_COMPILER_NVCC AND NOT KOKKOS_CXX_COMPILER_ID STREQUAL Clang
+    IF(Kokkos_COMPILE_LAUNCHER AND NOT INTERNAL_HAVE_COMPILER_NVCC AND NOT KOKKOS_CXX_COMPILER_ID STREQUAL Clang)
-       AND NOT (Kokkos_ENABLE_IMPL_NVHPC_AS_DEVICE_COMPILER AND KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC))
       IF(CMAKE_CXX_COMPILER_LAUNCHER)
-        IF(KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC)
-          MESSAGE(STATUS "Using nvc++ as device compiler requires Kokkos_ENABLE_IMPL_NVHPC_AS_DEVICE_COMPILER=ON!")
-        ENDIF()
         MESSAGE(FATAL_ERROR "Cannot use CMAKE_CXX_COMPILER_LAUNCHER if the CMAKE_CXX_COMPILER is not able to compile CUDA code, i.e. nvcc_wrapper or clang++!")
       ENDIF()
       # the first argument to launcher is always the C++ compiler defined by cmake
@@ -149,56 +145,85 @@ IF(KOKKOS_CXX_COMPILER_ID STREQUAL Fujitsu)
 ENDIF()
 
 # Enforce the minimum compilers supported by Kokkos.
-SET(KOKKOS_MESSAGE_TEXT "Compiler not supported by Kokkos.  Required compiler versions:")
+IF(NOT CMAKE_CXX_STANDARD)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(CPU)         8.0.0 or higher")
+  SET(CMAKE_CXX_STANDARD 17)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(CUDA)       10.0.0 or higher")
+ENDIF()
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(OpenMPTarget) 15.0.0 or higher")
+IF(CMAKE_CXX_STANDARD EQUAL 17)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    GCC                8.2.0 or higher")
+  SET(KOKKOS_CLANG_CPU_MINIMUM             8.0.0)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Intel             19.0.5 or higher")
+  SET(KOKKOS_CLANG_CUDA_MINIMUM           10.0.0)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    IntelLLVM(CPU)  2021.1.1 or higher")
+  SET(KOKKOS_CLANG_OPENMPTARGET_MINIMUM   15.0.0)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    IntelLLVM(SYCL) 2023.0.0 or higher")
+  SET(KOKKOS_GCC_MINIMUM                   8.2.0)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    NVCC              11.0.0 or higher")
+  SET(KOKKOS_INTEL_MINIMUM                19.0.5)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    HIPCC              5.2.0 or higher")
+  SET(KOKKOS_INTEL_LLVM_CPU_MINIMUM     2021.1.1)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    NVHPC/PGI           22.3 or higher")
+  SET(KOKKOS_INTEL_LLVM_SYCL_MINIMUM    2023.0.0)
-SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    MSVC               19.29 or higher")
+  SET(KOKKOS_NVCC_MINIMUM                 11.0.0)
+  SET(KOKKOS_HIPCC_MINIMUM                 5.2.0)
+  SET(KOKKOS_NVHPC_MINIMUM                  22.3)
+  SET(KOKKOS_MSVC_MINIMUM                  19.29)
+ELSE()
+  SET(KOKKOS_CLANG_CPU_MINIMUM            14.0.0)
+  SET(KOKKOS_CLANG_CUDA_MINIMUM           14.0.0)
+  SET(KOKKOS_CLANG_OPENMPTARGET_MINIMUM   15.0.0)
+  SET(KOKKOS_GCC_MINIMUM                  10.1.0)
+  SET(KOKKOS_INTEL_MINIMUM                "not supported")
+  SET(KOKKOS_INTEL_LLVM_CPU_MINIMUM     2022.0.0)
+  SET(KOKKOS_INTEL_LLVM_SYCL_MINIMUM    2023.0.0)
+  SET(KOKKOS_NVCC_MINIMUM                 12.0.0)
+  SET(KOKKOS_HIPCC_MINIMUM                 5.2.0)
+  SET(KOKKOS_NVHPC_MINIMUM                22.3)
+  SET(KOKKOS_MSVC_MINIMUM                 19.30)
+ENDIF()
+
+SET(KOKKOS_MESSAGE_TEXT "Compiler not supported by Kokkos for C++${CMAKE_CXX_STANDARD}. Required minimum compiler versions:")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(CPU)          ${KOKKOS_CLANG_CPU_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(CUDA)         ${KOKKOS_CLANG_CUDA_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Clang(OpenMPTarget) ${KOKKOS_CLANG_OPENMPTARGET_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    GCC                 ${KOKKOS_GCC_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    Intel               ${KOKKOS_INTEL_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    IntelLLVM(CPU)      ${KOKKOS_INTEL_LLVM_CPU_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    IntelLLVM(SYCL)     ${KOKKOS_INTEL_LLVM_SYCL_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    NVCC                ${KOKKOS_NVCC_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    HIPCC               ${KOKKOS_HIPCC_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    NVHPC/PGI           ${KOKKOS_NVHPC_MINIMUM}")
+SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    MSVC                ${KOKKOS_MSVC_MINIMUM}")
 SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\n    XL/XLClang          not supported")
 SET(KOKKOS_MESSAGE_TEXT "${KOKKOS_MESSAGE_TEXT}\nCompiler: ${KOKKOS_CXX_COMPILER_ID} ${KOKKOS_CXX_COMPILER_VERSION}\n")
 
 IF(KOKKOS_CXX_COMPILER_ID STREQUAL Clang AND NOT Kokkos_ENABLE_CUDA)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 8.0.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_CLANG_CPU_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL Clang AND Kokkos_ENABLE_CUDA)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 10.0.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_CLANG_CUDA_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL GNU)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 8.2.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_GCC_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL Intel)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 19.0.5)
+  IF((NOT CMAKE_CXX_STANDARD EQUAL 17) OR (KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_INTEL_MINIMUM}))
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL IntelLLVM AND NOT Kokkos_ENABLE_SYCL)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 2021.1.1)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_INTEL_LLVM_CPU_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL IntelLLVM AND Kokkos_ENABLE_SYCL)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 2023.0.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_INTEL_LLVM_SYCL_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 11.0.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_NVCC_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
   SET(CMAKE_CXX_EXTENSIONS OFF CACHE BOOL "Kokkos turns off CXX extensions" FORCE)
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL HIPCC)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 5.2.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_HIPCC_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL PGI OR KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 22.3)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_NVHPC_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
   # Treat PGI internally as NVHPC to simplify handling both compilers.
@@ -206,13 +231,13 @@ ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL PGI OR KOKKOS_CXX_COMPILER_ID STREQUAL NV
   # backward-compatible to pgc++.
   SET(KOKKOS_CXX_COMPILER_ID NVHPC CACHE STRING INTERNAL FORCE)
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL "MSVC")
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 19.29)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS ${KOKKOS_MSVC_MINIMUM})
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL XL OR KOKKOS_CXX_COMPILER_ID STREQUAL XLClang)
   MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
 ELSEIF(KOKKOS_CXX_COMPILER_ID STREQUAL Clang AND Kokkos_ENABLE_OPENMPTARGET)
-  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS 15.0.0)
+  IF(KOKKOS_CXX_COMPILER_VERSION VERSION_LESS KOKKOS_CLANG_OPENMPTARGET_MINIMUM)
     MESSAGE(FATAL_ERROR "${KOKKOS_MESSAGE_TEXT}")
   ENDIF()
 ENDIF()

lib/kokkos/cmake/kokkos_enable_options.cmake

@@ -48,6 +48,8 @@ KOKKOS_ENABLE_OPTION(CUDA_LAMBDA ${CUDA_LAMBDA_DEFAULT} "Whether to allow lambda
 # resolved but we keep the option around a bit longer to be safe.
 KOKKOS_ENABLE_OPTION(IMPL_CUDA_MALLOC_ASYNC ON  "Whether to enable CudaMallocAsync (requires CUDA Toolkit 11.2)")
 KOKKOS_ENABLE_OPTION(IMPL_NVHPC_AS_DEVICE_COMPILER OFF "Whether to allow nvc++ as Cuda device compiler")
+KOKKOS_ENABLE_OPTION(IMPL_CUDA_UNIFIED_MEMORY OFF "Whether to leverage unified memory architectures for CUDA")
+
 KOKKOS_ENABLE_OPTION(DEPRECATED_CODE_4    ON "Whether code deprecated in major release 4 is available" )
 KOKKOS_ENABLE_OPTION(DEPRECATION_WARNINGS ON "Whether to emit deprecation warnings" )
 KOKKOS_ENABLE_OPTION(HIP_RELOCATABLE_DEVICE_CODE  OFF "Whether to enable relocatable device code (RDC) for HIP")
@@ -75,8 +77,12 @@ KOKKOS_ENABLE_OPTION(IMPL_HIP_UNIFIED_MEMORY OFF "Whether to leverage unified me
 # This option will go away eventually, but allows fallback to old implementation when needed.
 KOKKOS_ENABLE_OPTION(DESUL_ATOMICS_EXTERNAL OFF "Whether to use an external desul installation")
 KOKKOS_ENABLE_OPTION(ATOMICS_BYPASS OFF "**NOT RECOMMENDED** Whether to make atomics non-atomic for non-threaded MPI-only use cases")
+KOKKOS_ENABLE_OPTION(IMPL_REF_COUNT_BRANCH_UNLIKELY ON "Whether to use the C++20 `[[unlikely]]` attribute in the view reference counting")
+mark_as_advanced(Kokkos_ENABLE_IMPL_REF_COUNT_BRANCH_UNLIKELY)
+KOKKOS_ENABLE_OPTION(IMPL_VIEW_OF_VIEWS_DESTRUCTOR_PRECONDITION_VIOLATION_WORKAROUND OFF "Whether to enable a workaround for invalid use of View of Views that causes program hang on destruction.")
+mark_as_advanced(Kokkos_ENABLE_IMPL_VIEW_OF_VIEWS_DESTRUCTOR_PRECONDITION_VIOLATION_WORKAROUND)
 
-KOKKOS_ENABLE_OPTION(IMPL_MDSPAN OFF "Whether to enable experimental mdspan support")
+KOKKOS_ENABLE_OPTION(IMPL_MDSPAN ON "Whether to enable experimental mdspan support")
 KOKKOS_ENABLE_OPTION(MDSPAN_EXTERNAL OFF BOOL "Whether to use an external version of mdspan")
 KOKKOS_ENABLE_OPTION(IMPL_SKIP_COMPILER_MDSPAN ON BOOL "Whether to use an internal version of mdspan even if the compiler supports mdspan")
 mark_as_advanced(Kokkos_ENABLE_IMPL_MDSPAN)
@@ -131,7 +137,7 @@ FUNCTION(check_device_specific_options)
   ENDIF()
 ENDFUNCTION()
 
-CHECK_DEVICE_SPECIFIC_OPTIONS(DEVICE CUDA OPTIONS CUDA_UVM CUDA_RELOCATABLE_DEVICE_CODE CUDA_LAMBDA CUDA_CONSTEXPR CUDA_LDG_INTRINSIC)
+CHECK_DEVICE_SPECIFIC_OPTIONS(DEVICE CUDA OPTIONS CUDA_UVM CUDA_RELOCATABLE_DEVICE_CODE CUDA_LAMBDA CUDA_CONSTEXPR CUDA_LDG_INTRINSIC IMPL_CUDA_UNIFIED_MEMORY)
 CHECK_DEVICE_SPECIFIC_OPTIONS(DEVICE HIP OPTIONS HIP_RELOCATABLE_DEVICE_CODE)
 CHECK_DEVICE_SPECIFIC_OPTIONS(DEVICE HPX OPTIONS IMPL_HPX_ASYNC_DISPATCH)
 

lib/kokkos/cmake/kokkos_functions.cmake

@@ -709,7 +709,12 @@ MACRO(kokkos_find_imported NAME)
   ENDIF()
 
   IF (NOT TPL_LIBRARY_SUFFIXES)
-    SET(TPL_LIBRARY_SUFFIXES lib lib64)
+    SET(TPL_LIBRARY_SUFFIXES lib)
+    IF(KOKKOS_IMPL_32BIT)
+      LIST(APPEND TPL_LIBRARY_SUFFIXES lib32)
+    ELSE()
+      LIST(APPEND TPL_LIBRARY_SUFFIXES lib64)
+    ENDIF()
   ENDIF()
 
   SET(${NAME}_INCLUDE_DIRS)

lib/kokkos/cmake/kokkos_test_cxx_std.cmake

@@ -124,12 +124,8 @@ IF(KOKKOS_ENABLE_CUDA)
     ELSEIF(CMAKE_CXX_EXTENSIONS)
       MESSAGE(FATAL_ERROR "Compiling CUDA code with clang doesn't support C++ extensions.  Set -DCMAKE_CXX_EXTENSIONS=OFF")
     ENDIF()
-  ELSEIF(NOT KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA AND NOT (Kokkos_ENABLE_IMPL_NVHPC_AS_DEVICE_COMPILER AND KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC))
+  ELSEIF(NOT KOKKOS_CXX_COMPILER_ID STREQUAL NVIDIA)
-    IF(KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC)
+    MESSAGE(FATAL_ERROR "Invalid compiler for CUDA. The compiler must be nvcc_wrapper or Clang or use kokkos_launch_compiler, but compiler ID was ${KOKKOS_CXX_COMPILER_ID}")
-      MESSAGE(FATAL_ERROR "Invalid compiler for CUDA. To allow nvc++ as Cuda compiler, Kokkos_ENABLE_IMPL_NVHPC_AS_DEVICE_COMPILER=ON must be set!")
-    ELSE()
-      MESSAGE(FATAL_ERROR "Invalid compiler for CUDA. The compiler must be nvcc_wrapper or Clang or NVC++ or use kokkos_launch_compiler, but compiler ID was ${KOKKOS_CXX_COMPILER_ID}")
-    ENDIF()
   ENDIF()
 ENDIF()
 

lib/kokkos/cmake/kokkos_tpls.cmake

@@ -103,13 +103,19 @@ if (Kokkos_ENABLE_IMPL_MDSPAN AND Kokkos_ENABLE_MDSPAN_EXTERNAL)
 endif()
 
 IF (Kokkos_ENABLE_OPENMP)
-  find_package(OpenMP REQUIRED)
+  find_package(OpenMP REQUIRED COMPONENTS CXX)
   # FIXME_TRILINOS Trilinos doesn't allow for Kokkos to use find_dependency
   # so we just append the flags here instead of linking with the OpenMP target.
   IF(KOKKOS_HAS_TRILINOS)
     COMPILER_SPECIFIC_FLAGS(DEFAULT ${OpenMP_CXX_FLAGS})
   ELSE()
-    KOKKOS_EXPORT_CMAKE_TPL(OpenMP REQUIRED)
+    KOKKOS_EXPORT_CMAKE_TPL(OpenMP REQUIRED COMPONENTS CXX)
+  ENDIF()
+  IF(Kokkos_ENABLE_HIP AND KOKKOS_COMPILE_LANGUAGE STREQUAL HIP)
+    GLOBAL_APPEND(KOKKOS_AMDGPU_OPTIONS ${OpenMP_CXX_FLAGS})
+  ENDIF()
+  IF(Kokkos_ENABLE_CUDA AND KOKKOS_COMPILE_LANGUAGE STREQUAL CUDA)
+    GLOBAL_APPEND(KOKKOS_CUDA_OPTIONS -Xcompiler ${OpenMP_CXX_FLAGS})
   ENDIF()
 ENDIF()
 

lib/kokkos/cmake/kokkos_tribits.cmake

@@ -160,6 +160,12 @@ FUNCTION(KOKKOS_ADD_EXECUTABLE_AND_TEST ROOT_NAME)
             )
         ENDIF()
     ENDIF()
+    # We noticed problems with -fvisibility=hidden for inline static variables
+    # if Kokkos was built as shared library.
+    IF(BUILD_SHARED_LIBS)
+      SET_PROPERTY(TARGET ${PACKAGE_NAME}_${ROOT_NAME} PROPERTY VISIBILITY_INLINES_HIDDEN ON)
+      SET_PROPERTY(TARGET ${PACKAGE_NAME}_${ROOT_NAME} PROPERTY CXX_VISIBILITY_PRESET hidden)
+    ENDIF()
 ENDFUNCTION()
 
 FUNCTION(KOKKOS_SET_EXE_PROPERTY ROOT_NAME)
@@ -241,34 +247,6 @@ MACRO(KOKKOS_CONFIGURE_CORE)
    KOKKOS_CONFIG_HEADER( KokkosCore_Config_HeaderSet.in KokkosCore_Config_FwdBackend.hpp "KOKKOS_FWD" "fwd/Kokkos_Fwd" "${KOKKOS_ENABLED_DEVICES}")
    KOKKOS_CONFIG_HEADER( KokkosCore_Config_HeaderSet.in KokkosCore_Config_SetupBackend.hpp "KOKKOS_SETUP" "setup/Kokkos_Setup" "${DEVICE_SETUP_LIST}")
    KOKKOS_CONFIG_HEADER( KokkosCore_Config_HeaderSet.in KokkosCore_Config_DeclareBackend.hpp "KOKKOS_DECLARE" "decl/Kokkos_Declare" "${KOKKOS_ENABLED_DEVICES}")
-   SET(_DEFAULT_HOST_MEMSPACE "::Kokkos::HostSpace")
-   KOKKOS_OPTION(DEFAULT_DEVICE_MEMORY_SPACE "" STRING "Override default device memory space")
-   KOKKOS_OPTION(DEFAULT_HOST_MEMORY_SPACE "" STRING "Override default host memory space")
-   KOKKOS_OPTION(DEFAULT_DEVICE_EXECUTION_SPACE "" STRING "Override default device execution space")
-   KOKKOS_OPTION(DEFAULT_HOST_PARALLEL_EXECUTION_SPACE "" STRING "Override default host parallel execution space")
-   IF (NOT Kokkos_DEFAULT_DEVICE_EXECUTION_SPACE STREQUAL "")
-      SET(_DEVICE_PARALLEL ${Kokkos_DEFAULT_DEVICE_EXECUTION_SPACE})
-      MESSAGE(STATUS "Override default device execution space: ${_DEVICE_PARALLEL}")
-      SET(KOKKOS_DEVICE_SPACE_ACTIVE ON)
-   ELSE()
-      IF (_DEVICE_PARALLEL STREQUAL "NoTypeDefined")
-         SET(KOKKOS_DEVICE_SPACE_ACTIVE OFF)
-      ELSE()
-         SET(KOKKOS_DEVICE_SPACE_ACTIVE ON)
-      ENDIF()
-   ENDIF()
-   IF (NOT Kokkos_DEFAULT_HOST_PARALLEL_EXECUTION_SPACE STREQUAL "")
-      SET(_HOST_PARALLEL ${Kokkos_DEFAULT_HOST_PARALLEL_EXECUTION_SPACE})
-      MESSAGE(STATUS "Override default host parallel execution space: ${_HOST_PARALLEL}")
-      SET(KOKKOS_HOSTPARALLEL_SPACE_ACTIVE ON)
-   ELSE()
-      IF (_HOST_PARALLEL STREQUAL "NoTypeDefined")
-         SET(KOKKOS_HOSTPARALLEL_SPACE_ACTIVE OFF)
-      ELSE()
-         SET(KOKKOS_HOSTPARALLEL_SPACE_ACTIVE ON)
-      ENDIF()
-   ENDIF()
-   #We are ready to configure the header
    CONFIGURE_FILE(cmake/KokkosCore_config.h.in KokkosCore_config.h @ONLY)
 ENDMACRO()
 
@@ -484,15 +462,10 @@ ENDFUNCTION()
 
 
 FUNCTION(KOKKOS_LIB_INCLUDE_DIRECTORIES TARGET)
-  IF(KOKKOS_HAS_TRILINOS)
-    #ignore the target, tribits doesn't do anything directly with targets
-    TRIBITS_INCLUDE_DIRECTORIES(${ARGN})
-  ELSE() #append to a list for later
   KOKKOS_LIB_TYPE(${TARGET} INCTYPE)
   FOREACH(DIR ${ARGN})
     TARGET_INCLUDE_DIRECTORIES(${TARGET} ${INCTYPE} $<BUILD_INTERFACE:${DIR}>)
   ENDFOREACH()
-  ENDIF()
 ENDFUNCTION()
 
 FUNCTION(KOKKOS_LIB_COMPILE_OPTIONS TARGET)

lib/kokkos/cmake/tpls/FindTPLCUSPARSE.cmake

@@ -1,26 +0,0 @@
-#@HEADER
-# ************************************************************************
-#
-#                        Kokkos v. 4.0
-#       Copyright (2022) National Technology & Engineering
-#               Solutions of Sandia, LLC (NTESS).
-#
-# Under the terms of Contract DE-NA0003525 with NTESS,
-# the U.S. Government retains certain rights in this software.
-#
-# Part of Kokkos, under the Apache License v2.0 with LLVM Exceptions.
-#
-# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-#
-#@HEADER
-
-# Check for CUDA support
-
-IF (NOT TPL_ENABLE_CUDA)
-  MESSAGE(FATAL_ERROR "\nCUSPARSE requires CUDA")
-ELSE()
-  GLOBAL_SET(TPL_CUSPARSE_LIBRARY_DIRS)
-  GLOBAL_SET(TPL_CUSPARSE_INCLUDE_DIRS ${TPL_CUDA_INCLUDE_DIRS})
-  GLOBAL_SET(TPL_CUSPARSE_LIBRARIES    ${CUDA_cusparse_LIBRARY})
-ENDIF()
-

lib/kokkos/containers/src/Kokkos_DualView.hpp

@@ -944,13 +944,13 @@ class DualView : public ViewTraits<DataType, Properties...> {
 
     if (sizeMismatch) {
       ::Kokkos::realloc(arg_prop, d_view, n0, n1, n2, n3, n4, n5, n6, n7);
-      if (alloc_prop_input::initialize) {
+      if constexpr (alloc_prop_input::initialize) {
         h_view = create_mirror_view(typename t_host::memory_space(), d_view);
       } else {
         h_view = create_mirror_view(Kokkos::WithoutInitializing,
                                     typename t_host::memory_space(), d_view);
       }
-    } else if (alloc_prop_input::initialize) {
+    } else if constexpr (alloc_prop_input::initialize) {
       if constexpr (alloc_prop_input::has_execution_space) {
         const auto& exec_space =
             Impl::get_property<Impl::ExecutionSpaceTag>(arg_prop);
@@ -1038,12 +1038,10 @@ class DualView : public ViewTraits<DataType, Properties...> {
       /* Resize on Device */
       if (sizeMismatch) {
         ::Kokkos::resize(properties, d_view, n0, n1, n2, n3, n4, n5, n6, n7);
-        if (alloc_prop_input::initialize) {
+        // this part of the lambda was relocated in a method as it contains a
-          h_view = create_mirror_view(typename t_host::memory_space(), d_view);
+        // `if constexpr`. In some cases, both branches were evaluated
-        } else {
+        // leading to a compile error
-          h_view = create_mirror_view(Kokkos::WithoutInitializing,
+        resync_host(properties);
-                                      typename t_host::memory_space(), d_view);
-        }
 
         /* Mark Device copy as modified */
         ++modified_flags(1);
@@ -1054,13 +1052,10 @@ class DualView : public ViewTraits<DataType, Properties...> {
       /* Resize on Host */
       if (sizeMismatch) {
         ::Kokkos::resize(properties, h_view, n0, n1, n2, n3, n4, n5, n6, n7);
-        if (alloc_prop_input::initialize) {
+        // this part of the lambda was relocated in a method as it contains a
-          d_view = create_mirror_view(typename t_dev::memory_space(), h_view);
+        // `if constexpr`. In some cases, both branches were evaluated
-
+        // leading to a compile error
-        } else {
+        resync_device(properties);
-          d_view = create_mirror_view(Kokkos::WithoutInitializing,
-                                      typename t_dev::memory_space(), h_view);
-        }
 
         /* Mark Host copy as modified */
         ++modified_flags(0);
@@ -1099,6 +1094,39 @@ class DualView : public ViewTraits<DataType, Properties...> {
     }
   }
 
+ private:
+  // resync host mirror from device
+  // this code was relocated from a lambda as it contains a `if constexpr`.
+  // In some cases, both branches were evaluated, leading to a compile error
+  template <class... ViewCtorArgs>
+  inline void resync_host(Impl::ViewCtorProp<ViewCtorArgs...> const&) {
+    using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
+
+    if constexpr (alloc_prop_input::initialize) {
+      h_view = create_mirror_view(typename t_host::memory_space(), d_view);
+    } else {
+      h_view = create_mirror_view(Kokkos::WithoutInitializing,
+                                  typename t_host::memory_space(), d_view);
+    }
+  }
+
+  // resync device mirror from host
+  // this code was relocated from a lambda as it contains a `if constexpr`
+  // In some cases, both branches were evaluated leading to a compile error
+  template <class... ViewCtorArgs>
+  inline void resync_device(Impl::ViewCtorProp<ViewCtorArgs...> const&) {
+    using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
+
+    if constexpr (alloc_prop_input::initialize) {
+      d_view = create_mirror_view(typename t_dev::memory_space(), h_view);
+
+    } else {
+      d_view = create_mirror_view(Kokkos::WithoutInitializing,
+                                  typename t_dev::memory_space(), h_view);
+    }
+  }
+
+ public:
   void resize(const size_t n0 = KOKKOS_IMPL_CTOR_DEFAULT_ARG,
               const size_t n1 = KOKKOS_IMPL_CTOR_DEFAULT_ARG,
               const size_t n2 = KOKKOS_IMPL_CTOR_DEFAULT_ARG,

lib/kokkos/containers/src/Kokkos_DynRankView.hpp

@@ -1657,8 +1657,7 @@ KOKKOS_FUNCTION auto as_view_of_rank_n(
 
   if constexpr (std::is_same_v<decltype(layout), Kokkos::LayoutLeft> ||
                 std::is_same_v<decltype(layout), Kokkos::LayoutRight> ||
-                std::is_same_v<decltype(layout), Kokkos::LayoutStride> ||
+                std::is_same_v<decltype(layout), Kokkos::LayoutStride>) {
-                is_layouttiled<decltype(layout)>::value) {
     for (int i = N; i < 7; ++i)
       layout.dimension[i] = KOKKOS_IMPL_CTOR_DEFAULT_ARG;
   }
@@ -1933,254 +1932,155 @@ struct MirrorDRVType {
 }  // namespace Impl
 
 namespace Impl {
+
+// create a mirror
+// private interface that accepts arbitrary view constructor args passed by a
+// view_alloc
 template <class T, class... P, class... ViewCtorArgs>
-inline typename DynRankView<T, P...>::HostMirror create_mirror(
+inline auto create_mirror(const DynRankView<T, P...>& src,
-    const DynRankView<T, P...>& src,
+                          const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
+  check_view_ctor_args_create_mirror<ViewCtorArgs...>();
-    std::enable_if_t<!Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* =
-        nullptr) {
-  using src_type = DynRankView<T, P...>;
-  using dst_type = typename src_type::HostMirror;
-
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-
-  static_assert(
-      !alloc_prop_input::has_label,
-      "The view constructor arguments passed to Kokkos::create_mirror "
-      "must not include a label!");
-  static_assert(
-      !alloc_prop_input::has_pointer,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not include a pointer!");
-  static_assert(
-      !alloc_prop_input::allow_padding,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not explicitly allow padding!");
 
   auto prop_copy = Impl::with_properties_if_unset(
       arg_prop, std::string(src.label()).append("_mirror"));
 
-  return dst_type(prop_copy, Impl::reconstructLayout(src.layout(), src.rank()));
+  if constexpr (Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space) {
-}
-
-template <class T, class... P, class... ViewCtorArgs>
-inline auto create_mirror(
-    const DynRankView<T, P...>& src,
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    std::enable_if_t<Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* =
-        nullptr) {
     using dst_type = typename Impl::MirrorDRVType<
         typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
         P...>::view_type;
 
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
+    return dst_type(prop_copy,
+                    Impl::reconstructLayout(src.layout(), src.rank()));
+  } else {
+    using src_type = DynRankView<T, P...>;
+    using dst_type = typename src_type::HostMirror;
 
-  static_assert(
+    return dst_type(prop_copy,
-      !alloc_prop_input::has_label,
+                    Impl::reconstructLayout(src.layout(), src.rank()));
-      "The view constructor arguments passed to Kokkos::create_mirror "
+  }
-      "must not include a label!");
+#if defined(KOKKOS_COMPILER_INTEL) ||                                 \
-  static_assert(
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
-      !alloc_prop_input::has_pointer,
+     !defined(KOKKOS_COMPILER_MSVC))
-      "The view constructor arguments passed to Kokkos::create_mirror must "
+  __builtin_unreachable();
-      "not include a pointer!");
+#endif
-  static_assert(
-      !alloc_prop_input::allow_padding,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not explicitly allow padding!");
-
-  auto prop_copy = Impl::with_properties_if_unset(
-      arg_prop, std::string(src.label()).append("_mirror"));
-
-  return dst_type(prop_copy, Impl::reconstructLayout(src.layout(), src.rank()));
 }
 
 }  // namespace Impl
 
-// Create a mirror in host space
+// public interface
-template <class T, class... P>
+template <class T, class... P,
-inline typename DynRankView<T, P...>::HostMirror create_mirror(
+          class Enable = std::enable_if_t<
-    const DynRankView<T, P...>& src,
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
-    std::enable_if_t<std::is_same<typename ViewTraits<T, P...>::specialize,
+inline auto create_mirror(const DynRankView<T, P...>& src) {
-                                  void>::value>* = nullptr) {
+  return Impl::create_mirror(src, Kokkos::view_alloc());
-  return Impl::create_mirror(src, Kokkos::Impl::ViewCtorProp<>{});
 }
 
-template <class T, class... P>
+// public interface that accepts a without initializing flag
-inline typename DynRankView<T, P...>::HostMirror create_mirror(
+template <class T, class... P,
-    Kokkos::Impl::WithoutInitializing_t wi, const DynRankView<T, P...>& src,
+          class Enable = std::enable_if_t<
-    std::enable_if_t<std::is_same<typename ViewTraits<T, P...>::specialize,
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
-                                  void>::value>* = nullptr) {
+inline auto create_mirror(Kokkos::Impl::WithoutInitializing_t wi,
+                          const DynRankView<T, P...>& src) {
   return Impl::create_mirror(src, Kokkos::view_alloc(wi));
 }
 
-template <class T, class... P, class... ViewCtorArgs>
+// public interface that accepts a space
-inline typename DynRankView<T, P...>::HostMirror create_mirror(
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    const DynRankView<T, P...>& src,
-    std::enable_if_t<
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
-        !Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* = nullptr) {
-  return Impl::create_mirror(src, arg_prop);
-}
-
-// Create a mirror in a new space
 template <class Space, class T, class... P,
-          typename Enable = std::enable_if_t<
+          class Enable = std::enable_if_t<
               Kokkos::is_space<Space>::value &&
-              std::is_void<typename ViewTraits<T, P...>::specialize>::value>>
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
-typename Impl::MirrorDRVType<Space, T, P...>::view_type create_mirror(
+auto create_mirror(const Space&, const Kokkos::DynRankView<T, P...>& src) {
-    const Space&, const Kokkos::DynRankView<T, P...>& src) {
   return Impl::create_mirror(
       src, Kokkos::view_alloc(typename Space::memory_space{}));
 }
 
-template <class Space, class T, class... P>
+// public interface that accepts a space and a without initializing flag
-typename Impl::MirrorDRVType<Space, T, P...>::view_type create_mirror(
+template <class Space, class T, class... P,
-    Kokkos::Impl::WithoutInitializing_t wi, const Space&,
+          class Enable = std::enable_if_t<
-    const Kokkos::DynRankView<T, P...>& src,
+              Kokkos::is_space<Space>::value &&
-    std::enable_if_t<std::is_same<typename ViewTraits<T, P...>::specialize,
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
-                                  void>::value>* = nullptr) {
+auto create_mirror(Kokkos::Impl::WithoutInitializing_t wi, const Space&,
+                   const Kokkos::DynRankView<T, P...>& src) {
   return Impl::create_mirror(
       src, Kokkos::view_alloc(wi, typename Space::memory_space{}));
 }
 
-template <class T, class... P, class... ViewCtorArgs>
+// public interface that accepts arbitrary view constructor args passed by a
-inline auto create_mirror(
+// view_alloc
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
+template <class T, class... P, class... ViewCtorArgs,
-    const DynRankView<T, P...>& src,
+          typename Enable = std::enable_if_t<
-    std::enable_if_t<
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
+inline auto create_mirror(const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-        Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* = nullptr) {
+                          const DynRankView<T, P...>& src) {
-  using ReturnType = typename Impl::MirrorDRVType<
+  return Impl::create_mirror(src, arg_prop);
-      typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-      P...>::view_type;
-  return ReturnType{Impl::create_mirror(src, arg_prop)};
 }
 
 namespace Impl {
-template <class T, class... P, class... ViewCtorArgs>
-inline std::enable_if_t<
-    !Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space &&
-        std::is_same<
-            typename DynRankView<T, P...>::memory_space,
-            typename DynRankView<T, P...>::HostMirror::memory_space>::value &&
-        std::is_same<
-            typename DynRankView<T, P...>::data_type,
-            typename DynRankView<T, P...>::HostMirror::data_type>::value,
-    typename DynRankView<T, P...>::HostMirror>
-create_mirror_view(const DynRankView<T, P...>& src,
-                   const typename Impl::ViewCtorProp<ViewCtorArgs...>&) {
-  return src;
-}
 
+// create a mirror view
+// private interface that accepts arbitrary view constructor args passed by a
+// view_alloc
 template <class T, class... P, class... ViewCtorArgs>
-inline std::enable_if_t<
+inline auto create_mirror_view(
-    !Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space &&
-        !(std::is_same<
-              typename DynRankView<T, P...>::memory_space,
-              typename DynRankView<T, P...>::HostMirror::memory_space>::value &&
-          std::is_same<
-              typename DynRankView<T, P...>::data_type,
-              typename DynRankView<T, P...>::HostMirror::data_type>::value),
-    typename DynRankView<T, P...>::HostMirror>
-create_mirror_view(
     const DynRankView<T, P...>& src,
-    const typename Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
+    [[maybe_unused]] const typename Impl::ViewCtorProp<ViewCtorArgs...>&
-  return Kokkos::Impl::create_mirror(src, arg_prop);
+        arg_prop) {
+  if constexpr (!Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space) {
+    if constexpr (std::is_same<typename DynRankView<T, P...>::memory_space,
+                               typename DynRankView<
+                                   T, P...>::HostMirror::memory_space>::value &&
+                  std::is_same<typename DynRankView<T, P...>::data_type,
+                               typename DynRankView<
+                                   T, P...>::HostMirror::data_type>::value) {
+      return typename DynRankView<T, P...>::HostMirror(src);
+    } else {
+      return Kokkos::Impl::choose_create_mirror(src, arg_prop);
+    }
+  } else {
+    if constexpr (Impl::MirrorDRViewType<typename Impl::ViewCtorProp<
+                                             ViewCtorArgs...>::memory_space,
+                                         T, P...>::is_same_memspace) {
+      return typename Impl::MirrorDRViewType<
+          typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
+          P...>::view_type(src);
+    } else {
+      return Kokkos::Impl::choose_create_mirror(src, arg_prop);
+    }
+  }
+#if defined(KOKKOS_COMPILER_INTEL) ||                                 \
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+     !defined(KOKKOS_COMPILER_MSVC))
+  __builtin_unreachable();
+#endif
 }
 
-template <class T, class... P, class... ViewCtorArgs,
-          class = std::enable_if_t<
-              Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>>
-inline std::enable_if_t<
-    Kokkos::is_space<
-        typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space>::value &&
-        Impl::MirrorDRViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace,
-    typename Impl::MirrorDRViewType<
-        typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-        P...>::view_type>
-create_mirror_view(const Kokkos::DynRankView<T, P...>& src,
-                   const typename Impl::ViewCtorProp<ViewCtorArgs...>&) {
-  return src;
-}
-
-template <class T, class... P, class... ViewCtorArgs,
-          class = std::enable_if_t<
-              Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>>
-inline std::enable_if_t<
-    Kokkos::is_space<
-        typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space>::value &&
-        !Impl::MirrorDRViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace,
-    typename Impl::MirrorDRViewType<
-        typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-        P...>::view_type>
-create_mirror_view(
-    const Kokkos::DynRankView<T, P...>& src,
-    const typename Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-  return Kokkos::Impl::create_mirror(src, arg_prop);
-}
 }  // namespace Impl
 
-// Create a mirror view in host space
+// public interface
 template <class T, class... P>
-inline std::enable_if_t<
+inline auto create_mirror_view(const Kokkos::DynRankView<T, P...>& src) {
-    (std::is_same<
+  return Impl::create_mirror_view(src, Kokkos::view_alloc());
-         typename DynRankView<T, P...>::memory_space,
-         typename DynRankView<T, P...>::HostMirror::memory_space>::value &&
-     std::is_same<typename DynRankView<T, P...>::data_type,
-                  typename DynRankView<T, P...>::HostMirror::data_type>::value),
-    typename DynRankView<T, P...>::HostMirror>
-create_mirror_view(const Kokkos::DynRankView<T, P...>& src) {
-  return src;
-}
-
-template <class T, class... P>
-inline std::enable_if_t<
-    !(std::is_same<
-          typename DynRankView<T, P...>::memory_space,
-          typename DynRankView<T, P...>::HostMirror::memory_space>::value &&
-      std::is_same<
-          typename DynRankView<T, P...>::data_type,
-          typename DynRankView<T, P...>::HostMirror::data_type>::value),
-    typename DynRankView<T, P...>::HostMirror>
-create_mirror_view(const Kokkos::DynRankView<T, P...>& src) {
-  return Kokkos::create_mirror(src);
 }
 
+// public interface that accepts a without initializing flag
 template <class T, class... P>
 inline auto create_mirror_view(Kokkos::Impl::WithoutInitializing_t wi,
                                const DynRankView<T, P...>& src) {
   return Impl::create_mirror_view(src, Kokkos::view_alloc(wi));
 }
 
-// Create a mirror view in a new space
+// public interface that accepts a space
-// FIXME_C++17 Improve SFINAE here.
 template <class Space, class T, class... P,
           class Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
-inline typename Impl::MirrorDRViewType<Space, T, P...>::view_type
+inline auto create_mirror_view(const Space&,
-create_mirror_view(
+                               const Kokkos::DynRankView<T, P...>& src) {
-    const Space&, const Kokkos::DynRankView<T, P...>& src,
+  return Impl::create_mirror_view(
-    std::enable_if_t<
+      src, Kokkos::view_alloc(typename Space::memory_space()));
-        Impl::MirrorDRViewType<Space, T, P...>::is_same_memspace>* = nullptr) {
-  return src;
 }
 
-// FIXME_C++17 Improve SFINAE here.
+// public interface that accepts a space and a without initializing flag
 template <class Space, class T, class... P,
-          class Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
+          typename Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
-inline typename Impl::MirrorDRViewType<Space, T, P...>::view_type
-create_mirror_view(
-    const Space& space, const Kokkos::DynRankView<T, P...>& src,
-    std::enable_if_t<
-        !Impl::MirrorDRViewType<Space, T, P...>::is_same_memspace>* = nullptr) {
-  return Kokkos::create_mirror(space, src);
-}
-
-template <class Space, class T, class... P>
 inline auto create_mirror_view(Kokkos::Impl::WithoutInitializing_t wi,
                                const Space&,
                                const Kokkos::DynRankView<T, P...>& src) {
@@ -2188,6 +2088,8 @@ inline auto create_mirror_view(Kokkos::Impl::WithoutInitializing_t wi,
       src, Kokkos::view_alloc(typename Space::memory_space{}, wi));
 }
 
+// public interface that accepts arbitrary view constructor args passed by a
+// view_alloc
 template <class T, class... P, class... ViewCtorArgs>
 inline auto create_mirror_view(
     const typename Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
@@ -2195,58 +2097,29 @@ inline auto create_mirror_view(
   return Impl::create_mirror_view(src, arg_prop);
 }
 
-template <class... ViewCtorArgs, class T, class... P>
+// create a mirror view and deep copy it
+// public interface that accepts arbitrary view constructor args passed by a
+// view_alloc
+template <class... ViewCtorArgs, class T, class... P,
+          class Enable = std::enable_if_t<
+              std::is_void<typename ViewTraits<T, P...>::specialize>::value>>
 auto create_mirror_view_and_copy(
-    const Impl::ViewCtorProp<ViewCtorArgs...>&,
+    [[maybe_unused]] const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    const Kokkos::DynRankView<T, P...>& src,
+    const Kokkos::DynRankView<T, P...>& src) {
-    std::enable_if_t<
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
-        Impl::MirrorDRViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace>* = nullptr) {
   using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-  static_assert(
-      alloc_prop_input::has_memory_space,
-      "The view constructor arguments passed to "
-      "Kokkos::create_mirror_view_and_copy must include a memory space!");
-  static_assert(!alloc_prop_input::has_pointer,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not include a pointer!");
-  static_assert(!alloc_prop_input::allow_padding,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not explicitly allow padding!");
 
+  Impl::check_view_ctor_args_create_mirror_view_and_copy<ViewCtorArgs...>();
+
+  if constexpr (Impl::MirrorDRViewType<
+                    typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
+                    T, P...>::is_same_memspace) {
     // same behavior as deep_copy(src, src)
-  if (!alloc_prop_input::has_execution_space)
+    if constexpr (!alloc_prop_input::has_execution_space)
       fence(
-        "Kokkos::create_mirror_view_and_copy: fence before returning src view");
+          "Kokkos::create_mirror_view_and_copy: fence before returning src "
+          "view");
     return src;
-}
+  } else {
-
-template <class... ViewCtorArgs, class T, class... P>
-auto create_mirror_view_and_copy(
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    const Kokkos::DynRankView<T, P...>& src,
-    std::enable_if_t<
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
-        !Impl::MirrorDRViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace>* = nullptr) {
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-  static_assert(
-      alloc_prop_input::has_memory_space,
-      "The view constructor arguments passed to "
-      "Kokkos::create_mirror_view_and_copy must include a memory space!");
-  static_assert(!alloc_prop_input::has_pointer,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not include a pointer!");
-  static_assert(!alloc_prop_input::allow_padding,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not explicitly allow padding!");
     using Space  = typename alloc_prop_input::memory_space;
     using Mirror = typename Impl::MirrorDRViewType<Space, T, P...>::view_type;
 
@@ -2265,6 +2138,11 @@ auto create_mirror_view_and_copy(
       deep_copy(mirror, src);
     return mirror;
   }
+#if defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+    !defined(KOKKOS_COMPILER_MSVC)
+  __builtin_unreachable();
+#endif
+}
 
 template <class Space, class T, class... P>
 auto create_mirror_view_and_copy(const Space&,

lib/kokkos/containers/src/Kokkos_DynamicView.hpp

@@ -590,62 +590,21 @@ struct MirrorDynamicViewType {
 }  // namespace Impl
 
 namespace Impl {
-template <class T, class... P, class... ViewCtorArgs>
-inline auto create_mirror(
-    const Kokkos::Experimental::DynamicView<T, P...>& src,
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    std::enable_if_t<!Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* =
-        nullptr) {
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
 
-  static_assert(
+// create a mirror
-      !alloc_prop_input::has_label,
+// private interface that accepts arbitrary view constructor args passed by a
-      "The view constructor arguments passed to Kokkos::create_mirror "
+// view_alloc
-      "must not include a label!");
+template <class T, class... P, class... ViewCtorArgs>
-  static_assert(
+inline auto create_mirror(const Kokkos::Experimental::DynamicView<T, P...>& src,
-      !alloc_prop_input::has_pointer,
+                          const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-      "The view constructor arguments passed to Kokkos::create_mirror must "
+  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-      "not include a pointer!");
+  check_view_ctor_args_create_mirror<ViewCtorArgs...>();
-  static_assert(
-      !alloc_prop_input::allow_padding,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not explicitly allow padding!");
 
   auto prop_copy = Impl::with_properties_if_unset(
       arg_prop, std::string(src.label()).append("_mirror"));
 
-  auto ret = typename Kokkos::Experimental::DynamicView<T, P...>::HostMirror(
+  if constexpr (Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space) {
-      prop_copy, src.chunk_size(), src.chunk_max() * src.chunk_size());
-
-  ret.resize_serial(src.extent(0));
-
-  return ret;
-}
-
-template <class T, class... P, class... ViewCtorArgs>
-inline auto create_mirror(
-    const Kokkos::Experimental::DynamicView<T, P...>& src,
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    std::enable_if_t<Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>* =
-        nullptr) {
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-
-  static_assert(
-      !alloc_prop_input::has_label,
-      "The view constructor arguments passed to Kokkos::create_mirror "
-      "must not include a label!");
-  static_assert(
-      !alloc_prop_input::has_pointer,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not include a pointer!");
-  static_assert(
-      !alloc_prop_input::allow_padding,
-      "The view constructor arguments passed to Kokkos::create_mirror must "
-      "not explicitly allow padding!");
-
     using MemorySpace = typename alloc_prop_input::memory_space;
-  auto prop_copy    = Impl::with_properties_if_unset(
-      arg_prop, std::string(src.label()).append("_mirror"));
 
     auto ret = typename Kokkos::Impl::MirrorDynamicViewType<
         MemorySpace, T, P...>::view_type(prop_copy, src.chunk_size(),
@@ -653,33 +612,59 @@ inline auto create_mirror(
 
     ret.resize_serial(src.extent(0));
 
+    return ret;
+  } else {
+    auto ret = typename Kokkos::Experimental::DynamicView<T, P...>::HostMirror(
+        prop_copy, src.chunk_size(), src.chunk_max() * src.chunk_size());
+
+    ret.resize_serial(src.extent(0));
+
     return ret;
   }
+#if defined(KOKKOS_COMPILER_INTEL) ||                                 \
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+     !defined(KOKKOS_COMPILER_MSVC))
+  __builtin_unreachable();
+#endif
+}
+
 }  // namespace Impl
 
-// Create a mirror in host space
+// public interface
-template <class T, class... P>
+template <class T, class... P,
+          typename Enable = std::enable_if_t<
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
 inline auto create_mirror(
     const Kokkos::Experimental::DynamicView<T, P...>& src) {
   return Impl::create_mirror(src, Impl::ViewCtorProp<>{});
 }
 
-template <class T, class... P>
+// public interface that accepts a without initializing flag
+template <class T, class... P,
+          typename Enable = std::enable_if_t<
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
 inline auto create_mirror(
     Kokkos::Impl::WithoutInitializing_t wi,
     const Kokkos::Experimental::DynamicView<T, P...>& src) {
   return Impl::create_mirror(src, Kokkos::view_alloc(wi));
 }
 
-// Create a mirror in a new space
+// public interface that accepts a space
-template <class Space, class T, class... P>
+template <class Space, class T, class... P,
+          typename Enable = std::enable_if_t<
+              Kokkos::is_space<Space>::value &&
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
 inline auto create_mirror(
     const Space&, const Kokkos::Experimental::DynamicView<T, P...>& src) {
   return Impl::create_mirror(
       src, Kokkos::view_alloc(typename Space::memory_space{}));
 }
 
-template <class Space, class T, class... P>
+// public interface that accepts a space and a without initializing flag
+template <class Space, class T, class... P,
+          typename Enable = std::enable_if_t<
+              Kokkos::is_space<Space>::value &&
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
 typename Kokkos::Impl::MirrorDynamicViewType<Space, T, P...>::view_type
 create_mirror(Kokkos::Impl::WithoutInitializing_t wi, const Space&,
               const Kokkos::Experimental::DynamicView<T, P...>& src) {
@@ -687,7 +672,11 @@ create_mirror(Kokkos::Impl::WithoutInitializing_t wi, const Space&,
       src, Kokkos::view_alloc(wi, typename Space::memory_space{}));
 }
 
-template <class T, class... P, class... ViewCtorArgs>
+// public interface that accepts arbitrary view constructor args passed by a
+// view_alloc
+template <class T, class... P, class... ViewCtorArgs,
+          typename Enable = std::enable_if_t<
+              std::is_void_v<typename ViewTraits<T, P...>::specialize>>>
 inline auto create_mirror(
     const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
     const Kokkos::Experimental::DynamicView<T, P...>& src) {
@@ -696,76 +685,56 @@ inline auto create_mirror(
 
 namespace Impl {
 
+// create a mirror view
+// private interface that accepts arbitrary view constructor args passed by a
+// view_alloc
 template <class T, class... P, class... ViewCtorArgs>
-inline std::enable_if_t<
+inline auto create_mirror_view(
-    !Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space &&
+    const Kokkos::Experimental::DynamicView<T, P...>& src,
-        (std::is_same<
+    [[maybe_unused]] const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-             typename Kokkos::Experimental::DynamicView<T, P...>::memory_space,
+  if constexpr (!Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space) {
+    if constexpr (std::is_same<typename Kokkos::Experimental::DynamicView<
+                                   T, P...>::memory_space,
                                typename Kokkos::Experimental::DynamicView<
                                    T, P...>::HostMirror::memory_space>::value &&
-         std::is_same<
+                  std::is_same<typename Kokkos::Experimental::DynamicView<
-             typename Kokkos::Experimental::DynamicView<T, P...>::data_type,
+                                   T, P...>::data_type,
                                typename Kokkos::Experimental::DynamicView<
-                 T, P...>::HostMirror::data_type>::value),
+                                   T, P...>::HostMirror::data_type>::value) {
-    typename Kokkos::Experimental::DynamicView<T, P...>::HostMirror>
+      return
-create_mirror_view(const Kokkos::Experimental::DynamicView<T, P...>& src,
+          typename Kokkos::Experimental::DynamicView<T, P...>::HostMirror(src);
-                   const Impl::ViewCtorProp<ViewCtorArgs...>&) {
+    } else {
-  return src;
+      return Kokkos::Impl::choose_create_mirror(src, arg_prop);
+    }
+  } else {
+    if constexpr (Impl::MirrorDynamicViewType<
+                      typename Impl::ViewCtorProp<
+                          ViewCtorArgs...>::memory_space,
+                      T, P...>::is_same_memspace) {
+      return typename Impl::MirrorDynamicViewType<
+          typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
+          P...>::view_type(src);
+    } else {
+      return Kokkos::Impl::choose_create_mirror(src, arg_prop);
+    }
+  }
+#if defined(KOKKOS_COMPILER_INTEL) ||                                 \
+    (defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+     !defined(KOKKOS_COMPILER_MSVC))
+  __builtin_unreachable();
+#endif
 }
 
-template <class T, class... P, class... ViewCtorArgs>
-inline std::enable_if_t<
-    !Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space &&
-        !(std::is_same<
-              typename Kokkos::Experimental::DynamicView<T, P...>::memory_space,
-              typename Kokkos::Experimental::DynamicView<
-                  T, P...>::HostMirror::memory_space>::value &&
-          std::is_same<
-              typename Kokkos::Experimental::DynamicView<T, P...>::data_type,
-              typename Kokkos::Experimental::DynamicView<
-                  T, P...>::HostMirror::data_type>::value),
-    typename Kokkos::Experimental::DynamicView<T, P...>::HostMirror>
-create_mirror_view(const Kokkos::Experimental::DynamicView<T, P...>& src,
-                   const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-  return Kokkos::create_mirror(arg_prop, src);
-}
-
-template <class T, class... P, class... ViewCtorArgs,
-          class = std::enable_if_t<
-              Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>>
-std::enable_if_t<Impl::MirrorDynamicViewType<
-                     typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
-                     T, P...>::is_same_memspace,
-                 typename Impl::MirrorDynamicViewType<
-                     typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
-                     T, P...>::view_type>
-create_mirror_view(const Kokkos::Experimental::DynamicView<T, P...>& src,
-                   const Impl::ViewCtorProp<ViewCtorArgs...>&) {
-  return src;
-}
-
-template <class T, class... P, class... ViewCtorArgs,
-          class = std::enable_if_t<
-              Impl::ViewCtorProp<ViewCtorArgs...>::has_memory_space>>
-std::enable_if_t<!Impl::MirrorDynamicViewType<
-                     typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
-                     T, P...>::is_same_memspace,
-                 typename Impl::MirrorDynamicViewType<
-                     typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
-                     T, P...>::view_type>
-create_mirror_view(const Kokkos::Experimental::DynamicView<T, P...>& src,
-                   const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop) {
-  return Kokkos::Impl::create_mirror(src, arg_prop);
-}
 }  // namespace Impl
 
-// Create a mirror view in host space
+// public interface
 template <class T, class... P>
 inline auto create_mirror_view(
     const typename Kokkos::Experimental::DynamicView<T, P...>& src) {
   return Impl::create_mirror_view(src, Impl::ViewCtorProp<>{});
 }
 
+// public interface that accepts a without initializing flag
 template <class T, class... P>
 inline auto create_mirror_view(
     Kokkos::Impl::WithoutInitializing_t wi,
@@ -773,15 +742,18 @@ inline auto create_mirror_view(
   return Impl::create_mirror_view(src, Kokkos::view_alloc(wi));
 }
 
-// Create a mirror in a new space
+// public interface that accepts a space
-template <class Space, class T, class... P>
+template <class Space, class T, class... P,
+          class Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
 inline auto create_mirror_view(
     const Space&, const Kokkos::Experimental::DynamicView<T, P...>& src) {
   return Impl::create_mirror_view(src,
                                   view_alloc(typename Space::memory_space{}));
 }
 
-template <class Space, class T, class... P>
+// public interface that accepts a space and a without initializing flag
+template <class Space, class T, class... P,
+          class Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
 inline auto create_mirror_view(
     Kokkos::Impl::WithoutInitializing_t wi, const Space&,
     const Kokkos::Experimental::DynamicView<T, P...>& src) {
@@ -789,6 +761,8 @@ inline auto create_mirror_view(
       src, Kokkos::view_alloc(wi, typename Space::memory_space{}));
 }
 
+// public interface that accepts arbitrary view constructor args passed by a
+// view_alloc
 template <class T, class... P, class... ViewCtorArgs>
 inline auto create_mirror_view(
     const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
@@ -985,58 +959,29 @@ struct ViewCopy<Kokkos::Experimental::DynamicView<DP...>,
 
 }  // namespace Impl
 
-template <class... ViewCtorArgs, class T, class... P>
+// create a mirror view and deep copy it
+// public interface that accepts arbitrary view constructor args passed by a
+// view_alloc
+template <class... ViewCtorArgs, class T, class... P,
+          class Enable = std::enable_if_t<
+              std::is_void<typename ViewTraits<T, P...>::specialize>::value>>
 auto create_mirror_view_and_copy(
-    const Impl::ViewCtorProp<ViewCtorArgs...>&,
+    [[maybe_unused]] const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    const Kokkos::Experimental::DynamicView<T, P...>& src,
+    const Kokkos::Experimental::DynamicView<T, P...>& src) {
-    std::enable_if_t<
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
-        Impl::MirrorDynamicViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace>* = nullptr) {
   using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-  static_assert(
-      alloc_prop_input::has_memory_space,
-      "The view constructor arguments passed to "
-      "Kokkos::create_mirror_view_and_copy must include a memory space!");
-  static_assert(!alloc_prop_input::has_pointer,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not include a pointer!");
-  static_assert(!alloc_prop_input::allow_padding,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not explicitly allow padding!");
 
+  Impl::check_view_ctor_args_create_mirror_view_and_copy<ViewCtorArgs...>();
+
+  if constexpr (Impl::MirrorDynamicViewType<
+                    typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space,
+                    T, P...>::is_same_memspace) {
     // same behavior as deep_copy(src, src)
-  if (!alloc_prop_input::has_execution_space)
+    if constexpr (!alloc_prop_input::has_execution_space)
       fence(
-        "Kokkos::create_mirror_view_and_copy: fence before returning src view");
+          "Kokkos::create_mirror_view_and_copy: fence before returning src "
+          "view");
     return src;
-}
+  } else {
-
-template <class... ViewCtorArgs, class T, class... P>
-auto create_mirror_view_and_copy(
-    const Impl::ViewCtorProp<ViewCtorArgs...>& arg_prop,
-    const Kokkos::Experimental::DynamicView<T, P...>& src,
-    std::enable_if_t<
-        std::is_void<typename ViewTraits<T, P...>::specialize>::value &&
-        !Impl::MirrorDynamicViewType<
-            typename Impl::ViewCtorProp<ViewCtorArgs...>::memory_space, T,
-            P...>::is_same_memspace>* = nullptr) {
-  using alloc_prop_input = Impl::ViewCtorProp<ViewCtorArgs...>;
-  static_assert(
-      alloc_prop_input::has_memory_space,
-      "The view constructor arguments passed to "
-      "Kokkos::create_mirror_view_and_copy must include a memory space!");
-  static_assert(!alloc_prop_input::has_pointer,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not include a pointer!");
-  static_assert(!alloc_prop_input::allow_padding,
-                "The view constructor arguments passed to "
-                "Kokkos::create_mirror_view_and_copy must "
-                "not explicitly allow padding!");
     using Space = typename alloc_prop_input::memory_space;
     using Mirror =
         typename Impl::MirrorDynamicViewType<Space, T, P...>::view_type;
@@ -1057,8 +1002,14 @@ auto create_mirror_view_and_copy(
       deep_copy(mirror, src);
     return mirror;
   }
+#if defined(KOKKOS_COMPILER_NVCC) && KOKKOS_COMPILER_NVCC >= 1130 && \
+    !defined(KOKKOS_COMPILER_MSVC)
+  __builtin_unreachable();
+#endif
+}
 
-template <class Space, class T, class... P>
+template <class Space, class T, class... P,
+          typename Enable = std::enable_if_t<Kokkos::is_space<Space>::value>>
 auto create_mirror_view_and_copy(
     const Space&, const Kokkos::Experimental::DynamicView<T, P...>& src,
     std::string const& name = "") {