Merge branch 'develop' into deprecate_pylammps

cmake/CMakeLists.txt

@@ -1078,12 +1078,15 @@ if(BUILD_TOOLS)
   message(STATUS "<<< Building Tools >>>")
 endif()
 if(BUILD_LAMMPS_GUI)
-  message(STATUS "<<< Building LAMMPS GUI >>>")
+  message(STATUS "<<< Building LAMMPS-GUI >>>")
   if(LAMMPS_GUI_USE_PLUGIN)
     message(STATUS "Loading LAMMPS library as plugin at run time")
   else()
     message(STATUS "Linking LAMMPS library at compile time")
   endif()
+  if(BUILD_WHAM)
+    message(STATUS "<<< Building WHAM >>>")
+  endif()
 endif()
 if(ENABLE_TESTING)
   message(STATUS "<<< Building Unit Tests >>>")

cmake/Modules/Packages/KOKKOS.cmake

@@ -7,26 +7,13 @@ endif()
 
 ########################################################################
 # consistency checks and Kokkos options/settings required by LAMMPS
-if(Kokkos_ENABLE_CUDA)
-  option(Kokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC "CUDA asynchronous malloc support" OFF)
-  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.
 if(Kokkos_ENABLE_OPENMP)
@@ -70,8 +57,8 @@ 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/4.4.01.tar.gz" CACHE STRING "URL for KOKKOS tarball")
+  set(KOKKOS_URL "https://github.com/kokkos/kokkos/archive/4.5.01.tar.gz" CACHE STRING "URL for KOKKOS tarball")
-  set(KOKKOS_MD5 "de6ee80d00b6212b02bfb7f1e71a8392" CACHE STRING "MD5 checksum of KOKKOS tarball")
+  set(KOKKOS_MD5 "4d832aa0284169d9e3fbae3165286bc6" CACHE STRING "MD5 checksum of KOKKOS tarball")
   mark_as_advanced(KOKKOS_URL)
   mark_as_advanced(KOKKOS_MD5)
   GetFallbackURL(KOKKOS_URL KOKKOS_FALLBACK)
@@ -96,7 +83,7 @@ if(DOWNLOAD_KOKKOS)
   add_dependencies(LAMMPS::KOKKOSCORE kokkos_build)
   add_dependencies(LAMMPS::KOKKOSCONTAINERS kokkos_build)
 elseif(EXTERNAL_KOKKOS)
-  find_package(Kokkos 4.4.01 REQUIRED CONFIG)
+  find_package(Kokkos 4.5.01 REQUIRED CONFIG)
   target_link_libraries(lammps PRIVATE Kokkos::kokkos)
 else()
   set(LAMMPS_LIB_KOKKOS_SRC_DIR ${LAMMPS_LIB_SOURCE_DIR}/kokkos)

cmake/Modules/Packages/ML-PACE.cmake

@@ -1,50 +1,62 @@
 # PACE library support for ML-PACE package
+find_package(pace QUIET)
 
-# set policy to silence warnings about timestamps of downloaded files. review occasionally if it may be set to NEW
+if(pace_FOUND)
-if(POLICY CMP0135)
+    find_package(pace)
-  cmake_policy(SET CMP0135 OLD)
+    target_link_libraries(lammps PRIVATE pace::pace)
-endif()
-
-set(PACELIB_URL "https://github.com/ICAMS/lammps-user-pace/archive/refs/tags/v.2023.11.25.fix.tar.gz" CACHE STRING "URL for PACE evaluator library sources")
-set(PACELIB_MD5 "b45de9a633f42ed65422567e3ce56f9f" CACHE STRING "MD5 checksum of PACE evaluator library tarball")
-mark_as_advanced(PACELIB_URL)
-mark_as_advanced(PACELIB_MD5)
-GetFallbackURL(PACELIB_URL PACELIB_FALLBACK)
-
-# LOCAL_ML-PACE points to top-level dir with local lammps-user-pace repo,
-# to make it easier to check local build without going through the public github releases
-if(LOCAL_ML-PACE)
- set(lib-pace "${LOCAL_ML-PACE}")
 else()
-  # download library sources to build folder
+    # set policy to silence warnings about timestamps of downloaded files. review occasionally if it may be set to NEW
-  if(EXISTS ${CMAKE_BINARY_DIR}/libpace.tar.gz)
+    if(POLICY CMP0135)
-    file(MD5 ${CMAKE_BINARY_DIR}/libpace.tar.gz DL_MD5)
+      cmake_policy(SET CMP0135 OLD)
-  endif()
-  if(NOT "${DL_MD5}" STREQUAL "${PACELIB_MD5}")
-    message(STATUS "Downloading ${PACELIB_URL}")
-    file(DOWNLOAD ${PACELIB_URL} ${CMAKE_BINARY_DIR}/libpace.tar.gz STATUS DL_STATUS SHOW_PROGRESS)
-    file(MD5 ${CMAKE_BINARY_DIR}/libpace.tar.gz DL_MD5)
-    if((NOT DL_STATUS EQUAL 0) OR (NOT "${DL_MD5}" STREQUAL "${PACELIB_MD5}"))
-      message(WARNING "Download from primary URL ${PACELIB_URL} failed\nTrying fallback URL ${PACELIB_FALLBACK}")
-      file(DOWNLOAD ${PACELIB_FALLBACK} ${CMAKE_BINARY_DIR}/libpace.tar.gz EXPECTED_HASH MD5=${PACELIB_MD5} SHOW_PROGRESS)
     endif()
-  else()
+
-    message(STATUS "Using already downloaded archive ${CMAKE_BINARY_DIR}/libpace.tar.gz")
+    set(PACELIB_URL "https://github.com/ICAMS/lammps-user-pace/archive/refs/tags/v.2023.11.25.fix2.tar.gz" CACHE STRING "URL for PACE evaluator library sources")
-  endif()
+    set(PACELIB_MD5 "a53bd87cfee8b07d9f44bc17aad69c3f" CACHE STRING "MD5 checksum of PACE evaluator library tarball")
+    mark_as_advanced(PACELIB_URL)
+    mark_as_advanced(PACELIB_MD5)
+    GetFallbackURL(PACELIB_URL PACELIB_FALLBACK)
+
+    # LOCAL_ML-PACE points to top-level dir with local lammps-user-pace repo,
+    # to make it easier to check local build without going through the public github releases
+    if(LOCAL_ML-PACE)
+     set(lib-pace "${LOCAL_ML-PACE}")
+    else()
+      # download library sources to build folder
+      if(EXISTS ${CMAKE_BINARY_DIR}/libpace.tar.gz)
+        file(MD5 ${CMAKE_BINARY_DIR}/libpace.tar.gz DL_MD5)
+      endif()
+      if(NOT "${DL_MD5}" STREQUAL "${PACELIB_MD5}")
+        message(STATUS "Downloading ${PACELIB_URL}")
+        file(DOWNLOAD ${PACELIB_URL} ${CMAKE_BINARY_DIR}/libpace.tar.gz STATUS DL_STATUS SHOW_PROGRESS)
+        file(MD5 ${CMAKE_BINARY_DIR}/libpace.tar.gz DL_MD5)
+        if((NOT DL_STATUS EQUAL 0) OR (NOT "${DL_MD5}" STREQUAL "${PACELIB_MD5}"))
+          message(WARNING "Download from primary URL ${PACELIB_URL} failed\nTrying fallback URL ${PACELIB_FALLBACK}")
+          file(DOWNLOAD ${PACELIB_FALLBACK} ${CMAKE_BINARY_DIR}/libpace.tar.gz EXPECTED_HASH MD5=${PACELIB_MD5} SHOW_PROGRESS)
+        endif()
+      else()
+        message(STATUS "Using already downloaded archive ${CMAKE_BINARY_DIR}/libpace.tar.gz")
+      endif()
 
 
-  # uncompress downloaded sources
+      # uncompress downloaded sources
-  execute_process(
+      execute_process(
-    COMMAND ${CMAKE_COMMAND} -E remove_directory lammps-user-pace*
+        COMMAND ${CMAKE_COMMAND} -E remove_directory lammps-user-pace*
-    COMMAND ${CMAKE_COMMAND} -E tar xzf libpace.tar.gz
+        COMMAND ${CMAKE_COMMAND} -E tar xzf libpace.tar.gz
-    WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
-  )
+      )
-  get_newest_file(${CMAKE_BINARY_DIR}/lammps-user-pace-* lib-pace)
+      get_newest_file(${CMAKE_BINARY_DIR}/lammps-user-pace-* lib-pace)
-endif()
+    endif()
-
+
-add_subdirectory(${lib-pace} build-pace)
+    # some preinstalled yaml-cpp versions don't provide a namespaced target
-set_target_properties(pace PROPERTIES CXX_EXTENSIONS ON OUTPUT_NAME lammps_pace${LAMMPS_MACHINE})
+    find_package(yaml-cpp QUIET)
-
+    if(TARGET yaml-cpp AND NOT TARGET yaml-cpp::yaml-cpp)
-if(CMAKE_PROJECT_NAME STREQUAL "lammps")
+      add_library(yaml-cpp::yaml-cpp ALIAS yaml-cpp)
-  target_link_libraries(lammps PRIVATE pace)
+    endif()
+
+    add_subdirectory(${lib-pace} build-pace)
+    set_target_properties(pace PROPERTIES CXX_EXTENSIONS ON OUTPUT_NAME lammps_pace${LAMMPS_MACHINE})
+
+    if(CMAKE_PROJECT_NAME STREQUAL "lammps")
+      target_link_libraries(lammps PRIVATE pace)
+    endif()
 endif()

cmake/Modules/Packages/VTK.cmake

@@ -1,3 +1,5 @@
+# FindVTK requires that C support is enabled when looking for MPI support
+enable_language(C)
 find_package(VTK REQUIRED NO_MODULE)
 target_compile_definitions(lammps PRIVATE -DLAMMPS_VTK)
 if (VTK_MAJOR_VERSION VERSION_LESS 9.0)

doc/doxygen/Doxyfile.in

@@ -2,7 +2,7 @@
 
 DOXYFILE_ENCODING      = UTF-8
 PROJECT_NAME           = "LAMMPS Programmer's Guide"
-PROJECT_NUMBER         = "4 May 2022"
+PROJECT_NUMBER         = "19 November 2024"
 PROJECT_BRIEF          = "Documentation of the LAMMPS library interface and Python wrapper"
 PROJECT_LOGO           = lammps-logo.png
 CREATE_SUBDIRS         = NO

doc/src/Build_basics.rst

@@ -160,7 +160,7 @@ with the OpenMP 3.1 semantics used in LAMMPS for maximal compatibility
 with compiler versions in use.  If compilation with OpenMP enabled fails
 because of your compiler requiring strict OpenMP 4.0 semantics, you can
 change the behavior by adding ``-D LAMMPS_OMP_COMPAT=4`` to the
-``LMP_INC`` variable in your makefile, or add it to the command line
+``LMP_INC`` variable in your makefile, or add it to the command-line flags
 while configuring with CMake.  LAMMPS will auto-detect a suitable setting
 for most GNU, Clang, and Intel compilers.
 
@@ -502,6 +502,8 @@ using CMake or Make.
                                       # chain.x, micelle2d.x, msi2lmp, phana,
                                       # stl_bin2txt
          -D BUILD_LAMMPS_GUI=value    # yes or no (default). Build LAMMPS-GUI
+         -D BUILD_WHAM=value          # yes (default). Download and build WHAM;
+                                      # only available for BUILD_LAMMPS_GUI=yes
 
       The generated binaries will also become part of the LAMMPS installation
       (see below).

doc/src/Build_cmake.rst

@@ -8,7 +8,7 @@ packages.  Links to those pages on the :doc:`Build overview <Build>`
 page.
 
 The following text assumes some familiarity with CMake and focuses on
-using the command line tool ``cmake`` and what settings are supported
+using the command-line tool ``cmake`` and what settings are supported
 for building LAMMPS.  A more detailed tutorial on how to use CMake
 itself, the text mode or graphical user interface, to change the
 generated output files for different build tools and development
@@ -42,9 +42,9 @@ that want to modify or extend LAMMPS.
   and adapt the LAMMPS default build configuration accordingly.
 - CMake can generate files for different build tools and integrated
   development environments (IDE).
-- CMake supports customization of settings with a command line, text
+- CMake supports customization of settings with a command-line, text
   mode, or graphical user interface.  No manual editing of files,
-  knowledge of file formats or complex command line syntax is required.
+  knowledge of file formats or complex command-line syntax is required.
 - All enabled components are compiled in a single build operation.
 - Automated dependency tracking for all files and configuration options.
 - Support for true out-of-source compilation. Multiple configurations
@@ -68,7 +68,7 @@ that purpose you can use either the command-line utility ``cmake`` (or
 graphical utility ``cmake-gui``, or use them interchangeably.  The
 second step is then the compilation and linking of all objects,
 libraries, and executables using the selected build tool.  Here is a
-minimal example using the command line version of CMake to build LAMMPS
+minimal example using the command-line version of CMake to build LAMMPS
 with no add-on packages enabled and no customization:
 
 .. code-block:: bash
@@ -131,7 +131,7 @@ file called ``CMakeLists.txt`` (for LAMMPS it is located in the
 configuration step.  The cache file contains all current CMake settings.
 
 To modify settings, enable or disable features, you need to set
-*variables* with either the ``-D`` command line flag (``-D
+*variables* with either the ``-D`` command-line flag (``-D
 VARIABLE1_NAME=value``) or change them in the text mode of the graphical
 user interface.  The ``-D`` flag can be used several times in one command.
 
@@ -141,11 +141,11 @@ a different compiler tool chain.  Those are loaded with the ``-C`` flag
 (``-C ../cmake/presets/basic.cmake``).  This step would only be needed
 once, as the settings from the preset files are stored in the
 ``CMakeCache.txt`` file. It is also possible to customize the build
-by adding one or more ``-D`` flags to the CMake command line.
+by adding one or more ``-D`` flags to the CMake command.
 
 Generating files for alternate build tools (e.g. Ninja) and project files
 for IDEs like Eclipse, CodeBlocks, or Kate can be selected using the ``-G``
-command line flag.  A list of available generator settings for your
+command-line flag.  A list of available generator settings for your
 specific CMake version is given when running ``cmake --help``.
 
 .. _cmake_multiconfig:

doc/src/Build_development.rst

@@ -263,9 +263,9 @@ will be skipped if prerequisite features are not available in LAMMPS.
    time.  Preference is given to parts of the code base that are easy to
    test or commonly used.
 
-Tests as shown by the ``ctest`` program are command lines defined in the
+Tests as shown by the ``ctest`` program are commands defined in the
 ``CMakeLists.txt`` files in the ``unittest`` directory tree.  A few
-tests simply execute LAMMPS with specific command line flags and check
+tests simply execute LAMMPS with specific command-line flags and check
 the output to the screen for expected content.  A large number of unit
 tests are special tests programs using the `GoogleTest framework
 <https://github.com/google/googletest/>`_ and linked to the LAMMPS
@@ -420,7 +420,7 @@ during MD timestepping and manipulate per-atom properties like
 positions, velocities, and forces.  For those fix styles, testing can be
 done in a very similar fashion as for force fields and thus there is a
 test program `test_fix_timestep` that shares a lot of code, properties,
-and command line flags with the force field style testers described in
+and command-line flags with the force field style testers described in
 the previous section.
 
 This tester will set up a small molecular system run with verlet run
@@ -642,10 +642,10 @@ The following target are available for both, GNU make and CMake:
 
 .. _gh-cli:
 
-GitHub command line interface
+GitHub command-line interface
 -----------------------------
 
-GitHub has developed a `command line tool <https://cli.github.com>`_
+GitHub has developed a `command-line tool <https://cli.github.com>`_
 to interact with the GitHub website via a command called ``gh``.
 This is extremely convenient when working with a Git repository hosted
 on GitHub (like LAMMPS).  It is thus highly recommended to install it

doc/src/Build_extras.rst

@@ -209,7 +209,7 @@ necessary for ``hipcc`` and the linker to work correctly.
 Using the CHIP-SPV implementation of HIP is supported. It allows one to
 run HIP code on Intel GPUs via the OpenCL or Level Zero back ends. To use
 CHIP-SPV, you must set ``-DHIP_USE_DEVICE_SORT=OFF`` in your CMake
-command line as CHIP-SPV does not yet support hipCUB. As of Summer 2022,
+command-line as CHIP-SPV does not yet support hipCUB. As of Summer 2022,
 the use of HIP for Intel GPUs is experimental. You should only use this
 option in preparations to run on Aurora system at Argonne.
 
@@ -232,7 +232,7 @@ option in preparations to run on Aurora system at Argonne.
 
 .. code:: bash
 
-   # CUDA target (not recommended, use GPU_ARCH=cuda)
+   # CUDA target (not recommended, use GPU_API=cuda)
    # !!! DO NOT set CMAKE_CXX_COMPILER !!!
    export HIP_PLATFORM=nvcc
    export HIP_PATH=/path/to/HIP/install
@@ -421,9 +421,10 @@ minutes to hours) to build.  Of course you only need to do that once.)
       cmake build system.  The ``lib/kim/Install.py`` script supports a
       ``CMAKE`` environment variable if the cmake executable is named other
       than ``cmake`` on your system.  Additional environment variables may be
-      provided on the command line for use by cmake.  For example, to use the
+      set with the ``make`` command for use by cmake.  For example, to use the
-      ``cmake3`` executable and tell it to use the gnu version 11 compilers
+      ``cmake3`` executable and tell it to use the GNU version 11 compilers
-      to build KIM, one could use the following command line.
+      called ``g++-11``, ``gcc-11`` and ``gfortran-11`` to build KIM, one
+      could use the following command.
 
       .. code-block:: bash
 
@@ -546,16 +547,7 @@ They must be specified in uppercase.
       - Local machine
    *  - AMDAVX
       - HOST
-      - AMD 64-bit x86 CPU (AVX 1)
+      - AMD chip
-   *  - ZEN
-      - HOST
-      - AMD Zen class CPU (AVX 2)
-   *  - ZEN2
-      - HOST
-      - AMD Zen2 class CPU (AVX 2)
-   *  - ZEN3
-      - HOST
-      - AMD Zen3 class CPU (AVX 2)
    *  - ARMV80
       - HOST
       - ARMv8.0 Compatible CPU
@@ -571,105 +563,126 @@ They must be specified in uppercase.
    *  - A64FX
       - HOST
       - ARMv8.2 with SVE Support
+   *  - ARMV9_GRACE
+      - HOST
+      - ARMv9 NVIDIA Grace CPU
    *  - SNB
       - HOST
-      - Intel Sandy/Ivy Bridge CPU (AVX 1)
+      - Intel Sandy/Ivy Bridge CPUs
    *  - HSW
       - HOST
-      - Intel Haswell CPU (AVX 2)
+      - Intel Haswell CPUs
    *  - BDW
       - HOST
-      - Intel Broadwell Xeon E-class CPU (AVX 2 + transactional mem)
+      - Intel Broadwell Xeon E-class CPUs
-   *  - SKL
-      - HOST
-      - Intel Skylake Client CPU
-   *  - SKX
-      - HOST
-      - Intel Skylake Xeon Server CPU (AVX512)
    *  - ICL
       - HOST
-      - Intel Ice Lake Client CPU (AVX512)
+      - Intel Ice Lake Client CPUs (AVX512)
    *  - ICX
       - HOST
-      - Intel Ice Lake Xeon Server CPU (AVX512)
+      - Intel Ice Lake Xeon Server CPUs (AVX512)
-   *  - SPR
+   *  - SKL
       - HOST
-      - Intel Sapphire Rapids Xeon Server CPU (AVX512)
+      - Intel Skylake Client CPUs
+   *  - SKX
+      - HOST
+      - Intel Skylake Xeon Server CPUs (AVX512)
    *  - KNC
       - HOST
       - Intel Knights Corner Xeon Phi
    *  - KNL
       - HOST
       - Intel Knights Landing Xeon Phi
+   *  - SPR
+      - HOST
+      - Intel Sapphire Rapids Xeon Server CPUs (AVX512)
    *  - POWER8
       - HOST
-      - IBM POWER8 CPU
+      - IBM POWER8 CPUs
    *  - POWER9
       - HOST
-      - IBM POWER9 CPU
+      - IBM POWER9 CPUs
+   *  - ZEN
+      - HOST
+      - AMD Zen architecture
+   *  - ZEN2
+      - HOST
+      - AMD Zen2 architecture
+   *  - ZEN3
+      - HOST
+      - AMD Zen3 architecture
    *  - RISCV_SG2042
       - HOST
-      - SG2042 (RISC-V) CPU
+      - SG2042 (RISC-V) CPUs
+   *  - RISCV_RVA22V
+      - HOST
+      - RVA22V (RISC-V) CPUs
    *  - KEPLER30
       - GPU
-      - NVIDIA Kepler generation CC 3.0 GPU
+      - NVIDIA Kepler generation CC 3.0
    *  - KEPLER32
       - GPU
-      - NVIDIA Kepler generation CC 3.2 GPU
+      - NVIDIA Kepler generation CC 3.2
    *  - KEPLER35
       - GPU
-      - NVIDIA Kepler generation CC 3.5 GPU
+      - NVIDIA Kepler generation CC 3.5
    *  - KEPLER37
       - GPU
-      - NVIDIA Kepler generation CC 3.7 GPU
+      - NVIDIA Kepler generation CC 3.7
    *  - MAXWELL50
       - GPU
-      - NVIDIA Maxwell generation CC 5.0 GPU
+      - NVIDIA Maxwell generation CC 5.0
    *  - MAXWELL52
       - GPU
-      - NVIDIA Maxwell generation CC 5.2 GPU
+      - NVIDIA Maxwell generation CC 5.2
    *  - MAXWELL53
       - GPU
-      - NVIDIA Maxwell generation CC 5.3 GPU
+      - NVIDIA Maxwell generation CC 5.3
    *  - PASCAL60
       - GPU
-      - NVIDIA Pascal generation CC 6.0 GPU
+      - NVIDIA Pascal generation CC 6.0
    *  - PASCAL61
       - GPU
-      - NVIDIA Pascal generation CC 6.1 GPU
+      - NVIDIA Pascal generation CC 6.1
    *  - VOLTA70
       - GPU
-      - NVIDIA Volta generation CC 7.0 GPU
+      - NVIDIA Volta generation CC 7.0
    *  - VOLTA72
       - GPU
-      - NVIDIA Volta generation CC 7.2 GPU
+      - NVIDIA Volta generation CC 7.2
    *  - TURING75
       - GPU
-      - NVIDIA Turing generation CC 7.5 GPU
+      - NVIDIA Turing generation CC 7.5
    *  - AMPERE80
       - GPU
-      - NVIDIA Ampere generation CC 8.0 GPU
+      - NVIDIA Ampere generation CC 8.0
    *  - AMPERE86
       - GPU
-      - NVIDIA Ampere generation CC 8.6 GPU
+      - NVIDIA Ampere generation CC 8.6
    *  - ADA89
       - GPU
-      - NVIDIA Ada Lovelace generation CC 8.9 GPU
+      - NVIDIA Ada generation CC 8.9
    *  - HOPPER90
       - GPU
-      - NVIDIA Hopper generation CC 9.0 GPU
+      - NVIDIA Hopper generation CC 9.0
    *  - AMD_GFX906
       - GPU
-      - AMD GPU MI50/MI60
+      - AMD GPU MI50/60
    *  - AMD_GFX908
       - GPU
       - AMD GPU MI100
    *  - AMD_GFX90A
       - GPU
       - AMD GPU MI200
+   *  - AMD_GFX940
+      - GPU
+      - AMD GPU MI300
    *  - AMD_GFX942
       - GPU
       - AMD GPU MI300
+   *  - AMD_GFX942_APU
+      - GPU
+      - AMD APU MI300A
    *  - AMD_GFX1030
       - GPU
       - AMD GPU V620/W6800
@@ -678,7 +691,7 @@ They must be specified in uppercase.
       - AMD GPU RX7900XTX
    *  - AMD_GFX1103
       - GPU
-      - AMD Phoenix APU with Radeon 740M/760M/780M/880M/890M
+      - AMD APU Phoenix
    *  - INTEL_GEN
       - GPU
       - SPIR64-based devices, e.g. Intel GPUs, using JIT
@@ -701,7 +714,7 @@ They must be specified in uppercase.
       - GPU
       - Intel GPU Ponte Vecchio
 
-This list was last updated for version 4.3.0 of the Kokkos library.
+This list was last updated for version 4.5.1 of the Kokkos library.
 
 .. tabs::
 
@@ -2191,7 +2204,7 @@ verified to work in February 2020 with Quantum Espresso versions 6.3 to
       from the sources in the *lib* folder (including the essential
       libqmmm.a) are not included in the static LAMMPS library and
       (currently) not installed, while their code is included in the
-      shared LAMMPS library.  Thus a typical command line to configure
+      shared LAMMPS library.  Thus a typical command to configure
       building LAMMPS for QMMM would be:
 
       .. code-block:: bash

doc/src/Build_windows.rst

@@ -100,9 +100,9 @@ procedure.
 
 It is possible to use both the integrated CMake support of the Visual
 Studio IDE or use an external CMake installation (e.g. downloaded from
-cmake.org) to create build files and compile LAMMPS from the command line.
+cmake.org) to create build files and compile LAMMPS from the command-line.
 
-Compilation via command line and unit tests are checked automatically
+Compilation via command-line and unit tests are checked automatically
 for the LAMMPS development branch through
 `GitHub Actions <https://github.com/lammps/lammps/actions/workflows/compile-msvc.yml>`_.
 
@@ -115,7 +115,7 @@ for the LAMMPS development branch through
 
 Please note, that for either approach CMake will create a so-called
 :ref:`"multi-configuration" build environment <cmake_multiconfig>`, and
-the command lines for building and testing LAMMPS must be adjusted
+the commands for building and testing LAMMPS must be adjusted
 accordingly.
 
 The LAMMPS cmake folder contains a ``CMakeSettings.json`` file with

doc/src/Classes_lammps.rst

@@ -4,7 +4,7 @@ LAMMPS Class
 The LAMMPS class is encapsulating an MD simulation state and thus it is
 the class that needs to be created when starting a new simulation system
 state.  The LAMMPS executable essentially creates one instance of this
-class and passes the command line flags and tells it to process the
+class and passes the command-line flags and tells it to process the
 provided input (a file or ``stdin``).  It shuts the class down when
 control is returned to it and then exits.  When using LAMMPS as a
 library from another code it is required to create an instance of this

doc/src/Commands_input.rst

@@ -69,7 +69,7 @@ WARNING message is printed.  The :doc:`Errors <Errors>` page gives
 more information on what errors mean.  The documentation for each
 command lists restrictions on how the command can be used.
 
-You can use the :ref:`-skiprun <skiprun>` command line flag
+You can use the :ref:`-skiprun <skiprun>` command-line flag
 to have LAMMPS skip the execution of any ``run``, ``minimize``, or similar
 commands to check the entire input for correct syntax to avoid crashes
 on typos or syntax errors in long runs.

doc/src/Commands_removed.rst

@@ -18,7 +18,7 @@ LAMMPS executable directly instead of having a separate tool.  A
 combination of the commands :doc:`read_restart <read_restart>` and
 :doc:`write_data <write_data>` can be used to the same effect.  For
 added convenience this conversion can also be triggered by
-:doc:`command line flags <Run_options>`
+:doc:`command-line flags <Run_options>`
 
 Fix ave/spatial and fix ave/spatial/sphere
 ------------------------------------------

doc/src/Developer_org.rst

@@ -94,12 +94,12 @@ represents what is generally referred to as an "instance of LAMMPS".  It
 is a composite holding pointers to instances of other core classes
 providing the core functionality of the MD engine in LAMMPS and through
 them abstractions of the required operations.  The constructor of the
-LAMMPS class will instantiate those instances, process the command line
+LAMMPS class will instantiate those instances, process the command-line
 flags, initialize MPI (if not already done) and set up file pointers for
 input and output.  The destructor will shut everything down and free all
 associated memory.  Thus code for the standalone LAMMPS executable in
 ``main.cpp`` simply initializes MPI, instantiates a single instance of
-LAMMPS while passing it the command line flags and input script. It
+LAMMPS while passing it the command-line flags and input script. It
 deletes the LAMMPS instance after the method reading the input returns
 and shuts down the MPI environment before it exits the executable.
 

doc/src/Developer_unittest.rst

@@ -227,12 +227,12 @@ Tests for the C-style library interface
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Tests for validating the LAMMPS C-style library interface are in the
-``unittest/c-library`` folder.  They are implemented either to be used
+``unittest/c-library`` folder.  They text either utility functions or
-for utility functions or for LAMMPS commands, but use the functions
+LAMMPS commands, but use the functions implemented in
-implemented in the ``src/library.cpp`` file as much as possible.  There
+``src/library.cpp`` as much as possible.  There may be some overlap with
-may be some overlap with other tests, but only in as much as is required
+other tests as far as the LAMMPS functionality is concerned, but the
-to test the C-style library API.  The tests are distributed over
+focus is on testing the C-style library API.  The tests are distributed
-multiple test programs which try to match the grouping of the
+over multiple test programs which try to match the grouping of the
 functions in the source code and :ref:`in the manual <lammps_c_api>`.
 
 This group of tests also includes tests invoking LAMMPS in parallel
@@ -258,7 +258,7 @@ Tests for the Python module and package
 
 The ``unittest/python`` folder contains primarily tests for classes and
 functions in the LAMMPS python module but also for commands in the
-PYTHON package.  These tests are only enabled if the necessary
+PYTHON package.  These tests are only enabled, if the necessary
 prerequisites are detected or enabled during configuration and
 compilation of LAMMPS (shared library build enabled, Python interpreter
 found, Python development files found).
@@ -272,29 +272,30 @@ Tests for the Fortran interface
 
 Tests for using the Fortran module are in the ``unittest/fortran``
 folder.  Since they are also using the GoogleTest library, they require
-implementing test wrappers in C++ that will call fortran functions
+test wrappers written in C++ that will call fortran functions with a C
-which provide a C function interface through ISO_C_BINDINGS that will in
+function interface through ISO_C_BINDINGS which will in turn call the
-turn call the functions in the LAMMPS Fortran module.
+functions in the LAMMPS Fortran module.
 
 Tests for the C++-style library interface
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 The tests in the ``unittest/cplusplus`` folder are somewhat similar to
 the tests for the C-style library interface, but do not need to test the
-several convenience and utility functions that are only available through
+convenience and utility functions that are only available through the
-the C-style interface.  Instead it can focus on the more generic features
+C-style library interface.  Instead they focus on the more generic
-that are used internally.  This part of the unit tests is currently still
+features that are used in LAMMPS internally.  This part of the unit
-mostly in the planning stage.
+tests is currently still mostly in the planning stage.
 
 Tests for reading and writing file formats
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 The ``unittest/formats`` folder contains test programs for reading and
-writing files like data files, restart files, potential files or dump files.
+writing files like data files, restart files, potential files or dump
-This covers simple things like the file i/o convenience functions in the
+files.  This covers simple things like the file i/o convenience
-``utils::`` namespace to complex tests of atom styles where creating and
+functions in the ``utils::`` namespace to complex tests of atom styles
-deleting atoms with different properties is tested in different ways
+where creating and deleting of atoms with different properties is tested
-and through script commands or reading and writing of data or restart files.
+in different ways and through script commands or reading and writing of
+data or restart files.
 
 Tests for styles computing or modifying forces
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -443,7 +444,7 @@ file for a style that is similar to one to be tested.  The file name should
 follow the naming conventions described above and after copying the file,
 the first step is to replace the style names where needed.  The coefficient
 values do not have to be meaningful, just in a reasonable range for the
-given system.  It does not matter if some forces are large, as long as
+given system.  It does not matter if some forces are large, for as long as
 they do not diverge.
 
 The template input files define a large number of index variables at the top
@@ -476,7 +477,7 @@ the tabulated coulomb, to test both code paths.  The reference results in the YA
 files then should be compared manually, if they agree well enough within the limits
 of those two approximations.
 
-The ``test_pair_style`` and equivalent programs have special command line options
+The ``test_pair_style`` and equivalent programs have special command-line options
 to update the YAML files. Running a command like
 
 .. code-block:: bash
@@ -531,19 +532,20 @@ Python module.
 Troubleshooting failed unit tests
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-The are by default no unit tests for newly added features (e.g. pair, fix,
+There are by default no unit tests for newly added features (e.g. pair,
-or compute styles) unless your pull request also includes tests for the
+fix, or compute styles) unless your pull request also includes tests for
-added features.  If you are modifying some features, you may see failures
+these added features.  If you are modifying some existing LAMMPS
-for existing tests, if your modifications have some unexpected side effects
+features, you may see failures for existing tests, if your modifications
-or your changes render the existing test invalid.  If you are adding an
+have some unexpected side effects or your changes render the existing
-accelerated version of an existing style, then only tests for INTEL,
+test invalid.  If you are adding an accelerated version of an existing
-KOKKOS (with OpenMP only), OPENMP, and OPT will be run automatically.
+style, then only tests for INTEL, KOKKOS (with OpenMP only), OPENMP, and
-Tests for the GPU package are time consuming and thus are only run
+OPT will be run automatically.  Tests for the GPU package are time
-*after* a merge, or when a special label, ``gpu_unit_tests`` is added
+consuming and thus are only run *after* a merge, or when a special
-to the pull request.  After the test has started, it is often best to
+label, ``gpu_unit_tests`` is added to the pull request.  After the test
-remove the label since every PR activity will re-trigger the test (that
+has started, it is often best to remove the label since every PR
-is a limitation of triggering a test with a label).  Support for unit
+activity will re-trigger the test (that is a limitation of triggering a
-tests when using KOKKOS with GPU acceleration is currently not supported.
+test with a label).  Support for unit tests using KOKKOS with GPU
+acceleration is currently not supported.
 
 When you see a failed build on GitHub, click on ``Details`` to be taken
 to the corresponding LAMMPS Jenkins CI web page.  Click on the "Exit"
@@ -588,7 +590,7 @@ While the epsilon (relative precision) for a single, `IEEE 754 compliant
 <https://en.wikipedia.org/wiki/IEEE_754>`_, double precision floating
 point operation is at about 2.2e-16, the achievable precision for the
 tests is lower due to most numbers being sums over intermediate results
-and the non-associativity of floating point math leading to larger
+for which the non-associativity of floating point math leads to larger
 errors.  As a rule of thumb, the test epsilon can often be in the range
 5.0e-14 to 1.0e-13.  But for "noisy" force kernels, e.g. those a larger
 amount of arithmetic operations involving `exp()`, `log()` or `sin()`
@@ -602,14 +604,14 @@ of floating point operations or that some or most intermediate operations
 may be done using approximations or with single precision floating point
 math.
 
-To rerun the failed unit test individually, change to the ``build`` directory
+To rerun a failed unit test individually, change to the ``build`` directory
 and run the test with verbose output. For example,
 
 .. code-block:: bash
 
     env TEST_ARGS=-v ctest -R ^MolPairStyle:lj_cut_coul_long -V
 
-``ctest`` with the ``-V`` flag also shows the exact command line
+``ctest`` with the ``-V`` flag also shows the exact command
 of the test. One can then use ``gdb --args`` to further debug and
 catch exceptions with the test command, for example,
 

doc/src/Developer_write_pair.rst

@@ -310,7 +310,7 @@ the constructor and the destructor.
 
 Pair styles are different from most classes in LAMMPS that define a
 "style", as their constructor only uses the LAMMPS class instance
-pointer as an argument, but **not** the command line arguments of the
+pointer as an argument, but **not** the arguments of the
 :doc:`pair_style command <pair_style>`.  Instead, those arguments are
 processed in the ``Pair::settings()`` function (or rather the version in
 the derived class).  The constructor is the place where global defaults
@@ -891,7 +891,7 @@ originally created from mixing or not).
 These data file output functions are only useful for true pair-wise
 additive potentials, where the potential parameters can be entered
 through *multiple* :doc:`pair_coeff commands <pair_coeff>`.  Pair styles
-that require a single "pair_coeff \* \*" command line are not compatible
+that require a single "pair_coeff \* \*" command are not compatible
 with reading their parameters from data files.  For pair styles like
 *born/gauss* that do support writing to data files, the potential
 parameters will be read from the data file, if present, and
@@ -1122,7 +1122,7 @@ once.  Thus, the ``coeff()`` function has to do three tasks, each of
 which is delegated to a function in the ``PairTersoff`` class:
 
 #. map elements to atom types.  Those follow the potential file name in the
-   command line arguments and are processed by the ``map_element2type()`` function.
+   command arguments and are processed by the ``map_element2type()`` function.
 #. read and parse the potential parameter file in the ``read_file()`` function.
 #. Build data structures where the original and derived parameters are
    indexed by all possible triples of atom types and thus can be looked
@@ -1356,8 +1356,8 @@ either 0 or 1.
 
 The ``morseflag`` variable defaults to 0 and is set to 1 in the
 ``PairAIREBOMorse::settings()`` function which is called by the
-:doc:`pair_style <pair_style>` command.  This function delegates
+:doc:`pair_style <pair_style>` command.  This function delegates all
-all command line processing and setting of other parameters to the
+command argument processing and setting of other parameters to the
 ``PairAIREBO::settings()`` function of the base class.
 
 .. code-block:: c++

doc/src/Errors_debug.rst

@@ -83,7 +83,7 @@ Run LAMMPS from within the debugger
 Running LAMMPS under the control of the debugger as shown below only
 works for a single MPI rank (for debugging a program running in parallel
 you usually need a parallel debugger program).  A simple way to launch
-GDB is to prefix the LAMMPS command line with ``gdb --args`` and then
+GDB is to prefix the LAMMPS command-line with ``gdb --args`` and then
 type the command "run" at the GDB prompt.  This will launch the
 debugger, load the LAMMPS executable and its debug info, and then run
 it.  When it reaches the code causing the segmentation fault, it will
@@ -180,7 +180,7 @@ inspect the behavior of a compiled program by essentially emulating a
 CPU and instrumenting the program while running.  This slows down
 execution quite significantly, but can also report issues that are not
 resulting in a crash.  The default valgrind tool is a memory checker and
-you can use it by prefixing the normal command line with ``valgrind``.
+you can use it by prefixing the normal command-line with ``valgrind``.
 Unlike GDB, this will also work for parallel execution, but it is
 recommended to redirect the valgrind output to a file (e.g. with
 ``--log-file=crash-%p.txt``, the %p will be substituted with the
@@ -235,3 +235,53 @@ from GDB. In addition you get a more specific hint about what cause the
 segmentation fault, i.e. that it is a NULL pointer dereference.  To find
 out which pointer exactly was NULL, you need to use the debugger, though.
 
+Debugging when LAMMPS appears to be stuck
+=========================================
+
+Sometimes the LAMMPS calculation appears to be stuck, that is the LAMMPS
+process or processes are active, but there is no visible progress.  This
+can have multiple reasons:
+
+- The selected styles are slow and require a lot of CPU time and the
+  system is large. When extrapolating the expected speed from smaller
+  systems, one has to factor in that not all models scale linearly with
+  system size, e.g. :doc:`kspace styles like ewald or pppm
+  <kspace_style>`. There is very little that can be done in this case.
+- The output interval is not set or set to a large value with the
+  :doc:`thermo <thermo>` command. I the first case, there will be output
+  only at the first and last step.
+- The output is block-buffered and instead of line-buffered. The output
+  will only be written to the screen after 4096 or 8192 characters of
+  output have accumulated.  This most often happens for files but also
+  with MPI parallel executables for output to the screen, since the
+  output to the screen is handled by the MPI library so that output from
+  all processes can be shown.  This can be suppressed by using the
+  ``-nonblock`` or ``-nb`` command-line flag, which turns off buffering
+  for screen and logfile output.
+- An MPI parallel calculation has a bug where a collective MPI function
+  is called (e.g. ``MPI_Barrier()``, ``MPI_Bcast()``,
+  ``MPI_Allreduce()`` and so on) before pending point-to-point
+  communications are completed or when the collective function is only
+  called from a subset of the MPI processes.  This also applies to some
+  internal LAMMPS functions like ``Error::all()`` which uses
+  ``MPI_Barrier()`` and thus ``Error::one()`` must be called, if the
+  error condition does not happen on all MPI processes simultaneously.
+- Some function in LAMMPS has a bug where a ``for`` or ``while`` loop
+  does not trigger the exit condition and thus will loop forever.  This
+  can happen when the wrong variable is incremented or when one value in
+  a comparison becomes ``NaN`` due to an overflow.
+
+In the latter two cases, further information and stack traces (see above)
+can be obtain by attaching a debugger to a running process.  For that the
+process ID (PID) is needed; this can be found on Linux machines with the
+``top``, ``htop``, ``ps``, or ``pstree`` commands.
+
+Then running the (GNU) debugger ``gdb`` with the ``-p`` flag followed by
+the process id will attach the process to the debugger and stop
+execution of that specific process.  From there on it is possible to
+issue all debugger commands in the same way as when LAMMPS was started
+from the debugger (see above).  Most importantly it is possible to
+obtain a stack trace with the ``where`` command and thus determine where
+in the execution of a timestep this process is.  Also internal data can
+be printed and execution single stepped or continued.  When the debugger
+is exited, the calculation will resume normally.

doc/src/Errors_messages.rst

@@ -7774,7 +7774,7 @@ Doc page with :doc:`WARNING messages <Errors_warnings>`
 *Too few values in body section of molecule file*
    Self-explanatory.
 
-*Too many -pk arguments in command line*
+*Too many -pk arguments in command-line*
    The string formed by concatenating the arguments is too long.  Use a
    package command in the input script instead.
 

doc/src/Examples.rst

@@ -146,6 +146,8 @@ Lowercase directories
 +-------------+------------------------------------------------------------------+
 | streitz     | use of Streitz/Mintmire potential with charge equilibration      |
 +-------------+------------------------------------------------------------------+
+| stress_vcm  | removing binned rigid body motion from binned stress profile     |
++-------------+------------------------------------------------------------------+
 | tad         | temperature-accelerated dynamics of vacancy diffusion in bulk Si |
 +-------------+------------------------------------------------------------------+
 | threebody   | regression test input for a variety of manybody potentials       |

doc/src/Fortran.rst

@@ -16,7 +16,7 @@ compiled alongside the code using it from the source code in
 ``fortran/lammps.f90`` *and* with the same compiler used to build the
 rest of the Fortran code that interfaces to LAMMPS.  When linking, you
 also need to :doc:`link to the LAMMPS library <Build_link>`.  A typical
-command line for a simple program using the Fortran interface would be:
+command for a simple program using the Fortran interface would be:
 
 .. code-block:: bash
 
@@ -91,12 +91,12 @@ function and triggered with the optional logical argument set to
      CALL lmp%close(.TRUE.)
    END PROGRAM testlib
 
-It is also possible to pass command line flags from Fortran to C/C++ and
+It is also possible to pass command-line flags from Fortran to C/C++ and
 thus make the resulting executable behave similarly to the standalone
 executable (it will ignore the `-in/-i` flag, though).  This allows
-using the command line to configure accelerator and suffix settings,
+using the command-line to configure accelerator and suffix settings,
 configure screen and logfile output, or to set index style variables
-from the command line and more.  Here is a correspondingly adapted
+from the command-line and more.  Here is a correspondingly adapted
 version of the previous example:
 
 .. code-block:: fortran
@@ -108,7 +108,7 @@ version of the previous example:
      CHARACTER(LEN=128), ALLOCATABLE :: command_args(:)
      INTEGER :: i, argc
 
-     ! copy command line flags to `command_args()`
+     ! copy command-line flags to `command_args()`
      argc = COMMAND_ARGUMENT_COUNT()
      ALLOCATE(command_args(0:argc))
      DO i=0, argc
@@ -448,7 +448,7 @@ of the contents of the :f:mod:`LIBLAMMPS` Fortran interface to LAMMPS.
    compiled with MPI support, it will also initialize MPI, if it has
    not already been initialized before.
 
-   The *args* argument with the list of command line parameters is
+   The *args* argument with the list of command-line parameters is
    optional and so it the *comm* argument with the MPI communicator.
    If *comm* is not provided, ``MPI_COMM_WORLD`` is assumed. For
    more details please see the documentation of :cpp:func:`lammps_open`.

doc/src/Howto_cmake.rst

@@ -56,7 +56,7 @@ using a shell like Bash or Zsh.
    Visual Studio IDE with the bundled CMake or from the Windows command prompt using
    a separately installed CMake package, both using the native Microsoft Visual C++
    compilers and (optionally) the Microsoft MPI SDK.  This tutorial, however, only
-   covers unix-like command line interfaces.
+   covers unix-like command-line interfaces.
 
 We also assume that you have downloaded and unpacked a recent LAMMPS source code package
 or used Git to create a clone of the LAMMPS sources on your compilation machine.
@@ -277,7 +277,7 @@ Setting options
 ---------------
 
 Options that enable, disable or modify settings are modified by setting
-the value of CMake variables. This is done on the command line with the
+the value of CMake variables. This is done on the command-line with the
 *-D* flag in the format ``-D VARIABLE=value``, e.g. ``-D
 CMAKE_BUILD_TYPE=Release`` or ``-D BUILD_MPI=on``.  There is one quirk:
 when used before the CMake directory, there may be a space between the
@@ -376,7 +376,7 @@ Using presets
 -------------
 
 Since LAMMPS has a lot of optional features and packages, specifying
-them all on the command line can be tedious. Or when selecting a
+them all on the command-line can be tedious. Or when selecting a
 different compiler toolchain, multiple options have to be changed
 consistently and that is rather error prone. Or when enabling certain
 packages, they require consistent settings to be operated in a
@@ -384,7 +384,7 @@ particular mode.  For this purpose, we are providing a selection of
 "preset files" for CMake in the folder ``cmake/presets``.  They
 represent a way to pre-load or override the CMake configuration cache by
 setting or changing CMake variables.  Preset files are loaded using the
-*-C* command line flag. You can combine loading multiple preset files or
+*-C* command-line flag. You can combine loading multiple preset files or
 change some variables later with additional *-D* flags.  A few examples:
 
 .. code-block:: bash

doc/src/Howto_github.rst

@@ -163,7 +163,7 @@ After everything is done, add the files to the branch and commit them:
    *git rm*, *git mv* for adding, removing, renaming individual files,
    respectively, and then *git commit* to finalize the commit.
    Carefully check all pending changes with *git status* before
-   committing them.  If you find doing this on the command line too
+   committing them.  If you find doing this on the command-line too
    tedious, consider using a GUI, for example the one included in git
    distributions written in Tk, i.e. use *git gui* (on some Linux
    distributions it may be required to install an additional package to

doc/src/Howto_lammps_gui.rst

@@ -20,8 +20,11 @@ to the online LAMMPS documentation for known LAMMPS commands and styles.
    (Ubuntu 20.04LTS or later and compatible), macOS (version 11 aka Big
    Sur or later), and Windows (version 10 or later) :ref:`are available
    <lammps_gui_install>` for download.  Non-MPI LAMMPS executables (as
-   ``lmp``) for running LAMMPS from the command line and :doc:`some
+   ``lmp``) for running LAMMPS from the command-line and :doc:`some
    LAMMPS tools <Tools>` compiled executables are also included.
+   Also, the pre-compiled LAMMPS-GUI packages include the WHAM executables
+   from http://membrane.urmc.rochester.edu/content/wham/ for use with
+   LAMMPS tutorials.
 
    The source code for LAMMPS-GUI is included in the LAMMPS source code
    distribution and can be found in the ``tools/lammps-gui`` folder.  It
@@ -29,16 +32,16 @@ to the online LAMMPS documentation for known LAMMPS commands and styles.
    <Build_cmake>`.
 
 LAMMPS-GUI tries to provide an experience similar to what people
-traditionally would have running LAMMPS using a command line window and
+traditionally would have running LAMMPS using a command-line window and
 the console LAMMPS executable but just rolled into a single executable:
 
 - writing & editing LAMMPS input files with a text editor
-- run LAMMPS on those input file with selected command line flags
+- run LAMMPS on those input file with selected command-line flags
 - extract data from the created files and visualize it with and
   external software
 
 That procedure is quite effective for people proficient in using the
-command line, as that allows them to use tools for the individual steps
+command-line, as that allows them to use tools for the individual steps
 that they are most comfortable with.  In fact, it is often *required* to
 adopt this workflow when running LAMMPS simulations on high-performance
 computing facilities.
@@ -100,10 +103,11 @@ MacOS 11 and later
 ^^^^^^^^^^^^^^^^^^
 
 After downloading the ``LAMMPS-macOS-multiarch-GUI-<version>.dmg``
-installer package, you need to double-click it and then, in the window
+application bundle disk image, you need to double-click it and then, in
-that opens, drag the app bundle as indicated into the "Applications"
+the window that opens, drag the app bundle as indicated into the
-folder.  The follow the instructions in the "README.txt" file to
+"Applications" folder.  Afterwards, the disk image can be unmounted.
-get access to the other included executables.
+Then follow the instructions in the "README.txt" file to get access to
+the other included command-line executables.
 
 Linux on x86\_64
 ^^^^^^^^^^^^^^^^
@@ -117,15 +121,25 @@ into the "LAMMPS_GUI" folder and execute "./lammps-gui" directly.
 
 The second variant uses `flatpak <https://www.flatpak.org>`_ and
 requires the flatpak management and runtime software to be installed.
-After downloading the ``LAMMPS-GUI-Linux-x86_64-GUI-<version>.tar.gz``
+After downloading the ``LAMMPS-GUI-Linux-x86_64-GUI-<version>.flatpak``
 flatpak bundle, you can install it with ``flatpak install --user
-LAMMPS-GUI-Linux-x86_64-GUI-<version>.tar.gz``.  After installation,
+LAMMPS-GUI-Linux-x86_64-GUI-<version>.flatpak``.  After installation,
 LAMMPS-GUI should be integrated into your desktop environment under
 "Applications > Science" but also can be launched from the console with
 ``flatpak run org.lammps.lammps-gui``.  The flatpak bundle also includes
 the console LAMMPS executable ``lmp`` which can be launched to run
-simulations with, for example: ``flatpak run --command=lmp
+simulations with, for example with:
-org.lammps.lammps-gui -in in.melt``.
+
+.. code-block:: sh
+
+   flatpak run --command=lmp org.lammps.lammps-gui -in in.melt
+
+Other bundled command-line executables are run the same way and can be
+listed with:
+
+.. code-block:: sh
+
+   ls $(flatpak info --show-location org.lammps.lammps-gui )/files/bin
 
 
 Compiling from Source
@@ -165,9 +179,9 @@ window is stored when exiting and restored when starting again.
 Opening Files
 ^^^^^^^^^^^^^
 
-The LAMMPS-GUI application can be launched without command line arguments
+The LAMMPS-GUI application can be launched without command-line arguments
 and then starts with an empty buffer in the *Editor* window.  If arguments
-are given LAMMPS will use first command line argument as the file name for
+are given LAMMPS will use first command-line argument as the file name for
 the *Editor* buffer and reads its contents into the buffer, if the file
 exists.  All further arguments are ignored.  Files can also be opened via
 the *File* menu, the `Ctrl-O` (`Command-O` on macOS) keyboard shortcut
@@ -261,14 +275,21 @@ Output Window
 
 By default, when starting a run, an *Output* window opens that displays
 the screen output of the running LAMMPS calculation, as shown below.
-This text would normally be seen in the command line window.
+This text would normally be seen in the command-line window.
 
 .. image:: JPG/lammps-gui-log.png
    :align: center
    :scale: 50%
 
 LAMMPS-GUI captures the screen output from LAMMPS as it is generated and
-updates the *Output* window regularly during a run.
+updates the *Output* window regularly during a run.  If there are any
+warnings or errors in the LAMMPS output, they are highlighted by using
+bold text colored in red.  There is a small panel at the bottom center
+of the *Output* window showing how many warnings and errors were
+detected and how many lines the entire output has.  By clicking on the
+button on the right with the warning symbol or by using the keyboard
+shortcut `Ctrl-N` (`Command-N` on macOS), you can jump to the next
+line with a warning or error.
 
 By default, the *Output* window is replaced each time a run is started.
 The runs are counted and the run number for the current run is displayed
@@ -398,7 +419,7 @@ below.
 Like for the *Output* and *Charts* windows, its content is continuously
 updated during a run.  It will show "(none)" if there are no variables
 defined.  Note that it is also possible to *set* :doc:`index style
-variables <variable>`, that would normally be set via command line
+variables <variable>`, that would normally be set via command-line
 flags, via the "Set Variables..." dialog from the *Run* menu.
 LAMMPS-GUI automatically defines the variable "gui_run" to the current
 value of the run counter.  That way it is possible to automatically
@@ -775,11 +796,11 @@ General Settings:
 
 - *Echo input to log:* when checked, all input commands, including
   variable expansions, are echoed to the *Output* window. This is
-  equivalent to using `-echo screen` at the command line.  There is no
+  equivalent to using `-echo screen` at the command-line.  There is no
   log *file* produced by default, since LAMMPS-GUI uses `-log none`.
 - *Include citation details:* when checked full citation info will be
   included to the log window.  This is equivalent to using `-cite
-  screen` on the command line.
+  screen` on the command-line.
 - *Show log window by default:* when checked, the screen output of a
   LAMMPS run will be collected in a log window during the run
 - *Show chart window by default:* when checked, the thermodynamic
@@ -828,7 +849,7 @@ Accelerators:
 
 This tab enables selection of an accelerator package for LAMMPS to use
 and is equivalent to using the `-suffix` and `-package` flags on the
-command line.  Only settings supported by the LAMMPS library and local
+command-line.  Only settings supported by the LAMMPS library and local
 hardware are available.  The `Number of threads` field allows setting
 the maximum number of threads for the accelerator packages that use
 threads.

doc/src/Howto_peri.rst

@@ -738,8 +738,8 @@ command.
 
 This can be done, for example, by using the built-in visualizer of the
 :doc:`dump image or dump movie <dump_image>` command to create snapshot
-images or a movie. Below are example command lines for using dump image
+images or a movie. Below are example command for using dump image with
-with the :ref:`example listed below <periexample>` and a set of images
+the :ref:`example listed below <periexample>` and a set of images
 created for steps 300, 600, and 2000 this way.
 
 .. code-block:: LAMMPS

doc/src/Howto_wsl.rst

@@ -260,7 +260,7 @@ Switch into the :code:`examples/melt` folder:
 
    cd ../examples/melt
 
-To run this example in serial, use the following command line:
+To run this example in serial, use the following command:
 
 .. code-block::
 

doc/src/Install_git.rst

@@ -60,7 +60,7 @@ between them at any time using "git checkout <branch name>".)
    files (mostly by accident).  If you do not need access to the entire
    commit history (most people don't), you can speed up the "cloning"
    process and reduce local disk space requirements by using the
-   ``--depth`` git command line flag.  That will create a "shallow clone"
+   ``--depth`` git command-line flag.  That will create a "shallow clone"
    of the repository, which contains only a subset of the git history.
    Using a depth of 1000 is usually sufficient to include the head
    commits of the *develop*, the *release*, and the *maintenance*

doc/src/Intro_authors.rst

@@ -8,6 +8,8 @@ send an email to all of them at this address: "developers at
 lammps.org".  General questions about LAMMPS should be posted in the
 `LAMMPS forum on MatSci <https://matsci.org/lammps/>`_.
 
+.. We need to keep this file in sync with https://www.lammps.org/authors.html
+
 .. raw:: latex
 
    \small
@@ -27,7 +29,7 @@ lammps.org".  General questions about LAMMPS should be posted in the
    * - `Steve Plimpton <sjp_>`_
      - SNL (retired)
      - sjplimp at gmail.com
-     - MD kernels, parallel algorithms & scalability, code structure and design
+     - original author, MD kernels, parallel algorithms & scalability, code structure and design
    * - `Aidan Thompson <at_>`_
      - SNL
      - athomps at sandia.gov

doc/src/Modify_style.rst

@@ -46,7 +46,7 @@ Include files (varied)
   but instead should be initialized either in the initializer list of
   the constructor or explicitly assigned in the body of the constructor.
   If the member variable is relevant to the functionality of a class
-  (for example when it stores a value from a command line argument), the
+  (for example when it stores a value from a command-line argument), the
   member variable declaration is followed by a brief comment explaining
   its purpose and what its values can be.  Class members that are
   pointers should always be initialized to ``nullptr`` in the

doc/src/Packages_details.rst

@@ -2171,8 +2171,8 @@ the :doc:`Build extras <Build_extras>` page.
 * ``src/OPENMP/README``
 * :doc:`Accelerator packages <Speed_packages>`
 * :doc:`OPENMP package <Speed_omp>`
-* :doc:`Command line option -suffix/-sf omp <Run_options>`
+* :doc:`Command-line option -suffix/-sf omp <Run_options>`
-* :doc:`Command line option -package/-pk omp <Run_options>`
+* :doc:`Command-line option -package/-pk omp <Run_options>`
 * :doc:`package omp <package>`
 * Search the :doc:`commands <Commands_all>` pages (:doc:`fix <Commands_fix>`, :doc:`compute <Commands_compute>`,
   :doc:`pair <Commands_pair>`, :doc:`bond, angle, dihedral, improper <Commands_bond>`,
@@ -2789,14 +2789,15 @@ implements smoothed particle hydrodynamics (SPH) for liquids.  See the
 related :ref:`MACHDYN package <PKG-MACHDYN>` package for smooth Mach dynamics
 (SMD) for solids.
 
-This package contains ideal gas, Lennard-Jones equation of states,
+This package contains ideal gas, Lennard-Jones equation of states, Tait,
-Tait, and full support for complete (i.e. internal-energy dependent)
+and full support for complete (i.e. internal-energy dependent) equations
-equations of state.  It allows for plain or Monaghans XSPH integration
+of state.  It allows for plain or Monaghans XSPH integration of the
-of the equations of motion.  It has options for density continuity or
+equations of motion.  It has options for density continuity or density
-density summation to propagate the density field.  It has
+summation to propagate the density field.  It has :doc:`set <set>`
-:doc:`set <set>` command options to set the internal energy and density
+command options to set the internal energy and density of particles from
-of particles from the input script and allows the same quantities to
+the input script and allows the same quantities to be output with
-be output with thermodynamic output or to dump files via the :doc:`compute property/atom <compute_property_atom>` command.
+thermodynamic output or to dump files via the :doc:`compute
+property/atom <compute_property_atom>` command.
 
 **Author:** Georg Ganzenmuller (Fraunhofer-Institute for High-Speed
 Dynamics, Ernst Mach Institute, Germany).
@@ -2809,6 +2810,17 @@ Dynamics, Ernst Mach Institute, Germany).
 * ``examples/PACKAGES/sph``
 * https://www.lammps.org/movies.html#sph
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
+.. note::
+
+   Please also note, that the :ref:`RHEO package <PKG-RHEO>` offers
+   similar functionality in a more modern and flexible implementation.
+
 ----------
 
 .. _PKG-SPIN:

doc/src/Run_basics.rst

@@ -1,8 +1,8 @@
 Basics of running LAMMPS
 ========================
 
-LAMMPS is run from the command line, reading commands from a file via
+LAMMPS is run from the command-line, reading commands from a file via
-the ``-in`` command line flag, or from standard input.  Using the ``-in
+the ``-in`` command-line flag, or from standard input.  Using the ``-in
 in.file`` variant is recommended (see note below).  The name of the
 LAMMPS executable is either ``lmp`` or ``lmp_<machine>`` with
 `<machine>` being the machine string used when compiling LAMMPS.  This
@@ -25,7 +25,7 @@ build LAMMPS:
 You normally run the LAMMPS command in the directory where your input
 script is located.  That is also where output files are produced by
 default, unless you provide specific other paths in your input script or
-on the command line.  As in some of the examples above, the LAMMPS
+on the command-line.  As in some of the examples above, the LAMMPS
 executable itself can be placed elsewhere.
 
 .. note::

doc/src/Run_options.rst

@@ -632,7 +632,7 @@ the ``-package omp`` command-line switch or the :doc:`package omp <package>` com
 
 The :doc:`suffix <suffix>` command can also be used within an input
 script to set a suffix, or to turn off or back on any suffix setting
-made via the command line.
+made via the command-line.
 
 ----------
 

doc/src/Run_windows.rst

@@ -20,7 +20,7 @@ To run with 4 threads, you can type this:
    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
+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`.
 

doc/src/Speed_gpu.rst

@@ -31,7 +31,8 @@ Coulombics.  It has the following general features:
   (for Nvidia GPUs, AMD GPUs, Intel GPUs, and multicore CPUs).
   so that the same functionality is supported on a variety of hardware.
 
-**Required hardware/software:**
+Required hardware/software
+""""""""""""""""""""""""""
 
 To compile and use this package in CUDA mode, you currently need
 to have an NVIDIA GPU and install the corresponding NVIDIA CUDA
@@ -69,12 +70,14 @@ To compile and use this package in HIP mode, you have to have the AMD ROCm
 software installed. Versions of ROCm older than 3.5 are currently deprecated
 by AMD.
 
-**Building LAMMPS with the GPU package:**
+Building LAMMPS with the GPU package
+""""""""""""""""""""""""""""""""""""
 
 See the :ref:`Build extras <gpu>` page for
 instructions.
 
-**Run with the GPU package from the command line:**
+Run with the GPU package from the command-line
+""""""""""""""""""""""""""""""""""""""""""""""
 
 The ``mpirun`` or ``mpiexec`` command sets the total number of MPI tasks
 used by LAMMPS (one or multiple per compute node) and the number of MPI
@@ -133,7 +136,8 @@ affect the setting for bonded interactions (LAMMPS default is "on").
 The "off" setting for pairwise interaction is currently required for
 GPU package pair styles.
 
-**Or run with the GPU package by editing an input script:**
+Run with the GPU package by editing an input script
+"""""""""""""""""""""""""""""""""""""""""""""""""""
 
 The discussion above for the ``mpirun`` or ``mpiexec`` command, MPI
 tasks/node, and use of multiple MPI tasks/GPU is the same.
@@ -149,7 +153,8 @@ You must also use the :doc:`package gpu <package>` command to enable the
 GPU package, unless the ``-sf gpu`` or ``-pk gpu`` :doc:`command-line switches <Run_options>` were used.  It specifies the number of
 GPUs/node to use, as well as other options.
 
-**Speed-ups to expect:**
+Speed-up to expect
+""""""""""""""""""
 
 The performance of a GPU versus a multicore CPU is a function of your
 hardware, which pair style is used, the number of atoms/GPU, and the
@@ -176,10 +181,13 @@ better with multiple OMP threads because the inter-process communication
 is higher for these styles with the GPU package in order to allow
 deterministic results.
 
-**Guidelines for best performance:**
+Guidelines for best performance
+"""""""""""""""""""""""""""""""
 
-* Using multiple MPI tasks per GPU will often give the best performance,
+* Using multiple MPI tasks (2-10) per GPU will often give the best
-  as allowed my most multicore CPU/GPU configurations.
+  performance, as allowed my most multicore CPU/GPU configurations.
+  Using too many MPI tasks will result in worse performance due to
+  growing overhead with the growing number of MPI tasks.
 * If the number of particles per MPI task is small (e.g. 100s of
   particles), it can be more efficient to run with fewer MPI tasks per
   GPU, even if you do not use all the cores on the compute node.
@@ -199,12 +207,13 @@ deterministic results.
   :doc:`angle <angle_style>`, :doc:`dihedral <dihedral_style>`,
   :doc:`improper <improper_style>`, and :doc:`long-range <kspace_style>`
   calculations will not be included in the "Pair" time.
-* Since only part of the pppm kspace style is GPU accelerated, it
+* Since only part of the pppm kspace style is GPU accelerated, it may be
-  may be faster to only use GPU acceleration for Pair styles with
+  faster to only use GPU acceleration for Pair styles with long-range
-  long-range electrostatics.  See the "pair/only" keyword of the
+  electrostatics.  See the "pair/only" keyword of the :doc:`package
-  package command for a shortcut to do that.  The work between kspace
+  command <package>` for a shortcut to do that.  The distribution of
-  on the CPU and non-bonded interactions on the GPU can be balanced
+  work between kspace on the CPU and non-bonded interactions on the GPU
-  through adjusting the coulomb cutoff without loss of accuracy.
+  can be balanced through adjusting the coulomb cutoff without loss of
+  accuracy.
 * When the *mode* setting for the package gpu command is force/neigh,
   the time for neighbor list calculations on the GPU will be added into
   the "Pair" time, not the "Neigh" time.  An additional breakdown of the
@@ -220,4 +229,6 @@ deterministic results.
 Restrictions
 """"""""""""
 
-None.
+When using :doc:`hybrid pair styles <pair_hybrid>`, the neighbor list
+must be generated on the host instead of the GPU and thus the potential
+GPU acceleration is reduced.

doc/src/Speed_intel.rst

@@ -1,5 +1,5 @@
 INTEL package
-==================
+=============
 
 The INTEL package is maintained by Mike Brown at Intel
 Corporation.  It provides two methods for accelerating simulations,
@@ -13,18 +13,18 @@ twice, once on the CPU and once with an offload flag. This allows
 LAMMPS to run on the CPU cores and co-processor cores simultaneously.
 
 Currently Available INTEL Styles
-"""""""""""""""""""""""""""""""""""""
+""""""""""""""""""""""""""""""""
 
 * Angle Styles: charmm, harmonic
-* Bond Styles: fene, fourier, harmonic
+* Bond Styles: fene, harmonic
 * Dihedral Styles: charmm, fourier, harmonic, opls
-* Fixes: nve, npt, nvt, nvt/sllod, nve/asphere
+* Fixes: nve, npt, nvt, nvt/sllod, nve/asphere, electrode/conp, electrode/conq, electrode/thermo
 * Improper Styles: cvff, harmonic
 * Pair Styles: airebo, airebo/morse, buck/coul/cut, buck/coul/long,
   buck, dpd, eam, eam/alloy, eam/fs, gayberne, lj/charmm/coul/charmm,
   lj/charmm/coul/long, lj/cut, lj/cut/coul/long, lj/long/coul/long,
-  rebo, sw, tersoff
+  rebo, snap, sw, tersoff
-* K-Space Styles: pppm, pppm/disp
+* K-Space Styles: pppm, pppm/disp, pppm/electrode
 
 .. warning::
 
@@ -33,7 +33,7 @@ Currently Available INTEL Styles
    input requires it, LAMMPS will abort with an error message.
 
 Speed-up to expect
-"""""""""""""""""""
+""""""""""""""""""
 
 The speedup will depend on your simulation, the hardware, which
 styles are used, the number of atoms, and the floating-point
@@ -312,21 +312,21 @@ almost all cases.
    recommended, especially when running on a machine with Intel
    Hyper-Threading technology disabled.
 
-Run with the INTEL package from the command line
+Run with the INTEL package from the command-line
-"""""""""""""""""""""""""""""""""""""""""""""""""""""
+""""""""""""""""""""""""""""""""""""""""""""""""
 
-To enable INTEL optimizations for all available styles used in
+To enable INTEL optimizations for all available styles used in the input
-the input script, the ``-sf intel`` :doc:`command-line switch <Run_options>` can be used without any requirement for
+script, the ``-sf intel`` :doc:`command-line switch <Run_options>` can
-editing the input script. This switch will automatically append
+be used without any requirement for editing the input script. This
-"intel" to styles that support it. It also invokes a default command:
+switch will automatically append "intel" to styles that support it. It
-:doc:`package intel 1 <package>`. This package command is used to set
+also invokes a default command: :doc:`package intel 1 <package>`. This
-options for the INTEL package.  The default package command will
+package command is used to set options for the INTEL package.  The
-specify that INTEL calculations are performed in mixed precision,
+default package command will specify that INTEL calculations are
-that the number of OpenMP threads is specified by the OMP_NUM_THREADS
+performed in mixed precision, that the number of OpenMP threads is
-environment variable, and that if co-processors are present and the
+specified by the OMP_NUM_THREADS environment variable, and that if
-binary was built with offload support, that 1 co-processor per node
+co-processors are present and the binary was built with offload support,
-will be used with automatic balancing of work between the CPU and the
+that 1 co-processor per node will be used with automatic balancing of
-co-processor.
+work between the CPU and the co-processor.
 
 You can specify different options for the INTEL package by using
 the ``-pk intel Nphi`` :doc:`command-line switch <Run_options>` with

doc/src/Speed_kokkos.rst

@@ -77,7 +77,7 @@ version 23 November 2023 and Kokkos version 4.2.
    rank.  When running with multiple MPI ranks, you may see segmentation
    faults without GPU-aware MPI support. These can be avoided by adding
    the flags :doc:`-pk kokkos gpu/aware off <Run_options>` to the
-   LAMMPS command line or by using the command :doc:`package kokkos
+   LAMMPS command-line or by using the command :doc:`package kokkos
    gpu/aware off <package>` in the input file.
 
 .. admonition:: Intel Data Center GPU support
@@ -423,7 +423,7 @@ in the ``kokkos-cuda.cmake`` CMake preset file.
    cmake -DKokkos_ENABLE_CUDA=yes -DKokkos_ENABLE_OPENMP=yes ../cmake
 
 The suffix "/kk" is equivalent to "/kk/device", and for Kokkos CUDA,
-using the ``-sf kk`` in the command line gives the default CUDA version
+using the ``-sf kk`` in the command-line gives the default CUDA version
 everywhere.  However, if the "/kk/host" suffix is added to a specific
 style in the input script, the Kokkos OpenMP (CPU) version of that
 specific style will be used instead.  Set the number of OpenMP threads
@@ -439,7 +439,7 @@ For example, the command to run with 1 GPU and 8 OpenMP threads is then:
 
    mpiexec -np 1 lmp_kokkos_cuda_openmpi -in in.lj -k on g 1 t 8 -sf kk
 
-Conversely, if the ``-sf kk/host`` is used in the command line and then
+Conversely, if the ``-sf kk/host`` is used in the command-line and then
 the "/kk" or "/kk/device" suffix is added to a specific style in your
 input script, then only that specific style will run on the GPU while
 everything else will run on the CPU in OpenMP mode. Note that the
@@ -451,7 +451,7 @@ on the host CPU can overlap with a pair style running on the
 GPU. First compile with ``--default-stream per-thread`` added to ``CCFLAGS``
 in the Kokkos CUDA Makefile.  Then explicitly use the "/kk/host"
 suffix for kspace and bonds, angles, etc.  in the input file and the
-"kk" suffix (equal to "kk/device") on the command line.  Also make
+"kk" suffix (equal to "kk/device") on the command-line.  Also make
 sure the environment variable ``CUDA_LAUNCH_BLOCKING`` is not set to "1"
 so CPU/GPU overlap can occur.
 

doc/src/Speed_omp.rst

@@ -21,7 +21,7 @@ Building LAMMPS with the OPENMP package
 See the :ref:`Build extras <openmp>` page for
 instructions.
 
-Run with the OPENMP package from the command line
+Run with the OPENMP package from the command-line
 """""""""""""""""""""""""""""""""""""""""""""""""""
 
 These examples assume one or more 16-core nodes.

doc/src/Speed_opt.rst

@@ -17,7 +17,7 @@ Building LAMMPS with the OPT package
 
 See the :ref:`Build extras <opt>` page for instructions.
 
-Run with the OPT package from the command line
+Run with the OPT package from the command-line
 """"""""""""""""""""""""""""""""""""""""""""""
 
 .. code-block:: bash

doc/src/Tools.rst

@@ -501,7 +501,7 @@ Here are a few highlights of LAMMPS-GUI
 - Indicator for line that caused an error
 - Visualization of current state in Image Viewer (via calling :doc:`write_dump image <dump_image>`)
 - Capture of images created via :doc:`dump image <dump_image>` in Slide show window
-- Dialog to set variables, similar to the LAMMPS command line flag '-v' / '-var'
+- Dialog to set variables, similar to the LAMMPS command-line flag '-v' / '-var'
 - Support for GPU, INTEL, KOKKOS/OpenMP, OPENMAP, and OPT and accelerator packages
 
 Parallelization
@@ -550,7 +550,7 @@ will be found automatically.  2) you can download the `Flatpak file
 *flatpak* command: ``flatpak install --user
 LAMMPS-Linux-x86_64-GUI-<version>.flatpak`` and run it with ``flatpak
 run org.lammps.lammps-gui``.  The flatpak bundle also includes the
-command line version of LAMMPS and some LAMMPS tools like msi2lmp.  The
+command-line version of LAMMPS and some LAMMPS tools like msi2lmp.  The
 can be launched by using the ``--command`` flag. For example to run
 LAMMPS directly on the ``in.lj`` benchmark input you would type in the
 ``bench`` folder: ``flatpak run --command=lmp -in in.lj`` The flatpak
@@ -608,10 +608,10 @@ would be the ``examples/COUPLE/plugin`` folder of the LAMMPS
 distribution.
 
 When compiling LAMMPS-GUI with plugin support, there is an additional
-command line flag (``-p <path>`` or ``--pluginpath <path>``) which
+command-line flag (``-p <path>`` or ``--pluginpath <path>``) which
 allows to override the path to LAMMPS shared library used by the GUI.
 This is usually auto-detected on the first run and can be changed in the
-LAMMPS-GUI *Preferences* dialog.  The command line flag allows to reset
+LAMMPS-GUI *Preferences* dialog.  The command-line flag allows to reset
 this path to a valid value in case the original setting has become
 invalid.  An empty path ("") as argument restores the default setting.
 
@@ -656,7 +656,7 @@ it will create a compressed ``LAMMPS-Win10-amd64.zip`` zip file with the
 executables and required dependent .dll files.  This zip file can be
 uncompressed and ``lammps-gui.exe`` run directly from there.  The
 uncompressed folder can be added to the ``PATH`` environment and LAMMPS
-and LAMMPS-GUI can be launched from anywhere from the command line.
+and LAMMPS-GUI can be launched from anywhere from the command-line.
 
 **MinGW64 Cross-compiler**
 
@@ -876,7 +876,7 @@ the same ``LAMMPS_CACHING_DIR``. This script does the following:
  #. Start a simple local HTTP server using Python to host files for CMake
 
 Afterwards, it will print out instruction on how to modify the CMake
-command line to make sure it uses the local HTTP server.
+commands to make sure it uses the local HTTP server.
 
 To undo the environment changes and shutdown the local HTTP server,
 run the ``deactivate_caches`` command.
@@ -1025,7 +1025,7 @@ with those in the provided log file with the same number of processors
 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
+the mpirun command, the LAMMPS command-line arguments, and the
 tolerances for individual thermo quantities can be specified in a
 configuration file in YAML format.
 

doc/src/compute_sph_e_atom.rst

@@ -33,6 +33,12 @@ particle.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The value of the internal energy will be 0.0 for atoms not in the
 specified compute group.
 

doc/src/compute_sph_rho_atom.rst

@@ -32,6 +32,12 @@ kernel function interpolation using "pair style sph/rhosum".
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The value of the SPH density will be 0.0 for atoms not in the
 specified compute group.
 

doc/src/compute_sph_t_atom.rst

@@ -37,6 +37,12 @@ particles, i.e. a Smooth-Particle Hydrodynamics particle.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The value of the internal energy will be 0.0 for atoms not in the
 specified compute group.
 

doc/src/compute_temp_chunk.rst

@@ -184,11 +184,24 @@ temp/chunk calculation to a file is to use the
 The keyword/value option pairs are used in the following ways.
 
 The *com* keyword can be used with a value of *yes* to subtract the
-velocity of the center-of-mass for each chunk from the velocity of the
+velocity of the center-of-mass (VCM) for each chunk from the velocity of
-atoms in that chunk, before calculating either the global or per-chunk
+the atoms in that chunk, before calculating either the global or per-chunk
-temperature.  This can be useful if the atoms are streaming or
+temperature. This can be useful if the atoms are streaming or
 otherwise moving collectively, and you wish to calculate only the
-thermal temperature.
+thermal temperature. This per-chunk VCM bias can be used in other fixes and
+computes that can incorporate a temperature bias. If this compute is used
+as a temperature bias in other commands then this bias is subtracted from
+each atom, the command runs with the remaining thermal velocities, and
+then the bias is added back in. This includes thermostatting
+fixes like :doc:`fix nvt <fix_nh>`,
+:doc:`fix temp/rescale <fix_temp_rescale>`,
+:doc:`fix temp/berendsen <fix_temp_berendsen>`, and
+:doc:`fix langevin <fix_langevin>`, and computes like
+:doc:`compute stress/atom <compute_stress_atom>` and
+:doc:`compute pressure <compute_pressure>`. See the input script in
+examples/stress_vcm for an example of how to use the *com* keyword in
+conjunction with compute stress/atom to create a stress profile of a rigid
+body while removing the overall motion of the rigid body.
 
 For the *bias* keyword, *bias-ID* refers to the ID of a temperature
 compute that removes a "bias" velocity from each atom.  This also

doc/src/dump_image.rst

@@ -681,7 +681,7 @@ MPEG or other movie file you can use:
 
 * c) Use FFmpeg
 
-  FFmpeg is a command line tool that is available on many platforms and
+  FFmpeg is a command-line tool that is available on many platforms and
   allows extremely flexible encoding and decoding of movies.
 
   .. code-block:: bash

doc/src/fix_imd.rst

@@ -97,7 +97,7 @@ adjustments.
 To connect VMD to a listening LAMMPS simulation on the same machine
 with fix imd enabled, one needs to start VMD and load a coordinate or
 topology file that matches the fix group.  When the VMD command
-prompts appears, one types the command line:
+prompts appears, one types the command:
 
 .. parsed-literal::
 

doc/src/fix_precession_spin.rst

@@ -135,7 +135,7 @@ directions for the forces. Only the direction of the vector is
 important; its length is ignored (the entered vectors are
 normalized).
 
-Those styles can be combined within one single command line.
+Those styles can be combined within one single command.
 
 .. note::
 

doc/src/fix_sph.rst

@@ -32,6 +32,12 @@ Hydrodynamics.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 Restart, fix_modify, output, run start/stop, minimize info
 """""""""""""""""""""""""""""""""""""""""""""""""""""""""""
 

doc/src/fix_sph_stationary.rst

@@ -32,6 +32,12 @@ space.  SPH stands for Smoothed Particle Hydrodynamics.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 Restart, fix_modify, output, run start/stop, minimize info
 """""""""""""""""""""""""""""""""""""""""""""""""""""""""""
 

doc/src/kim_commands.rst

@@ -1084,10 +1084,11 @@ the form of *key_name_key*-*key_name_value* pairs).  For example,
    kim property modify 1 key mass    source-value 26.98154
    kim property modify 1 key mass    source-unit  amu
 
-where the special keyword "key" is followed by a *key_name* ("species" or
+where the special keyword "key" is followed by a *key_name* ("species"
-"mass" in the above) and one or more key-value pairs.  These key-value pairs
+or "mass" in the above) and one or more key-value pairs.  These
-may continue until either another "key" keyword is given or the end of the
+key-value pairs may continue until either another "key" keyword is given
-command line is reached.  Thus, the above could equivalently be written as
+or the end of the line is reached.  Thus, the above could equivalently
+be written as
 
 .. code-block:: LAMMPS
 

doc/src/label.rst

@@ -24,12 +24,12 @@ Description
 """""""""""
 
 Label this line of the input script with the chosen ID.  Unless a jump
-command was used previously, this does nothing.  But if a
+command was used previously, this does nothing.  But if a :doc:`jump
-:doc:`jump <jump>` command was used with a label argument to begin
+<jump>` command was used with a label argument to begin invoking this
-invoking this script file, then all command lines in the script prior
+script file, then all commands in the script prior to this line will be
-to this line will be ignored.  I.e. execution of the script will begin
+ignored.  I.e. execution of the script will begin at this line.  This is
-at this line.  This is useful for looping over a section of the input
+useful for looping over a section of the input script as discussed in
-script as discussed in the :doc:`jump <jump>` command.
+the :doc:`jump <jump>` command.
 
 Restrictions
 """"""""""""

doc/src/package.rst

@@ -504,7 +504,7 @@ as it is for non-accelerated pair styles
 The *binsize* keyword sets the size of bins used to bin atoms during
 neighbor list builds. The same value can be set by the
 :doc:`neigh_modify binsize <neigh_modify>` command. Making it an option
-in the package kokkos command allows it to be set from the command line.
+in the package kokkos command allows it to be set from the command-line.
 The default value for CPUs is 0.0, which means the LAMMPS default will be
 used, which is bins = 1/2 the size of the pairwise cutoff + neighbor skin
 distance. This is fine when neighbor lists are built on the CPU. For GPU
@@ -664,7 +664,7 @@ too.
    Also note that if the :doc:`-sf hybrid intel omp command-line switch <Run_options>` is used, it invokes a "package intel" command, followed by a
    "package omp" command, both with a setting of *Nthreads* = 0. Likewise
    for a hybrid suffix for gpu and omp. Note that KOKKOS also supports
-   setting the number of OpenMP threads from the command line using the
+   setting the number of OpenMP threads from the command-line using the
    "-k on" :doc:`command-line switch <Run_options>`. The default for
    KOKKOS is 1 thread per MPI task, so any other number of threads should
    be explicitly set using the "-k on" command-line switch (and this

doc/src/pair_mgpt.rst

@@ -111,8 +111,8 @@ routines.  For x-86 machines, there is a provided Makefile.mgptfast
 which enables the fast algebra routines, i.e. build LAMMPS with "make
 mgptfast".  The user will be informed in the output files of the
 matrix kernels in use. To further improve speed, on x86 the option
-precision single can be added to the :doc:`pair_coeff <pair_coeff>`
+*precision single* can be added to the :doc:`pair_coeff <pair_coeff>`
-command line, which improves speed (up to a factor of two) at the cost
+command, which improves speed (up to a factor of two) at the cost
 of doing matrix calculations with 7 digit precision instead of the
 default 16. For consistency the default option can be specified
 explicitly by the option precision double.

doc/src/pair_sph_heatconduction.rst

@@ -30,6 +30,12 @@ The transport model is the diffusion equation for the internal energy.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_sph_idealgas.rst

@@ -36,6 +36,12 @@ particles from interpenetrating :ref:`(Monaghan) <ideal-Monoghan>`.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_sph_lj.rst

@@ -34,6 +34,12 @@ interpenetrating :ref:`(Monaghan) <Monoghan>`.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_sph_rhosum.rst

@@ -29,6 +29,12 @@ SPH particles by kernel function interpolation, every Nstep timesteps.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_sph_taitwater.rst

@@ -41,6 +41,12 @@ prevent particles from interpenetrating :ref:`(Monaghan) <Monaghan>`.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_sph_taitwater_morris.rst

@@ -37,6 +37,12 @@ This pair style also computes laminar viscosity :ref:`(Morris) <Morris>`.
 See `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in
 LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 The following coefficients must be defined for each pair of atoms
 types via the :doc:`pair_coeff <pair_coeff>` command as in the examples
 above.

doc/src/pair_thole.rst

@@ -131,7 +131,7 @@ command.
 * LJ cutoff (distance units)
 
 The last two coefficients are optional and default to the global values from
-the *pair_style* command line.
+the *pair_style* command.
 
 ----------
 

doc/src/read_restart.rst

@@ -48,9 +48,9 @@ meaning that the trajectories of a restarted run will precisely match
 those produced by the original run had it continued on.
 
 Some information about a restart file can be gathered directly from the
-command line when using LAMMPS with the :ref:`-restart2info
+command-line when using LAMMPS with the :ref:`-restart2info
-<restart2info>` command line flag.  On Unix-like operating systems (like
+<restart2info>` command-line flag.  On Unix-like operating systems (like
-Linux or macOS), one can also :ref:`configure the "file" command line
+Linux or macOS), one can also :ref:`configure the "file" command-line
 program <magic>` to display basic information about a restart file
 
 The binary restart file format was not designed with backward, forward,
@@ -60,9 +60,9 @@ Changes to the architecture, compilation settings, or LAMMPS version can
 render a restart file unreadable or it may read the data incorrectly.
 If you want a more portable format, you can use the data file format as
 created by the :doc:`write_data <write_data>` command.  Binary restart
-files can also be converted into a data file from the command line by
+files can also be converted into a data file from the command-line by
 the LAMMPS executable that wrote them using the :ref:`-restart2data
-<restart2data>` command line flag.
+<restart2data>` command-line flag.
 
 Several things can prevent exact restarts due to round-off effects, in
 which case the trajectories in the 2 runs will slowly diverge.  These

doc/src/set.rst

@@ -516,6 +516,12 @@ Keywords *sph/e*, *sph/cv*, and *sph/rho* set the energy, heat capacity,
 and density of smoothed particle hydrodynamics (SPH) particles.  See
 `this PDF guide <PDF/SPH_LAMMPS_userguide.pdf>`_ to using SPH in LAMMPS.
 
+.. note::
+
+   Please note that the SPH PDF guide file has not been updated for
+   many years and thus does not reflect the current *syntax* of the
+   SPH package commands. For that please refer to the LAMMPS manual.
+
 Keyword *smd/mass/density* sets the mass of all selected particles, but
 it is only applicable to the Smooth Mach Dynamics package MACHDYN.  It
 assumes that the particle volume has already been correctly set and

doc/src/suffix.rst

@@ -30,7 +30,7 @@ Description
 
 This command allows you to use variants of various styles if they
 exist.  In that respect it operates the same as the :doc:`-suffix command-line switch <Run_options>`.  It also has options to turn
-off or back on any suffix setting made via the command line.
+off or back on any suffix setting made via the command-line.
 
 The specified style can be *gpu*, *intel*, *kk*, *omp*, *opt* or
 *hybrid*\ . These refer to optional packages that LAMMPS can be built

doc/src/variable.rst

@@ -71,9 +71,9 @@ Syntax
          feature functions = is_available(category,feature), is_active(category,feature), is_defined(category,id)
          atom value = id[i], mass[i], type[i], mol[i], x[i], y[i], z[i], vx[i], vy[i], vz[i], fx[i], fy[i], fz[i], q[i]
          atom vector = id, mass, type, mol, radius, q, x, y, z, vx, vy, vz, fx, fy, fz
-         custom atom property = i_name, d_name, i_name[i], d_name[i], i2_name[i], d2_name[i], i2_name[i][j], d_name[i][j]
+         custom atom property = i_name, d_name, i_name[i], d_name[i], i2_name[i], d2_name[i], i2_name[i][j], d2_name[i][j]
-         compute references = c_ID, c_ID[i], c_ID[i][j], C_ID, C_ID[i]
+         compute references = c_ID, c_ID[i], c_ID[i][j], C_ID, C_ID[i], C_ID[i][j]
-         fix references = f_ID, f_ID[i], f_ID[i][j], F_ID, F_ID[i]
+         fix references = f_ID, f_ID[i], f_ID[i][j], F_ID, F_ID[i], F_ID[i][j]
          variable references = v_name, v_name[i]
          vector initialization = [1,3,7,10] (for *vector* variables only)
 

examples/README

@@ -112,6 +112,7 @@ snap:     examples for using several bundled SNAP potentials
 srd:      stochastic rotation dynamics (SRD) particles as solvent
 steinhardt: Steinhardt-Nelson Q_l and W_l parameters usng orientorder/atom
 streitz:  Streitz-Mintmire potential for Al2O3
+stress_vcm: removing binned rigid body motion from binned stress profile
 tad:      temperature-accelerated dynamics of vacancy diffusion in bulk Si
 template: examples for using atom_style template and comparing to atom style molecular
 tersoff:  regression test input for Tersoff variants

examples/comb/README

@@ -20,5 +20,3 @@ Examples:
 4. in.comb.Cu2O.elastic: Cu2O crystal, qeq on, minimizes, then calculates
    elastic constants
 5. in.comb.HfO2: HfO2 polymorphs: Monoclinic HfO2 NVT @ 300K
-6. in.comb.CuaS: Metallic Cu and amorphous silica interface, qeq on,
-	five step NVE run

examples/stress_vcm/README

@@ -0,0 +1,32 @@
+README stress_vcm 
+=================
+
+Contents:
+
+-   in.stress_vcm: Example script showing how to remove binned 
+    velocities of center of mass (VCM) from stress calculations.
+-   stress_comparison.19Nov24.png: Plot shows the stress 
+    calculated in bars on the y axis for each positional bin on 
+    the x axis. Plotted are three different time steps from 
+    stress profiles with and without the VCM removed. Plot 
+    generated using Python.
+-   stress_xx.19Nov24.out: Output file generated by fix ave/time.
+-   log.19Nov24.stress_vcm.g++.1: LAMMPS log file with 1 proc.
+-   log.19Nov24.stress_vcm.g++.4: LAMMPS log file with 4 procs.
+
+Notes:
+
+-   Running this script as-is will generate two files. A log 
+    file with thermodynamic data and a stress_xx.out file 
+    containing the binned stress profile with the VCM removed.
+-   To generate the binned stress profile without removing the 
+    VCM then the compute stress/atom command at step three 
+    needs the last keyword "ch_temp_vcm" to be replaced with 
+    "NULL". 
+-   Uncommenting the line under "Atom dump" will generate an 
+    all atom dump file every 50 time steps containing atom ID,
+    type, and xyz coordinates.
+-   Uncommenting the lines under "Image dumps" will generate 
+    .jpg image files every 250 timesteps.
+-   Uncommenting lines under "Movie dump" will generate a .avi 
+    movie file showing timesteps every 125 timesteps. 

examples/stress_vcm/in.stress_vcm

@@ -0,0 +1,113 @@
+# Removing Binned Velocities of Center of Mass (VCM) from Stress 
+
+# This example shows how to remove rigid body motion from 
+# binned stress calculations. This uses a combination of commands
+# from compute chunk/atom, compute temp/chunk, compute 
+# stress/atom and fix ave/time. We'll show how these commands 
+# work in the context of a shockwave experiment on a cube of 
+# atoms. To shock the cube, a rectangular region of atoms is 
+# frozen, moved into the cube with a constant velocity along the
+# x direction, and then unfrozen. As the shockwave begins 
+# propagating, the body of the cube also moves along the x 
+# direction. To better understand the stress dynamics of the 
+# cube we remove the velocity component belonging to the overall 
+# motion of each bin.
+
+units		    metal
+boundary        p p p
+atom_style	    atomic
+lattice		    fcc 5.3589
+processors 		1 * *
+
+# Defining regions for box and atoms.
+# In this experiment an elongated simulation cell is 
+# defined in the x direction to allow for non-periodic 
+# motion of the atoms.
+
+region		    box1 block -3 24 0 12 0 12 units lattice
+region		    box2 block 0 12 0 12 0 12 units lattice
+
+# Creating box and atoms
+
+create_box	    1 box1
+create_atoms    1 region box2
+
+mass		    1 40.00
+
+# Adding energy to the system
+
+velocity        all create 600.0 9999
+
+pair_style	    lj/cut 10 
+pair_coeff	    1 1 0.04 3.405
+
+# Begin time integration
+
+timestep        2e-3
+
+fix	            fix_nve all nve
+
+thermo		    100
+
+run             500
+
+#--------------------------------------#
+# Chunk, Stress, and VCM removal steps #
+#--------------------------------------#
+
+# 1. Create 20 equispaced bins sliced along the x direction.
+# -"units reduced" normalizes the distance from 0.0 to 1.0
+variable        nbins index 20
+variable        fraction equal 1.0/v_nbins
+variable        volfrac equal 1/(vol*${fraction}) 
+compute         ch_id all chunk/atom bin/1d x lower ${fraction} units reduced 
+
+# 2. Calculate temperature bins with VCM aka COM velocities removed.
+compute         ch_temp_vcm all temp/chunk ch_id com yes
+
+# 3. Compute per atom stress with VCM removed via temp-ID.
+# -The velocities from specified temp-ID are used to compute stress.
+# -Stress/atom units are pressure*volume! Optionally handled next step.
+compute         atom_stress_vcm all stress/atom ch_temp_vcm
+
+# 4. Divide out bin volume from xx stress component.
+variable        stress atom -(c_atom_stress_vcm[1])/(vol*${fraction})
+
+# 5. Sum the per atom stresses in each bin.
+compute         ch_stress_vcm all reduce/chunk ch_id sum v_stress
+
+# 6. Average and output to file.
+# -The average output is every 100 steps with samples collected 20 times with 5 step intervals.
+fix             ave_stress_vcm all ave/time 5 20 100 c_ch_stress_vcm mode vector file stress_xx.out
+
+#--------------------------------------#
+
+# Piston compressing along x direction
+
+region 			piston block -1 1 INF INF INF INF units lattice
+group          	piston region piston
+fix 			fix_piston piston move linear 5 0 0 units box # strain rate ~ 8e10 1/s 
+
+thermo_style    custom step temp ke pe lx ly lz pxx pyy pzz econserve 
+
+# Atom dump
+
+# dump		    atom_dump all atom 50 dump.vcm
+
+# # Image dumps
+
+# dump		    2 all image 250 image.*.jpg type type &
+# 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    2 pad 1
+
+# # Movie dump
+
+# dump		    3 all movie 125 movie.avi type type &
+# 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    3 pad 1
+
+run	            500
+
+unfix           fix_piston
+
+run             1500

examples/stress_vcm/log.19Nov24.stress_vcm.g++.1

@@ -0,0 +1,253 @@
+LAMMPS (19 Nov 2024)
+OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:99)
+  using 1 OpenMP thread(s) per MPI task
+# Removing Binned Velocities of Center of Mass (VCM) from Stress
+
+# This example shows how to remove rigid body motion from
+# binned stress calculations. This uses a combination of commands
+# from compute chunk/atom, compute temp/chunk, compute
+# stress/atom and fix ave/time. We'll show how these commands
+# work in the context of a shockwave experiment on a cube of
+# atoms. To shock the cube, a rectangular region of atoms is
+# frozen, moved into the cube with a constant velocity along the
+# x direction, and then unfrozen. As the shockwave begins
+# propagating, the body of the cube also moves along the x
+# direction. To better understand the stress dynamics of the
+# cube we remove the velocity component belonging to the overall
+# motion of each bin.
+
+units		    metal
+boundary        p p p
+atom_style	    atomic
+lattice		    fcc 5.3589
+Lattice spacing in x,y,z = 5.3589 5.3589 5.3589
+processors 		1 * *
+
+# Defining regions for box and atoms.
+# In this experiment an elongated simulation cell is
+# defined in the x direction to allow for non-periodic
+# motion of the atoms.
+
+region		    box1 block -3 24 0 12 0 12 units lattice
+region		    box2 block 0 12 0 12 0 12 units lattice
+
+# Creating box and atoms
+
+create_box	    1 box1
+Created orthogonal box = (-16.0767 0 0) to (128.6136 64.3068 64.3068)
+  1 by 1 by 1 MPI processor grid
+create_atoms    1 region box2
+Created 7200 atoms
+  using lattice units in orthogonal box = (-16.0767 0 0) to (128.6136 64.3068 64.3068)
+  create_atoms CPU = 0.002 seconds
+
+mass		    1 40.00
+
+# Adding energy to the system
+
+velocity        all create 600.0 9999
+
+pair_style	    lj/cut 10
+pair_coeff	    1 1 0.04 3.405
+
+# Begin time integration
+
+timestep        2e-3
+
+fix	            fix_nve all nve
+
+thermo		    100
+
+run             500
+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 = 12
+  ghost atom cutoff = 12
+  binsize = 6, bins = 25 11 11
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair lj/cut, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/3d
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 5.721 | 5.721 | 5.721 Mbytes
+   Step          Temp          E_pair         E_mol          TotEng         Press     
+         0   600           -2252.7567      0             -1694.4304     -974.62456    
+       100   284.72172     -1977.4291      0             -1712.483       2453.7429    
+       200   304.44519     -1994.7937      0             -1711.4941      1822.2699    
+       300   304.28012     -1993.2958      0             -1710.1498      1498.3794    
+       400   296.76492     -1985.1364      0             -1708.9836      1259.9474    
+       500   295.00895     -1982.4224      0             -1707.9036      964.9526     
+Loop time of 3.01696 on 1 procs for 500 steps with 7200 atoms
+
+Performance: 28.638 ns/day, 0.838 hours/ns, 165.730 timesteps/s, 1.193 Matom-step/s
+99.4% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 2.8439     | 2.8439     | 2.8439     |   0.0 | 94.26
+Neigh   | 0.11212    | 0.11212    | 0.11212    |   0.0 |  3.72
+Comm    | 0.015585   | 0.015585   | 0.015585   |   0.0 |  0.52
+Output  | 0.003747   | 0.003747   | 0.003747   |   0.0 |  0.12
+Modify  | 0.026097   | 0.026097   | 0.026097   |   0.0 |  0.87
+Other   |            | 0.01551    |            |       |  0.51
+
+Nlocal:           7200 ave        7200 max        7200 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:           6410 ave        6410 max        6410 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:         615095 ave      615095 max      615095 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 615095
+Ave neighs/atom = 85.429861
+Neighbor list builds = 9
+Dangerous builds = 0
+
+#------------------------------------#
+# Chunk, Stress, and VCM removal steps
+#------------------------------------#
+
+# 1. Create 20 equispaced bins sliced along the x direction.
+# "units reduced" normalizes the distance from 0 to 1
+variable        nbins index 20
+variable        fraction equal 1.0/v_nbins
+variable        volfrac equal 1/(vol*${fraction})
+variable        volfrac equal 1/(vol*0.05)
+compute         ch_id all chunk/atom bin/1d x lower ${fraction} units reduced
+compute         ch_id all chunk/atom bin/1d x lower 0.05 units reduced
+
+# 2. Calculate temperature bins with VCM aka COM velocities removed.
+compute         ch_temp_vcm all temp/chunk ch_id com yes
+
+# 3. Compute per atom stress with VCM removed via temp-ID.
+# The velocities from specified temp-ID are used to compute stress
+# Stress/atom units are pressure*volume! Optionally handled next step.
+compute         atom_stress_vcm all stress/atom ch_temp_vcm
+
+# 4. Divide out bin volume from xx stress component.
+variable        stress atom -(c_atom_stress_vcm[1])/(vol*${fraction})
+variable        stress atom -(c_atom_stress_vcm[1])/(vol*0.05)
+
+# 5. Sum the per atom stresses in each bin.
+compute         ch_stress_vcm all reduce/chunk ch_id sum v_stress
+
+# 6. Average and output to file.
+# The average output is every 100 steps with samples collected 20 times with 5 step intervals
+# fix             ave_stress_vcm all ave/time 5 20 100 c_ch_stress_vcm mode vector file stress_xx.out
+
+#------------------------------------#
+
+# Piston compressing along x direction
+
+region 			piston block -1 1 INF INF INF INF units lattice
+group          	piston region piston
+863 atoms in group piston
+fix 			fix_piston piston move linear 5 0 0 units box # strain rate ~ 8e10 1/s
+
+thermo_style    custom step temp ke pe lx ly lz pxx pyy pzz econserve
+
+# Atom dump
+
+# dump		    atom_dump all atom 50 dump.vcm
+
+# # Image dumps
+
+# dump		    2 all image 250 image.*.jpg type type # 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    2 pad 1
+
+# # Movie dump
+
+# dump		    3 all movie 125 movie.avi type type # 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    3 pad 1
+
+run	            500
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+WARNING: One or more atoms are time integrated more than once (src/modify.cpp:296)
+Per MPI rank memory allocation (min/avg/max) = 6.975 | 6.975 | 6.975 Mbytes
+   Step          Temp          KinEng         PotEng           Lx             Ly             Lz            Pxx            Pyy            Pzz         Econserve   
+       500   295.00895      274.51875     -1982.4224      144.6903       64.3068        64.3068        631.89976      1127.2965      1135.6616     -1707.9036    
+       600   357.38902      332.56613     -1951.3422      144.6903       64.3068        64.3068        2236.6706      2003.2726      1943.6815     -1618.7761    
+       700   420.30268      391.11005     -1911.8178      144.6903       64.3068        64.3068        3761.5011      3065.4699      3140.3169     -1520.7077    
+       800   484.96279      451.27911     -1875.379       144.6903       64.3068        64.3068        5362.254       4174.4201      4166.0818     -1424.0999    
+       900   587.78954      546.96391     -1871.217       144.6903       64.3068        64.3068        6481.4714      4875.705       4676.6083     -1324.2531    
+      1000   684.07997      636.56636     -1868.1639      144.6903       64.3068        64.3068        7734.6158      5271.3524      5272.1276     -1231.5975    
+Loop time of 3.09383 on 1 procs for 500 steps with 7200 atoms
+
+Performance: 27.927 ns/day, 0.859 hours/ns, 161.612 timesteps/s, 1.164 Matom-step/s
+100.0% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 2.8485     | 2.8485     | 2.8485     |   0.0 | 92.07
+Neigh   | 0.18767    | 0.18767    | 0.18767    |   0.0 |  6.07
+Comm    | 0.011533   | 0.011533   | 0.011533   |   0.0 |  0.37
+Output  | 0.003323   | 0.003323   | 0.003323   |   0.0 |  0.11
+Modify  | 0.031777   | 0.031777   | 0.031777   |   0.0 |  1.03
+Other   |            | 0.01107    |            |       |  0.36
+
+Nlocal:           7200 ave        7200 max        7200 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:           6409 ave        6409 max        6409 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:         646408 ave      646408 max      646408 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 646408
+Ave neighs/atom = 89.778889
+Neighbor list builds = 15
+Dangerous builds = 0
+
+unfix           fix_piston
+
+run             1500
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Per MPI rank memory allocation (min/avg/max) = 6.6 | 6.6 | 6.6 Mbytes
+   Step          Temp          KinEng         PotEng           Lx             Ly             Lz            Pxx            Pyy            Pzz         Econserve   
+      1000   684.07997      636.56636     -1868.1639      144.6903       64.3068        64.3068        7734.6158      5271.3524      5272.1276     -1231.5975    
+      1100   710.19886      660.87113     -1894.0485      144.6903       64.3068        64.3068        8048.3485      5396.6668      5376.5956     -1233.1774    
+      1200   717.16487      667.35331     -1901.3849      144.6903       64.3068        64.3068        8009.7984      5634.5121      5349.4113     -1234.0316    
+      1300   710.26037      660.92837     -1894.9802      144.6903       64.3068        64.3068        8063.4125      5572.1245      5530.174      -1234.0519    
+      1400   715.93921      666.21278     -1898.8885      144.6903       64.3068        64.3068        7752.0927      5293.5463      5322.2312     -1232.6757    
+      1500   748.85411      696.84154     -1926.4891      144.6903       64.3068        64.3068        6030.5428      4076.8886      4012.7653     -1229.6475    
+      1600   767.98982      714.64815     -1939.8556      144.6903       64.3068        64.3068        4200.3475      2532.5711      2530.5518     -1225.2075    
+      1700   757.22042      704.62675     -1925.553       144.6903       64.3068        64.3068        2686.7843      1482.2796      1505.8073     -1220.9262    
+      1800   727.30327      676.78754     -1894.6635      144.6903       64.3068        64.3068        1764.2793      781.37451      801.18668     -1217.8759    
+      1900   688.82146      640.97853     -1856.5007      144.6903       64.3068        64.3068        1022.805       417.32394      359.74951     -1215.5221    
+      2000   655.91228      610.35509     -1823.954       144.6903       64.3068        64.3068        551.98825     -20.148643     -56.976652     -1213.5989    
+      2100   620.22468      577.14622     -1789.1761      144.6903       64.3068        64.3068        264.05975     -266.8323      -314.45533     -1212.0299    
+      2200   589.13325      548.21428     -1758.9252      144.6903       64.3068        64.3068        41.369707     -533.503       -525.69401     -1210.7109    
+      2300   563.20394      524.08593     -1733.6036      144.6903       64.3068        64.3068       -220.99189     -810.90513     -774.65084     -1209.5176    
+      2400   540.44236      502.90528     -1711.3384      144.6903       64.3068        64.3068       -358.01508     -962.31635     -977.3253      -1208.4332    
+      2500   523.5718       487.20648     -1694.7088      144.6903       64.3068        64.3068       -521.87444     -1152.8386     -1231.7615     -1207.5023    
+Loop time of 9.34327 on 1 procs for 1500 steps with 7200 atoms
+
+Performance: 27.742 ns/day, 0.865 hours/ns, 160.543 timesteps/s, 1.156 Matom-step/s
+98.5% CPU use with 1 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 8.4692     | 8.4692     | 8.4692     |   0.0 | 90.65
+Neigh   | 0.7512     | 0.7512     | 0.7512     |   0.0 |  8.04
+Comm    | 0.031189   | 0.031189   | 0.031189   |   0.0 |  0.33
+Output  | 0.010584   | 0.010584   | 0.010584   |   0.0 |  0.11
+Modify  | 0.053052   | 0.053052   | 0.053052   |   0.0 |  0.57
+Other   |            | 0.02803    |            |       |  0.30
+
+Nlocal:           7200 ave        7200 max        7200 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Nghost:           6380 ave        6380 max        6380 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+Neighs:         515773 ave      515773 max      515773 min
+Histogram: 1 0 0 0 0 0 0 0 0 0
+
+Total # of neighbors = 515773
+Ave neighs/atom = 71.635139
+Neighbor list builds = 57
+Dangerous builds = 0
+Total wall time: 0:00:15

examples/stress_vcm/log.19Nov24.stress_vcm.g++.4

@@ -0,0 +1,253 @@
+LAMMPS (19 Nov 2024)
+OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:99)
+  using 1 OpenMP thread(s) per MPI task
+# Removing Binned Velocities of Center of Mass (VCM) from Stress
+
+# This example shows how to remove rigid body motion from
+# binned stress calculations. This uses a combination of commands
+# from compute chunk/atom, compute temp/chunk, compute
+# stress/atom and fix ave/time. We'll show how these commands
+# work in the context of a shockwave experiment on a cube of
+# atoms. To shock the cube, a rectangular region of atoms is
+# frozen, moved into the cube with a constant velocity along the
+# x direction, and then unfrozen. As the shockwave begins
+# propagating, the body of the cube also moves along the x
+# direction. To better understand the stress dynamics of the
+# cube we remove the velocity component belonging to the overall
+# motion of each bin.
+
+units		    metal
+boundary        p p p
+atom_style	    atomic
+lattice		    fcc 5.3589
+Lattice spacing in x,y,z = 5.3589 5.3589 5.3589
+processors 		1 * *
+
+# Defining regions for box and atoms.
+# In this experiment an elongated simulation cell is
+# defined in the x direction to allow for non-periodic
+# motion of the atoms.
+
+region		    box1 block -3 24 0 12 0 12 units lattice
+region		    box2 block 0 12 0 12 0 12 units lattice
+
+# Creating box and atoms
+
+create_box	    1 box1
+Created orthogonal box = (-16.0767 0 0) to (128.6136 64.3068 64.3068)
+  1 by 2 by 2 MPI processor grid
+create_atoms    1 region box2
+Created 7200 atoms
+  using lattice units in orthogonal box = (-16.0767 0 0) to (128.6136 64.3068 64.3068)
+  create_atoms CPU = 0.001 seconds
+
+mass		    1 40.00
+
+# Adding energy to the system
+
+velocity        all create 600.0 9999
+
+pair_style	    lj/cut 10
+pair_coeff	    1 1 0.04 3.405
+
+# Begin time integration
+
+timestep        2e-3
+
+fix	            fix_nve all nve
+
+thermo		    100
+
+run             500
+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 = 12
+  ghost atom cutoff = 12
+  binsize = 6, bins = 25 11 11
+  1 neighbor lists, perpetual/occasional/extra = 1 0 0
+  (1) pair lj/cut/opt, perpetual
+      attributes: half, newton on
+      pair build: half/bin/atomonly/newton
+      stencil: half/bin/3d
+      bin: standard
+Per MPI rank memory allocation (min/avg/max) = 3.662 | 3.662 | 3.662 Mbytes
+   Step          Temp          E_pair         E_mol          TotEng         Press     
+         0   600           -2252.7567      0             -1694.4304     -974.62456    
+       100   284.1896      -1976.961       0             -1712.5101      2462.6396    
+       200   308.58965     -1998.6349      0             -1711.4787      1789.0033    
+       300   300.55093     -1989.9838      0             -1710.308       1545.8576    
+       400   297.91491     -1986.2519      0             -1709.029       1247.7121    
+       500   294.66041     -1982.1097      0             -1707.9153      961.03073    
+Loop time of 0.942408 on 4 procs for 500 steps with 7200 atoms
+
+Performance: 91.680 ns/day, 0.262 hours/ns, 530.556 timesteps/s, 3.820 Matom-step/s
+82.1% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.61287    | 0.63781    | 0.65858    |   2.1 | 67.68
+Neigh   | 0.030246   | 0.031529   | 0.034546   |   1.0 |  3.35
+Comm    | 0.23074    | 0.25145    | 0.27819    |   3.7 | 26.68
+Output  | 0.000282   | 0.0003735  | 0.000463   |   0.0 |  0.04
+Modify  | 0.005566   | 0.0057635  | 0.005989   |   0.2 |  0.61
+Other   |            | 0.01548    |            |       |  1.64
+
+Nlocal:           1800 ave        1814 max        1787 min
+Histogram: 1 0 1 0 0 0 0 1 0 1
+Nghost:         3713.5 ave        3727 max        3699 min
+Histogram: 1 0 1 0 0 0 0 1 0 1
+Neighs:         153532 ave      154995 max      152312 min
+Histogram: 1 0 1 0 0 1 0 0 0 1
+
+Total # of neighbors = 614128
+Ave neighs/atom = 85.295556
+Neighbor list builds = 9
+Dangerous builds = 0
+
+#------------------------------------#
+# Chunk, Stress, and VCM removal steps
+#------------------------------------#
+
+# 1. Create 20 equispaced bins sliced along the x direction.
+# "units reduced" normalizes the distance from 0 to 1
+variable        nbins index 20
+variable        fraction equal 1.0/v_nbins
+variable        volfrac equal 1/(vol*${fraction})
+variable        volfrac equal 1/(vol*0.05)
+compute         ch_id all chunk/atom bin/1d x lower ${fraction} units reduced
+compute         ch_id all chunk/atom bin/1d x lower 0.05 units reduced
+
+# 2. Calculate temperature bins with VCM aka COM velocities removed.
+compute         ch_temp_vcm all temp/chunk ch_id com yes
+
+# 3. Compute per atom stress with VCM removed via temp-ID.
+# The velocities from specified temp-ID are used to compute stress
+# Stress/atom units are pressure*volume! Optionally handled next step.
+compute         atom_stress_vcm all stress/atom ch_temp_vcm
+
+# 4. Divide out bin volume from xx stress component.
+variable        stress atom -(c_atom_stress_vcm[1])/(vol*${fraction})
+variable        stress atom -(c_atom_stress_vcm[1])/(vol*0.05)
+
+# 5. Sum the per atom stresses in each bin.
+compute         ch_stress_vcm all reduce/chunk ch_id sum v_stress
+
+# 6. Average and output to file.
+# The average output is every 100 steps with samples collected 20 times with 5 step intervals
+# fix             ave_stress_vcm all ave/time 5 20 100 c_ch_stress_vcm mode vector file stress_xx.out
+
+#------------------------------------#
+
+# Piston compressing along x direction
+
+region 			piston block -1 1 INF INF INF INF units lattice
+group          	piston region piston
+864 atoms in group piston
+fix 			fix_piston piston move linear 5 0 0 units box # strain rate ~ 8e10 1/s
+
+thermo_style    custom step temp ke pe lx ly lz pxx pyy pzz econserve
+
+# Atom dump
+
+# dump		    atom_dump all atom 50 dump.vcm
+
+# # Image dumps
+
+# dump		    2 all image 250 image.*.jpg type type # 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    2 pad 1
+
+# # Movie dump
+
+# dump		    3 all movie 125 movie.avi type type # 		        axes yes 0.8 0.02 view 60 -30
+# dump_modify	    3 pad 1
+
+run	            500
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+WARNING: One or more atoms are time integrated more than once (src/modify.cpp:296)
+Per MPI rank memory allocation (min/avg/max) = 4.916 | 4.916 | 4.916 Mbytes
+   Step          Temp          KinEng         PotEng           Lx             Ly             Lz            Pxx            Pyy            Pzz         Econserve   
+       500   294.66041      274.19441     -1982.1097      144.6903       64.3068        64.3068        645.25795      1119.5337      1118.3006     -1707.9153    
+       600   357.88641      333.02897     -1951.8158      144.6903       64.3068        64.3068        2176.0343      1929.2787      1981.8479     -1618.7869    
+       700   418.41159      389.3503      -1912.8337      144.6903       64.3068        64.3068        3702.2875      3043.7607      3081.1607     -1523.4834    
+       800   483.71102      450.11428     -1875.7955      144.6903       64.3068        64.3068        5254.3875      4190.9789      4158.3561     -1425.6813    
+       900   586.0893       545.38176     -1870.9313      144.6903       64.3068        64.3068        6509.1439      4756.2216      4724.7086     -1325.5495    
+      1000   686.32946      638.65962     -1874.811       144.6903       64.3068        64.3068        7515.1606      5193.049       5261.8688     -1236.1514    
+Loop time of 0.656417 on 4 procs for 500 steps with 7200 atoms
+
+Performance: 131.624 ns/day, 0.182 hours/ns, 761.711 timesteps/s, 5.484 Matom-step/s
+92.8% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 0.51672    | 0.52334    | 0.53259    |   0.8 | 79.73
+Neigh   | 0.045091   | 0.045915   | 0.047402   |   0.4 |  6.99
+Comm    | 0.060735   | 0.071794   | 0.079302   |   2.6 | 10.94
+Output  | 0.000208   | 0.000389   | 0.000926   |   0.0 |  0.06
+Modify  | 0.006007   | 0.0061595  | 0.00626    |   0.1 |  0.94
+Other   |            | 0.008815   |            |       |  1.34
+
+Nlocal:           1800 ave        1811 max        1785 min
+Histogram: 1 0 0 1 0 0 0 0 0 2
+Nghost:        3713.25 ave        3727 max        3702 min
+Histogram: 2 0 0 0 0 0 0 1 0 1
+Neighs:         161477 ave      162958 max      159732 min
+Histogram: 1 0 0 0 1 0 0 1 0 1
+
+Total # of neighbors = 645909
+Ave neighs/atom = 89.709583
+Neighbor list builds = 15
+Dangerous builds = 0
+
+unfix           fix_piston
+
+run             1500
+Generated 0 of 0 mixed pair_coeff terms from geometric mixing rule
+Per MPI rank memory allocation (min/avg/max) = 4.541 | 4.541 | 4.541 Mbytes
+   Step          Temp          KinEng         PotEng           Lx             Ly             Lz            Pxx            Pyy            Pzz         Econserve   
+      1000   686.32946      638.65962     -1874.811       144.6903       64.3068        64.3068        7515.1606      5193.049       5261.8688     -1236.1514    
+      1100   709.7333       660.43791     -1898.2844      144.6903       64.3068        64.3068        7932.8638      5334.6171      5364.5335     -1237.8465    
+      1200   713.27253      663.73132     -1902.4588      144.6903       64.3068        64.3068        7957.2574      5500.6231      5538.0516     -1238.7275    
+      1300   705.44796      656.45022     -1895.1575      144.6903       64.3068        64.3068        7996.7265      5584.6233      5538.2494     -1238.7072    
+      1400   711.86463      662.42121     -1899.8416      144.6903       64.3068        64.3068        7674.2462      5292.4915      5294.5366     -1237.4204    
+      1500   742.18946      690.63979     -1924.9562      144.6903       64.3068        64.3068        6047.915       4056.6156      4014.4446     -1234.3164    
+      1600   762.81764      709.83522     -1939.8563      144.6903       64.3068        64.3068        4185.5873      2530.0572      2576.1943     -1230.0211    
+      1700   754.40428      702.00621     -1927.7337      144.6903       64.3068        64.3068        2662.7604      1509.1985      1484.7252     -1225.7275    
+      1800   721.03504      670.95468     -1893.5556      144.6903       64.3068        64.3068        1765.8783      835.89765      861.9432      -1222.6009    
+      1900   689.64162      641.74172     -1861.8886      144.6903       64.3068        64.3068        941.58148      312.93205      409.79901     -1220.1469    
+      2000   650.79664      605.59477     -1823.9889      144.6903       64.3068        64.3068        543.39234      28.48735       80.396505     -1218.3941    
+      2100   616.04072      573.25286     -1790.1764      144.6903       64.3068        64.3068        308.16444     -235.20997     -248.22531     -1216.9235    
+      2200   587.18712      546.40333     -1761.8878      144.6903       64.3068        64.3068        37.044801     -476.50396     -470.83059     -1215.4845    
+      2300   562.84178      523.74892     -1738.2239      144.6903       64.3068        64.3068       -139.28348     -711.17273     -730.80877     -1214.475     
+      2400   540.48362      502.94367     -1716.3529      144.6903       64.3068        64.3068       -320.98222     -951.2066      -943.93966     -1213.4093    
+      2500   519.80431      483.70067     -1696.1896      144.6903       64.3068        64.3068       -471.61317     -1088.8457     -1131.5396     -1212.4889    
+Loop time of 1.97213 on 4 procs for 1500 steps with 7200 atoms
+
+Performance: 131.431 ns/day, 0.183 hours/ns, 760.598 timesteps/s, 5.476 Matom-step/s
+95.3% CPU use with 4 MPI tasks x 1 OpenMP threads
+
+MPI task timing breakdown:
+Section |  min time  |  avg time  |  max time  |%varavg| %total
+---------------------------------------------------------------
+Pair    | 1.5455     | 1.5599     | 1.5723     |   0.8 | 79.10
+Neigh   | 0.16844    | 0.1704     | 0.17237    |   0.4 |  8.64
+Comm    | 0.19002    | 0.2047     | 0.22068    |   2.4 | 10.38
+Output  | 0.000525   | 0.0006785  | 0.001077   |   0.0 |  0.03
+Modify  | 0.012434   | 0.012601   | 0.012777   |   0.1 |  0.64
+Other   |            | 0.02388    |            |       |  1.21
+
+Nlocal:           1800 ave        1833 max        1776 min
+Histogram: 1 0 1 0 1 0 0 0 0 1
+Nghost:           3702 ave        3732 max        3674 min
+Histogram: 1 0 0 1 0 0 1 0 0 1
+Neighs:         129380 ave      132578 max      127003 min
+Histogram: 1 0 0 2 0 0 0 0 0 1
+
+Total # of neighbors = 517520
+Ave neighs/atom = 71.877778
+Neighbor list builds = 54
+Dangerous builds = 0
+Total wall time: 0:00:03

examples/stress_vcm/stress_xx.19Nov24.out

@@ -0,0 +1,423 @@
+# Time-averaged data for fix ave_stress_vcm
+# TimeStep Number-of-rows
+# Row c_ch_stress_vcm
+600 20
+1 0
+2 -142.965
+3 2142.79
+4 12968.3
+5 -336.7
+6 2638.09
+7 4214.83
+8 3187.61
+9 -488.891
+10 -49.3553
+11 151.373
+12 -317.663
+13 0
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+700 20
+1 0
+2 -14.3195
+3 -1238.9
+4 30664.3
+5 18805.2
+6 498.562
+7 930.874
+8 660.655
+9 -266.903
+10 -317.877
+11 -386.989
+12 -304.697
+13 0
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+800 20
+1 0
+2 0
+3 -1656.7
+4 30424.3
+5 37003.5
+6 15562.5
+7 -2441.9
+8 -1766.09
+9 272.718
+10 -664.774
+11 -72.6933
+12 -469.765
+13 0
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+900 20
+1 0
+2 0
+3 -1567.21
+4 24987.6
+5 38068.9
+6 31595
+7 8864.94
+8 -3423.99
+9 -753.063
+10 125.21
+11 -50.4895
+12 -172.14
+13 0
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1000 20
+1 0
+2 0
+3 -893.168
+4 15591.6
+5 32690.6
+6 30183
+7 27172
+8 9459.75
+9 -1416.35
+10 -432.731
+11 444.323
+12 -424.357
+13 0
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1100 20
+1 0
+2 0
+3 -601.805
+4 8890.79
+5 23345.1
+6 28529.2
+7 29111.9
+8 25846.2
+9 7451.83
+10 -1624.2
+11 320.704
+12 -50.9865
+13 -5.50481
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1200 20
+1 0
+2 0
+3 1435.39
+4 8818.29
+5 7129.61
+6 20281.7
+7 28026.1
+8 28327.7
+9 26918.6
+10 8277.12
+11 -249.644
+12 -171.806
+13 -7.19065
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1300 20
+1 0
+2 0
+3 -718.118
+4 3021.9
+5 9010.51
+6 9500.87
+7 19432.8
+8 27254.3
+9 28638.5
+10 25568.5
+11 8094.66
+12 -368.293
+13 -2.20997
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1400 20
+1 0
+2 0
+3 -650.581
+4 190.19
+5 5465.38
+6 7489.23
+7 7575.16
+8 18433.5
+9 26975.3
+10 28981.5
+11 26987.9
+12 7502.07
+13 0.117312
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1500 20
+1 0
+2 0
+3 -619.311
+4 561.257
+5 461.5
+6 4105.68
+7 9272.68
+8 10445.6
+9 18826.1
+10 25434.8
+11 25653.8
+12 10981.2
+13 33.682
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1600 20
+1 0
+2 0
+3 -349.345
+4 513.579
+5 -471.384
+6 1257.81
+7 7122.9
+8 8659.35
+9 8452.08
+10 16013.5
+11 17091
+12 5476.24
+13 -136.183
+14 0
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1700 20
+1 0
+2 0
+3 -273.839
+4 -907.407
+5 -272.136
+6 594.363
+7 3302.77
+8 5564.07
+9 8689.92
+10 6446.06
+11 1779.37
+12 338.998
+13 -171.408
+14 -1.21548
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1800 20
+1 0
+2 0
+3 -164.819
+4 383.877
+5 -140.681
+6 -10.0153
+7 907.937
+8 3269.05
+9 5325.22
+10 395.73
+11 -4103.73
+12 -2787.16
+13 -1357.04
+14 -35.2044
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+1900 20
+1 0
+2 0
+3 -80.813
+4 334.225
+5 248.55
+6 82.0566
+7 207.763
+8 185.714
+9 -55.8635
+10 -2758.51
+11 -4619.33
+12 -5521.92
+13 -2346.36
+14 -415.324
+15 0
+16 0
+17 0
+18 0
+19 0
+20 0
+2000 20
+1 0
+2 0
+3 -83.1832
+4 264.023
+5 596.087
+6 40.8157
+7 -267.093
+8 -2288.15
+9 -3387.64
+10 -5566.79
+11 -5640.76
+12 -4925.74
+13 -3096.01
+14 -757.817
+15 -1.13042
+16 0
+17 0
+18 0
+19 0
+20 0
+2100 20
+1 0
+2 0
+3 -17.4378
+4 62.1251
+5 740.988
+6 357.467
+7 -1137.61
+8 -4266.83
+9 -4962.9
+10 -5322.45
+11 -5437.58
+12 -4846.56
+13 -3651.28
+14 -1151.01
+15 -28.3074
+16 0
+17 0
+18 0
+19 0
+20 0
+2200 20
+1 0
+2 0
+3 -10.8779
+4 -56.7926
+5 400.261
+6 -568.63
+7 -2193.36
+8 -3856.71
+9 -6603
+10 -5717.11
+11 -4868.64
+12 -4173.5
+13 -3402.64
+14 -1712.44
+15 -80.6771
+16 -0.123189
+17 0
+18 0
+19 0
+20 0
+2300 20
+1 0
+2 0
+3 -22.8402
+4 -44.5496
+5 -365.476
+6 -1285.6
+7 -2887.76
+8 -4022.77
+9 -6280.86
+10 -6055.26
+11 -4921.51
+12 -4445.37
+13 -3531.69
+14 -1360.49
+15 -258.99
+16 0.196931
+17 0
+18 0
+19 0
+20 0
+2400 20
+1 0
+2 0
+3 -0.594396
+4 -148.921
+5 -1118.18
+6 -2071.85
+7 -3989.41
+8 -4567.01
+9 -4939.36
+10 -5170.94
+11 -4922.25
+12 -4587.5
+13 -3748.19
+14 -1785.46
+15 -460.491
+16 2.54038
+17 0
+18 0
+19 0
+20 0
+2500 20
+1 0
+2 0
+3 5.64755
+4 -485.854
+5 -2525.68
+6 -2642.35
+7 -5066.15
+8 -4546.03
+9 -4429.45
+10 -4579.15
+11 -4829.56
+12 -4384.77
+13 -3525.99
+14 -1708.9
+15 -627.176
+16 -23.5581
+17 0
+18 0
+19 0
+20 0

lib/kokkos/CHANGELOG.md

@@ -1,7 +1,112 @@
 # CHANGELOG
 
+## 4.5.01
+
+[Full Changelog](https://github.com/kokkos/kokkos/compare/4.5.00...4.5.01)
+
+### Bug Fixes
+
+* Fix re-builds after cleaning the binary tree when doing `add_subdirectory` on the Kokkos source [\#7557](https://github.com/kokkos/kokkos/pull/7557)
+* Update mdspan to include fix for submdspan and bracket operator with clang 15&16 [\#7559](https://github.com/kokkos/kokkos/pull/7559)
+* Fix DynRankView performance regression by re-introducing shortcut operator() impls [\#7606](https://github.com/kokkos/kokkos/pull/7606)
+* Add missing MI300A (`GFX942_APU`) option to Makefile build-system
+
+## 4.5.00
+
+[Full Changelog](https://github.com/kokkos/kokkos/compare/4.4.01...4.5.00)
+
+### Features
+
+* SYCL backend graduated to production ready
+* Introduce new `SequentialHostInit` view allocation property [\#7229](https://github.com/kokkos/kokkos/pull/7229) (backported in 4.4.01)
+* Support building with Run-Time Type Information (RTTI) disabled
+* Add new `KOKKOS_RELOCATABLE_FUNCTION` function annotation macro [\#5993](https://github.com/kokkos/kokkos/pull/5993)
+
+### Backend and Architecture Enhancements
+
+#### CUDA
+
+* Adding occupancy tuning for CUDA architectures [\#6788](https://github.com/kokkos/kokkos/pull/6788)
+* By default disable `cudaMallocAsync` (i.e., revert the change made in version 4.2) [\#7353](https://github.com/kokkos/kokkos/pull/7353)
+
+#### HIP
+
+* Add support for AMD Phoenix APUs with Radeon 740M/760M/780M/880M/890M [\#7162](https://github.com/kokkos/kokkos/pull/7162)
+* Update maximum waves per CU values for consumer card [\#7347](https://github.com/kokkos/kokkos/pull/7347)
+* Check that Kokkos is running on the architecture it was compiled for [\#7379](https://github.com/kokkos/kokkos/pull/7379)
+* Add opt-in option to use `hipMallocAsync` instead of `hipMalloc` [\#7324](https://github.com/kokkos/kokkos/pull/7324)
+* Introduce new architecture option `AMD_GFX942_APU` for MI300A [\#7462](https://github.com/kokkos/kokkos/pull/7462)
+
+#### SYCL
+
+* Move the `SYCL` backend out of the `Experimental` namespace [\#7171](https://github.com/kokkos/kokkos/pull/7171)
+* Introduce `KOKKOS_ENABLE_SYCL_RELOCATABLE_DEVICE_CODE` as CMake option [\#5993](https://github.com/kokkos/kokkos/pull/5993)
+
+#### OpenACC
+
+* Add support for building with the Clacc compiler [\#7198](https://github.com/kokkos/kokkos/pull/7198)
+* Workaround NVHPC collapse clause bug for `MDRangePolicy` [\#7425](https://github.com/kokkos/kokkos/pull/7425)
+
+#### HPX
+
+* Implement `Experimental::partition_space` to produce truly independent execution spaces [\#7287](https://github.com/kokkos/kokkos/pull/7287)
+
+#### Threads
+
+* Fix compilation for `parallel_reduce` `MDRange` with `Dynamic` scheduling [\#7478](https://github.com/kokkos/kokkos/pull/7478)
+* Fix race conditions on ARM architectures [\#7498](https://github.com/kokkos/kokkos/pull/7498)
+
+#### OpenMP
+
+* Fix run time behavior when compiling with `-fvisibility-hidden` [\#7284](https://github.com/kokkos/kokkos/pull/7284) (backported in 4.4.01)
+* Fix linking with Cray Clang compiler [\#7341](https://github.com/kokkos/kokkos/pull/7341)
+
+#### Serial
+
+* Allow `Kokkos_ENABLE_ATOMICS_BYPASS` to skip mutexes to remediate performance regression in 4.4 [\#7369](https://github.com/kokkos/kokkos/pull/7369)
+
+### General Enhancements
+
+* Improve `View` initialization/destruction for non-scalar trivial and trivially-destructible types [\#7219](https://github.com/kokkos/kokkos/pull/7219) [\#7225](https://github.com/kokkos/kokkos/pull/7225)
+* Add getters for default tile sizes used in `MDRangePolicy` [\#6839](https://github.com/kokkos/kokkos/pull/6839)
+* Improve performance of `Kokkos::sort` when `std::sort` is used [\#7264](https://github.com/kokkos/kokkos/pull/7264)
+* Add range-based for loop support for `Array<T, N>` [\#7293](https://github.com/kokkos/kokkos/pull/7293)
+* Allow functors as reducers for nested team parallel reduce [\#6921](https://github.com/kokkos/kokkos/pull/6921)
+* Avoid making copies of string rvalue reference arguments to `view_alloc()` [\#7364](https://github.com/kokkos/kokkos/pull/7364)
+* Add `atomic_{mod,xor,nand,lshift,rshift}` [\#7458](https://github.com/kokkos/kokkos/pull/7458)
+* Allow using `SequentialHostInit` with `Kokkos::DualView` [\#7456](https://github.com/kokkos/kokkos/pull/7456)
+* Add `Graph::instantiate()` [\#7240](https://github.com/kokkos/kokkos/pull/7240)
+* Allow an arbitrary execution space instance to be used in `Kokkos::Graph::submit()` [\#7249](https://github.com/kokkos/kokkos/pull/7249)
+* Enable compile-time diagnostic of illegal reduction target for graphs [\#7460](https://github.com/kokkos/kokkos/pull/7460)
+
+### Build System Changes
+
+* Make sure backend-specific options such as `IMPL_CUDA_MALLOC_ASYNC` only show when that backend is actually enabled [\#7228](https://github.com/kokkos/kokkos/pull/7228)
+* Major refactoring removing `TriBITS` paths [\#6164](https://github.com/kokkos/kokkos/pull/6164)
+* Add support for SpacemiT K60 (RISC-V) [\#7160](https://github.com/kokkos/kokkos/pull/7160)
+
+### Deprecations
+
+* Deprecate Tasking interface [\#7393](https://github.com/kokkos/kokkos/pull/7393)
+* Deprecate `atomic_query_version`, `atomic_assign`, `atomic_compare_exchange_strong`, `atomic_{inc, dec}rement` [\#7458](https://github.com/kokkos/kokkos/pull/7458)
+* Deprecate `{OpenMP,HPX}::is_asynchronous()` [\#7322](https://github.com/kokkos/kokkos/pull/7322)
+
+### Bug Fixes
+
+* Fix undefined behavior in `BinSort` when sorting within bins on host [\#7223](https://github.com/kokkos/kokkos/pull/7223)
+* Using CUDA limits to set extents for blocks, grids [\#7235](https://github.com/kokkos/kokkos/pull/7235)
+* Fix `deep_copy (serial_exec, dst, src)` with multiple host backends [\#7245](https://github.com/kokkos/kokkos/pull/7245)
+* Skip `RangePolicy` bounds conversion checks if roundtrip convertibility is not provided [\#7172](https://github.com/kokkos/kokkos/pull/7172)
+* Allow extracting host and device views from `DualView` with `const` value type [\#7242](https://github.com/kokkos/kokkos/pull/7242)
+* Fix `TeamPolicy` array reduction for CUDA and HIP [\#6296](https://github.com/kokkos/kokkos/pull/6296)
+* Fix implicit copy assignment operators in few AVX2 masks being deleted [\#7296](https://github.com/kokkos/kokkos/pull/7296)
+* Fix configuring without architecture flags for SYCL [\#7303](https://github.com/kokkos/kokkos/pull/7303)
+* Set an initial value index during join of `MinLoc`, `MaxLoc` or `MinMaxLoc` [\#7330](https://github.com/kokkos/kokkos/pull/7330)
+* Fix storage lifetime of driver for global launch of graph nodes for CUDA and HIP [\#7365](https://github.com/kokkos/kokkos/pull/7365)
+* Make `value_type` for `RandomAccessIterator` non-`const` [\#7485](https://github.com/kokkos/kokkos/pull/7485)
+
 ## [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)
+[Full Changelog](https://github.com/kokkos/kokkos/compare/4.4.00...4.4.01)
 
 ### Features:
 * Introduce new SequentialHostInit view allocation property [\#7229](https://github.com/kokkos/kokkos/pull/7229)
@@ -13,7 +118,7 @@
 
 ### 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)
+* 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)
@@ -57,6 +162,7 @@
 * 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)
+* Introduce `KOKKOS_DEDUCTION_GUIDE` macro to allow for portable user-defined deduction guides [\#6954](https://github.com/kokkos/kokkos/pull/6954)
 
 ### Build System Changes
 * Do not require OpenMP support for languages other than CXX [\#6965](https://github.com/kokkos/kokkos/pull/6965)
@@ -1388,7 +1494,7 @@
 **Closed issues:**
 
 - Silent error (Validate storage level arg to set_scratch_size) [\#3097](https://github.com/kokkos/kokkos/issues/3097)
-- Remove KOKKKOS\_ENABLE\_PROFILING Option [\#3095](https://github.com/kokkos/kokkos/issues/3095)
+- Remove KOKKOS\_ENABLE\_PROFILING Option [\#3095](https://github.com/kokkos/kokkos/issues/3095)
 - Cuda 11 -\> allow C++17 [\#3083](https://github.com/kokkos/kokkos/issues/3083)
 - In source build failure not explained [\#3081](https://github.com/kokkos/kokkos/issues/3081)
 - Allow naming of Views for initialization kernel [\#3070](https://github.com/kokkos/kokkos/issues/3070)

lib/kokkos/CMakeLists.txt

@@ -1,12 +1,11 @@
 cmake_minimum_required(VERSION 3.16 FATAL_ERROR)
 
 # Disable in-source builds to prevent source tree corruption.
-if( "${CMAKE_SOURCE_DIR}" STREQUAL "${CMAKE_BINARY_DIR}" )
+if("${CMAKE_SOURCE_DIR}" STREQUAL "${CMAKE_BINARY_DIR}")
-  message( FATAL_ERROR "FATAL: In-source builds are not allowed. You should create a separate directory for build files and delete CMakeCache.txt." )
+  message(
-endif()
+    FATAL_ERROR
-
+      "FATAL: In-source builds are not allowed. You should create a separate directory for build files and delete CMakeCache.txt."
-if (COMMAND TRIBITS_PACKAGE)
+  )
-  TRIBITS_PACKAGE(Kokkos)
 endif()
 
 # We want to determine if options are given with the wrong case
@@ -15,142 +14,141 @@ endif()
 # form a list of all the given variables. If it begins with any
 # case of KoKkOS, we add it to the list.
 
-GET_CMAKE_PROPERTY(_variableNames VARIABLES)
+get_cmake_property(_variableNames VARIABLES)
-SET(KOKKOS_GIVEN_VARIABLES)
+set(KOKKOS_GIVEN_VARIABLES)
-FOREACH (var ${_variableNames})
+foreach(var ${_variableNames})
-  STRING(TOUPPER ${var} UC_VAR)
+  string(TOUPPER ${var} UC_VAR)
-  STRING(FIND ${UC_VAR} KOKKOS IDX)
+  string(FIND ${UC_VAR} KOKKOS IDX)
-  IF (${IDX} EQUAL 0)
+  if(${IDX} EQUAL 0)
-    LIST(APPEND KOKKOS_GIVEN_VARIABLES ${var})
+    list(APPEND KOKKOS_GIVEN_VARIABLES ${var})
-  ENDIF()
+  endif()
-ENDFOREACH()
+endforeach()
 
 # Basic initialization (Used in KOKKOS_SETTINGS)
-SET(Kokkos_SOURCE_DIR    ${CMAKE_CURRENT_SOURCE_DIR})
+set(Kokkos_SOURCE_DIR ${CMAKE_CURRENT_SOURCE_DIR})
-SET(KOKKOS_SOURCE_DIR    ${CMAKE_CURRENT_SOURCE_DIR})
+set(KOKKOS_SOURCE_DIR ${CMAKE_CURRENT_SOURCE_DIR})
-SET(KOKKOS_SRC_PATH      ${Kokkos_SOURCE_DIR})
+set(KOKKOS_SRC_PATH ${Kokkos_SOURCE_DIR})
-SET(KOKKOS_PATH          ${Kokkos_SOURCE_DIR})
+set(KOKKOS_PATH ${Kokkos_SOURCE_DIR})
-SET(KOKKOS_TOP_BUILD_DIR ${CMAKE_CURRENT_BINARY_DIR})
+set(KOKKOS_TOP_BUILD_DIR ${CMAKE_CURRENT_BINARY_DIR})
+
+set(PACKAGE_NAME Kokkos)
+set(PACKAGE_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}")
 
-# Is this a build as part of Trilinos?
-IF(COMMAND TRIBITS_PACKAGE_DECL)
-  SET(KOKKOS_HAS_TRILINOS ON)
-ELSE()
-  SET(KOKKOS_HAS_TRILINOS OFF)
-  SET(PACKAGE_NAME Kokkos)
-  SET(PACKAGE_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}")
-ENDIF()
 # Is this build a subdirectory of another project
-GET_DIRECTORY_PROPERTY(HAS_PARENT PARENT_DIRECTORY)
+get_directory_property(HAS_PARENT PARENT_DIRECTORY)
 
+include(${KOKKOS_SRC_PATH}/cmake/kokkos_functions.cmake)
+include(${KOKKOS_SRC_PATH}/cmake/kokkos_pick_cxx_std.cmake)
 
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_functions.cmake)
+set(KOKKOS_ENABLED_OPTIONS) #exported in config file
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_pick_cxx_std.cmake)
+set(KOKKOS_ENABLED_DEVICES) #exported in config file
-
+set(KOKKOS_ENABLED_TPLS) #exported in config file
-SET(KOKKOS_ENABLED_OPTIONS)      #exported in config file
+set(KOKKOS_ENABLED_ARCH_LIST) #exported in config file
-SET(KOKKOS_ENABLED_DEVICES)      #exported in config file
-SET(KOKKOS_ENABLED_TPLS)         #exported in config file
-SET(KOKKOS_ENABLED_ARCH_LIST)    #exported in config file
 
 #These are helper flags used for sanity checks during config
 #Certain features should depend on other features being configured first
-SET(KOKKOS_CFG_DAG_NONE              On) #sentinel to indicate no dependencies
+set(KOKKOS_CFG_DAG_NONE On) #sentinel to indicate no dependencies
-SET(KOKKOS_CFG_DAG_DEVICES_DONE      Off)
+set(KOKKOS_CFG_DAG_DEVICES_DONE Off)
-SET(KOKKOS_CFG_DAG_OPTIONS_DONE      Off)
+set(KOKKOS_CFG_DAG_OPTIONS_DONE Off)
-SET(KOKKOS_CFG_DAG_ARCH_DONE         Off)
+set(KOKKOS_CFG_DAG_ARCH_DONE Off)
-SET(KOKKOS_CFG_DAG_CXX_STD_DONE      Off)
+set(KOKKOS_CFG_DAG_CXX_STD_DONE Off)
-SET(KOKKOS_CFG_DAG_COMPILER_ID_DONE  Off)
+set(KOKKOS_CFG_DAG_COMPILER_ID_DONE Off)
-FUNCTION(KOKKOS_CFG_DEPENDS SUCCESSOR PRECURSOR)
+function(KOKKOS_CFG_DEPENDS SUCCESSOR PRECURSOR)
-  SET(PRE_FLAG  KOKKOS_CFG_DAG_${PRECURSOR})
+  set(PRE_FLAG KOKKOS_CFG_DAG_${PRECURSOR})
-  SET(POST_FLAG KOKKOS_CFG_DAG_${SUCCESSOR})
+  set(POST_FLAG KOKKOS_CFG_DAG_${SUCCESSOR})
-  IF (NOT ${PRE_FLAG})
+  if(NOT ${PRE_FLAG})
-    MESSAGE(FATAL_ERROR "Bad CMake refactor: feature ${SUCCESSOR} cannot be configured until ${PRECURSOR} is configured")
+    message(
-  ENDIF()
+      FATAL_ERROR "Bad CMake refactor: feature ${SUCCESSOR} cannot be configured until ${PRECURSOR} is configured"
-  GLOBAL_SET(${POST_FLAG} On)
+    )
-ENDFUNCTION()
+  endif()
+  global_set(${POST_FLAG} On)
+endfunction()
 
+list(APPEND CMAKE_MODULE_PATH cmake/Modules)
 
-LIST(APPEND CMAKE_MODULE_PATH cmake/Modules)
+set(CMAKE_DISABLE_SOURCE_CHANGES ON)
+set(CMAKE_DISABLE_IN_SOURCE_BUILD ON)
 
-IF(NOT KOKKOS_HAS_TRILINOS)
+# What language are we compiling Kokkos as
-  set(CMAKE_DISABLE_SOURCE_CHANGES ON)
+# downstream dependencies need to match this!
-  set(CMAKE_DISABLE_IN_SOURCE_BUILD ON)
+set(KOKKOS_COMPILE_LANGUAGE CXX)
+# use lower case here since we didn't parse options yet
+if(Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE AND Kokkos_ENABLE_CUDA)
 
-  # What language are we compiling Kokkos as
+  # Without this as a language for the package we would get a C++ compiler enabled.
-  # downstream dependencies need to match this!
+  # but we still need a C++ compiler even if we build all our cpp files as CUDA only
-  SET(KOKKOS_COMPILE_LANGUAGE CXX)
+  # because otherwise the C++ features don't work etc.
-  # use lower case here since we didn't parse options yet
+  # This is just the rather odd way CMake does this, since CUDA doesn't imply C++ even
-  IF (Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE AND Kokkos_ENABLE_CUDA)
+  # though it is a C++ extension ... (but I guess it didn't use to be back in CUDA 4 or 5
+  # days.
+  set(KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE CXX)
 
-    # Without this as a language for the package we would get a C++ compiler enabled.
+  set(KOKKOS_COMPILE_LANGUAGE CUDA)
-    # but we still need a C++ compiler even if we build all our cpp files as CUDA only
+endif()
-    # because otherwise the C++ features don't work etc.
+# use lower case here since we haven't parsed options yet
-    # This is just the rather odd way CMake does this, since CUDA doesn't imply C++ even
+if(Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE AND Kokkos_ENABLE_HIP)
-    # though it is a C++ extension ... (but I guess it didn't use to be back in CUDA 4 or 5
-    # days.
-    SET(KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE CXX)
 
-    SET(KOKKOS_COMPILE_LANGUAGE CUDA)
+  # Without this as a language for the package we would get a C++ compiler enabled.
-  ENDIF()
+  # but we still need a C++ compiler even if we build all our cpp files as HIP only
-  # use lower case here since we haven't parsed options yet
+  # because otherwise the C++ features don't work etc.
-  IF (Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE AND Kokkos_ENABLE_HIP)
+  set(KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE CXX)
 
-    # Without this as a language for the package we would get a C++ compiler enabled.
+  set(KOKKOS_COMPILE_LANGUAGE HIP)
-    # but we still need a C++ compiler even if we build all our cpp files as HIP only
+endif()
-    # because otherwise the C++ features don't work etc.
-    SET(KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE CXX)
 
-    SET(KOKKOS_COMPILE_LANGUAGE HIP)
+if(Spack_WORKAROUND)
-  ENDIF()
+  if(Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE)
+    message(FATAL_ERROR "Can't currently use Kokkos_ENABLE_COMPILER_AS_CMAKE_LANGUAGE in a spack installation!")
+  endif()
 
-  IF (Spack_WORKAROUND)
+  #if we are explicitly using Spack for development,
-    IF (Kokkos_ENABLE_COMPILE_AS_CMAKE_LANGUAGE)
+  #nuke the Spack compiler
-      MESSAGE(FATAL_ERROR "Can't currently use Kokkos_ENABLE_COMPILER_AS_CMAKE_LANGUAGE in a spack installation!")
+  set(SPACK_CXX $ENV{SPACK_CXX})
-    ENDIF()
+  if(SPACK_CXX)
-
+    set(CMAKE_CXX_COMPILER ${SPACK_CXX} CACHE STRING "the C++ compiler" FORCE)
-    #if we are explicitly using Spack for development,
+    set(ENV{CXX} ${SPACK_CXX})
-    #nuke the Spack compiler
+  endif()
-    SET(SPACK_CXX $ENV{SPACK_CXX})
+endif()
-    IF(SPACK_CXX)
+# Always call the project command to define Kokkos_ variables
-      SET(CMAKE_CXX_COMPILER ${SPACK_CXX} CACHE STRING "the C++ compiler" FORCE)
+# and to make sure that C++ is an enabled language
-      SET(ENV{CXX} ${SPACK_CXX})
+project(Kokkos ${KOKKOS_COMPILE_LANGUAGE} ${KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE})
-    ENDIF()
+if(NOT HAS_PARENT)
-  ENDIF()
+  if(NOT CMAKE_BUILD_TYPE)
-  # Always call the project command to define Kokkos_ variables
+    set(DEFAULT_BUILD_TYPE "RelWithDebInfo")
-  # and to make sure that C++ is an enabled language
+    message(STATUS "Setting build type to '${DEFAULT_BUILD_TYPE}' as none was specified.")
-  PROJECT(Kokkos ${KOKKOS_COMPILE_LANGUAGE} ${KOKKOS_INTERNAL_EXTRA_COMPILE_LANGUAGE})
+    set(CMAKE_BUILD_TYPE "${DEFAULT_BUILD_TYPE}"
-  IF(NOT HAS_PARENT)
+        CACHE STRING "Choose the type of build, options are: Debug, Release, RelWithDebInfo and MinSizeRel." FORCE
-    IF (NOT CMAKE_BUILD_TYPE)
+    )
-      SET(DEFAULT_BUILD_TYPE "RelWithDebInfo")
+  endif()
-      MESSAGE(STATUS "Setting build type to '${DEFAULT_BUILD_TYPE}' as none was specified.")
+endif()
-      SET(CMAKE_BUILD_TYPE "${DEFAULT_BUILD_TYPE}" CACHE STRING
-          "Choose the type of build, options are: Debug, Release, RelWithDebInfo and MinSizeRel."
-          FORCE)
-    ENDIF()
-  ENDIF()
-ELSE()
-  SET(KOKKOS_COMPILE_LANGUAGE CXX)
-ENDIF()
-
-IF (NOT CMAKE_SIZEOF_VOID_P)
-  STRING(FIND ${CMAKE_CXX_COMPILER} nvcc_wrapper FIND_IDX)
-  IF (NOT FIND_IDX STREQUAL -1)
-    MESSAGE(FATAL_ERROR "Kokkos did not configure correctly and failed to validate compiler. The most likely cause is CUDA linkage using nvcc_wrapper. Please ensure your CUDA environment is correctly configured.")
-  ELSE()
-    MESSAGE(FATAL_ERROR "Kokkos did not configure correctly and failed to validate compiler. The most likely cause is linkage errors during CMake compiler validation. Please consult the CMake error log shown below for the exact error during compiler validation")
-  ENDIF()
-ELSEIF (NOT CMAKE_SIZEOF_VOID_P EQUAL 8)
-  IF(CMAKE_SIZEOF_VOID_P EQUAL 4)
-    MESSAGE(WARNING "32-bit builds are experimental and not officially supported.")
-    SET(KOKKOS_IMPL_32BIT ON)
-  ELSE()
-    MESSAGE(FATAL_ERROR "Kokkos assumes a 64-bit build, i.e., 8-byte pointers, but found ${CMAKE_SIZEOF_VOID_P}-byte pointers instead;")
-  ENDIF()
-ENDIF()
 
+if(NOT CMAKE_SIZEOF_VOID_P)
+  string(FIND ${CMAKE_CXX_COMPILER} nvcc_wrapper FIND_IDX)
+  if(NOT FIND_IDX STREQUAL -1)
+    message(
+      FATAL_ERROR
+        "Kokkos did not configure correctly and failed to validate compiler. The most likely cause is CUDA linkage using nvcc_wrapper. Please ensure your CUDA environment is correctly configured."
+    )
+  else()
+    message(
+      FATAL_ERROR
+        "Kokkos did not configure correctly and failed to validate compiler. The most likely cause is linkage errors during CMake compiler validation. Please consult the CMake error log shown below for the exact error during compiler validation"
+    )
+  endif()
+elseif(NOT CMAKE_SIZEOF_VOID_P EQUAL 8)
+  if(CMAKE_SIZEOF_VOID_P EQUAL 4)
+    message(WARNING "32-bit builds are experimental and not officially supported.")
+    set(KOKKOS_IMPL_32BIT ON)
+  else()
+    message(
+      FATAL_ERROR
+        "Kokkos assumes a 64-bit build, i.e., 8-byte pointers, but found ${CMAKE_SIZEOF_VOID_P}-byte pointers instead;"
+    )
+  endif()
+endif()
 
 set(Kokkos_VERSION_MAJOR 4)
-set(Kokkos_VERSION_MINOR 4)
+set(Kokkos_VERSION_MINOR 5)
 set(Kokkos_VERSION_PATCH 1)
 set(Kokkos_VERSION "${Kokkos_VERSION_MAJOR}.${Kokkos_VERSION_MINOR}.${Kokkos_VERSION_PATCH}")
 message(STATUS "Kokkos version: ${Kokkos_VERSION}")
@@ -164,58 +162,54 @@ math(EXPR KOKKOS_VERSION_PATCH "${KOKKOS_VERSION} % 100")
 
 # Load either the real TriBITS or a TriBITS wrapper
 # for certain utility functions that are universal (like GLOBAL_SET)
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/fake_tribits.cmake)
+include(${KOKKOS_SRC_PATH}/cmake/fake_tribits.cmake)
 
-IF (Kokkos_ENABLE_CUDA)
+if(Kokkos_ENABLE_CUDA)
   # If we are building CUDA, we have tricked CMake because we declare a CXX project
   # If the default C++ standard for a given compiler matches the requested
   # standard, then CMake just omits the -std flag in later versions of CMake
   # This breaks CUDA compilation (CUDA compiler can have a different default
   # -std then the underlying host compiler by itself). Setting this variable
   # forces CMake to always add the -std flag even if it thinks it doesn't need it
-  GLOBAL_SET(CMAKE_CXX_STANDARD_DEFAULT 98)
+  global_set(CMAKE_CXX_STANDARD_DEFAULT 98)
-ENDIF()
+endif()
 
 # These are the variables we will append to as we go
 # I really wish these were regular variables
 # but scoping issues can make it difficult
-GLOBAL_SET(KOKKOS_COMPILE_OPTIONS)
+global_set(KOKKOS_COMPILE_OPTIONS)
-GLOBAL_SET(KOKKOS_LINK_OPTIONS)
+global_set(KOKKOS_LINK_OPTIONS)
-GLOBAL_SET(KOKKOS_AMDGPU_OPTIONS)
+global_set(KOKKOS_AMDGPU_OPTIONS)
-GLOBAL_SET(KOKKOS_CUDA_OPTIONS)
+global_set(KOKKOS_CUDA_OPTIONS)
-GLOBAL_SET(KOKKOS_CUDAFE_OPTIONS)
+global_set(KOKKOS_CUDAFE_OPTIONS)
-GLOBAL_SET(KOKKOS_XCOMPILER_OPTIONS)
+global_set(KOKKOS_XCOMPILER_OPTIONS)
 # We need to append text here for making sure TPLs
 # we import are available for an installed Kokkos
-GLOBAL_SET(KOKKOS_TPL_EXPORTS)
+global_set(KOKKOS_TPL_EXPORTS)
 # KOKKOS_DEPENDENCE is used by kokkos_launch_compiler
-GLOBAL_SET(KOKKOS_COMPILE_DEFINITIONS KOKKOS_DEPENDENCE)
+global_set(KOKKOS_COMPILE_DEFINITIONS KOKKOS_DEPENDENCE)
 # MSVC never goes through kokkos_launch_compiler
-IF(NOT MSVC)
+if(NOT MSVC)
-    GLOBAL_APPEND(KOKKOS_LINK_OPTIONS -DKOKKOS_DEPENDENCE)
+  global_append(KOKKOS_LINK_OPTIONS -DKOKKOS_DEPENDENCE)
-ENDIF()
+endif()
 
-IF(Kokkos_ENABLE_TESTS AND NOT KOKKOS_HAS_TRILINOS)
+include(${CMAKE_CURRENT_SOURCE_DIR}/cmake/kokkos_configure_trilinos.cmake)
+
+if(Kokkos_ENABLE_TESTS)
   find_package(GTest QUIET)
-ENDIF()
+endif()
 
 # Include a set of Kokkos-specific wrapper functions that
 # will either call raw CMake or TriBITS
 # These are functions like KOKKOS_INCLUDE_DIRECTORIES
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_tribits.cmake)
+include(${KOKKOS_SRC_PATH}/cmake/kokkos_tribits.cmake)
-
 
 # Check the environment and set certain variables
 # to allow platform-specific checks
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_check_env.cmake)
+include(${KOKKOS_SRC_PATH}/cmake/kokkos_check_env.cmake)
 
-IF(NOT KOKKOS_HAS_TRILINOS)
+include(${KOKKOS_SRC_PATH}/cmake/build_env_info.cmake)
-  # This does not work in Trilinos and we simply don't care
+check_git_setup()
-  # to fix it for Trilinos
-  # Gather information about the runtime environment
-  INCLUDE(${KOKKOS_SRC_PATH}/cmake/build_env_info.cmake)
-  check_git_setup()
-ENDIF()
 
 # The build environment setup goes in the following steps
 # 1) Check all the enable options. This includes checking Kokkos_DEVICES
@@ -223,102 +217,54 @@ ENDIF()
 # 3) Check the CXX standard and select important CXX flags
 # 4) Check for any third-party libraries (TPLs) like hwloc
 # 5) Check if optimizing for a particular architecture and add arch-specific flags
-KOKKOS_SETUP_BUILD_ENVIRONMENT()
+kokkos_setup_build_environment()
 
 # Finish off the build
 # 6) Recurse into subdirectories and configure individual libraries
 # 7) Export and install targets
 
-OPTION(BUILD_SHARED_LIBS "Build shared libraries" OFF)
+option(BUILD_SHARED_LIBS "Build shared libraries" OFF)
 
-SET(KOKKOS_COMPONENT_LIBRARIES kokkoscore kokkoscontainers kokkosalgorithms kokkossimd)
+set(KOKKOS_COMPONENT_LIBRARIES kokkoscore kokkoscontainers kokkosalgorithms kokkossimd)
-SET_PROPERTY(GLOBAL PROPERTY KOKKOS_INT_LIBRARIES kokkos ${KOKKOS_COMPONENT_LIBRARIES})
+set_property(GLOBAL PROPERTY KOKKOS_INT_LIBRARIES kokkos ${KOKKOS_COMPONENT_LIBRARIES})
 
-IF (KOKKOS_HAS_TRILINOS)
+if(HAS_PARENT)
-  SET(TRILINOS_INCDIR ${${PROJECT_NAME}_INSTALL_INCLUDE_DIR})
+  set(KOKKOS_HEADER_DIR "include/kokkos")
-  SET(KOKKOS_HEADER_DIR ${TRILINOS_INCDIR})
+  set(KOKKOS_IS_SUBDIRECTORY TRUE)
-  SET(KOKKOS_IS_SUBDIRECTORY TRUE)
+else()
-ELSEIF(HAS_PARENT)
+  set(KOKKOS_HEADER_DIR "${CMAKE_INSTALL_INCLUDEDIR}")
-  SET(KOKKOS_HEADER_DIR "include/kokkos")
+  set(KOKKOS_IS_SUBDIRECTORY FALSE)
-  SET(KOKKOS_IS_SUBDIRECTORY TRUE)
+endif()
-ELSE()
-  SET(KOKKOS_HEADER_DIR "${CMAKE_INSTALL_INCLUDEDIR}")
-  SET(KOKKOS_IS_SUBDIRECTORY FALSE)
-ENDIF()
 
 #------------------------------------------------------------------------------
 #
 # A) Forward declare the package so that certain options are also defined for
 # subpackages
 
-## This restores the old behavior of ProjectCompilerPostConfig.cmake
-# We must do this before KOKKOS_PACKAGE_DECL
-IF (KOKKOS_HAS_TRILINOS)
-  # Overwrite the old flags at the top-level
-  # Because Tribits doesn't use lists, it uses spaces for the list of CXX flags
-  # we have to match the annoying behavior, also we have to preserve quotes
-  # which needs another workaround.
-  SET(KOKKOS_COMPILE_OPTIONS_TMP)
-  IF (KOKKOS_ENABLE_HIP)
-    LIST(APPEND KOKKOS_COMPILE_OPTIONS ${KOKKOS_AMDGPU_OPTIONS})
-  ENDIF()
-  FOREACH(OPTION ${KOKKOS_COMPILE_OPTIONS})
-    STRING(FIND "${OPTION}" " " OPTION_HAS_WHITESPACE)
-    IF(OPTION_HAS_WHITESPACE EQUAL -1)
-      LIST(APPEND KOKKOS_COMPILE_OPTIONS_TMP "${OPTION}")
-    ELSE()
-      LIST(APPEND KOKKOS_COMPILE_OPTIONS_TMP "\"${OPTION}\"")
-    ENDIF()
-  ENDFOREACH()
-  STRING(REPLACE ";" " " KOKKOSCORE_COMPILE_OPTIONS "${KOKKOS_COMPILE_OPTIONS_TMP}")
-  LIST(APPEND KOKKOS_ALL_COMPILE_OPTIONS ${KOKKOS_COMPILE_OPTIONS})
-  IF (KOKKOS_ENABLE_CUDA)
-    LIST(APPEND KOKKOS_ALL_COMPILE_OPTIONS ${KOKKOS_CUDA_OPTIONS})
-  ENDIF()
-  FOREACH(XCOMP_FLAG ${KOKKOS_XCOMPILER_OPTIONS})
-    SET(KOKKOSCORE_XCOMPILER_OPTIONS "${KOKKOSCORE_XCOMPILER_OPTIONS} -Xcompiler ${XCOMP_FLAG}")
-    LIST(APPEND KOKKOS_ALL_COMPILE_OPTIONS -Xcompiler ${XCOMP_FLAG})
-  ENDFOREACH()
-  IF (KOKKOS_ENABLE_CUDA)
-    STRING(REPLACE ";" " " KOKKOSCORE_CUDA_OPTIONS    "${KOKKOS_CUDA_OPTIONS}")
-    FOREACH(CUDAFE_FLAG ${KOKKOS_CUDAFE_OPTIONS})
-      SET(KOKKOSCORE_CUDAFE_OPTIONS "${KOKKOSCORE_CUDAFE_OPTIONS} -Xcudafe ${CUDAFE_FLAG}")
-      LIST(APPEND KOKKOS_ALL_COMPILE_OPTIONS -Xcudafe ${CUDAFE_FLAG})
-    ENDFOREACH()
-  ENDIF()
-  #These flags get set up in KOKKOS_PACKAGE_DECL, which means they
-  #must be configured before KOKKOS_PACKAGE_DECL
-  SET(KOKKOS_ALL_COMPILE_OPTIONS
-    $<$<COMPILE_LANGUAGE:CXX>:${KOKKOS_ALL_COMPILE_OPTIONS}>)
-ENDIF()
-
-
 #------------------------------------------------------------------------------
 #
 # D) Process the subpackages (subdirectories) for Kokkos
 #
-KOKKOS_PROCESS_SUBPACKAGES()
+kokkos_process_subpackages()
-
 
 #------------------------------------------------------------------------------
 #
 # E) If Kokkos itself is enabled, process the Kokkos package
 #
 
-KOKKOS_PACKAGE_POSTPROCESS()
+kokkos_configure_core()
-KOKKOS_CONFIGURE_CORE()
 
-IF (NOT KOKKOS_HAS_TRILINOS AND NOT Kokkos_INSTALL_TESTING)
+if(NOT Kokkos_INSTALL_TESTING)
-  ADD_LIBRARY(kokkos INTERFACE)
+  add_library(kokkos INTERFACE)
   #Make sure in-tree projects can reference this as Kokkos::
   #to match the installed target names
-  ADD_LIBRARY(Kokkos::kokkos ALIAS kokkos)
+  add_library(Kokkos::kokkos ALIAS kokkos)
   # all_libs target is required for TriBITS-compliance
-  ADD_LIBRARY(Kokkos::all_libs ALIAS kokkos)
+  add_library(Kokkos::all_libs ALIAS kokkos)
-  TARGET_LINK_LIBRARIES(kokkos INTERFACE ${KOKKOS_COMPONENT_LIBRARIES})
+  target_link_libraries(kokkos INTERFACE ${KOKKOS_COMPONENT_LIBRARIES})
-  KOKKOS_INTERNAL_ADD_LIBRARY_INSTALL(kokkos)
+  kokkos_internal_add_library_install(kokkos)
-ENDIF()
+endif()
-INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_install.cmake)
+include(${KOKKOS_SRC_PATH}/cmake/kokkos_install.cmake)
 
 # nvcc_wrapper is Kokkos' wrapper for NVIDIA's NVCC CUDA compiler.
 # Kokkos needs nvcc_wrapper in order to build.  Other libraries and
@@ -327,16 +273,15 @@ INCLUDE(${KOKKOS_SRC_PATH}/cmake/kokkos_install.cmake)
 # as relative to ${CMAKE_INSTALL_PATH}.
 # KOKKOS_INSTALL_ADDITIONAL_FILES will install nvcc wrapper and other generated
 # files
-KOKKOS_INSTALL_ADDITIONAL_FILES()
+kokkos_install_additional_files()
-
 
 #  Finally - if we are a subproject - make sure the enabled devices are visible
-IF (HAS_PARENT)
+if(HAS_PARENT)
-  FOREACH(DEV Kokkos_ENABLED_DEVICES)
+  foreach(DEV Kokkos_ENABLED_DEVICES)
     #I would much rather not make these cache variables or global properties, but I can't
     #make any guarantees on whether PARENT_SCOPE is good enough to make
     #these variables visible where I need them
-    SET(Kokkos_ENABLE_${DEV} ON PARENT_SCOPE)
+    set(Kokkos_ENABLE_${DEV} ON PARENT_SCOPE)
-    SET_PROPERTY(GLOBAL PROPERTY Kokkos_ENABLE_${DEV} ON)
+    set_property(GLOBAL PROPERTY Kokkos_ENABLE_${DEV} ON)
-  ENDFOREACH()
+  endforeach()
-ENDIF()
+endif()

lib/kokkos/CONTRIBUTING.md

@@ -7,6 +7,8 @@ We actively welcome pull requests.
 3. If you've changed APIs, update the documentation.
 4. Ensure the test suite passes.
 
+Before sending your patch for review, please try to ensure that it is formatted properly. We use clang-format version 16 for this.
+
 ## Issues
 We use GitHub issues to track public bugs. Please ensure your description is clear and has sufficient instructions to be able to reproduce the issue.
 

lib/kokkos/HOW_TO_SNAPSHOT

@@ -1,73 +0,0 @@
-
-Developers of Kokkos (those who commit modifications to Kokkos)
-must maintain the snapshot of Kokkos in the Trilinos repository.
-
-This file contains instructions for how to
-snapshot Kokkos from github.com/kokkos to Trilinos.
-
-------------------------------------------------------------------------
-*** EVERYTHING GOES RIGHT WORKFLOW ***
-
-1) Given a 'git clone' of Kokkos and of Trilinos repositories.
-1.1) Let ${KOKKOS} be the absolute path to the Kokkos clone.
-     This path *must* terminate with the directory name 'kokkos';
-     e.g., ${HOME}/kokkos .
-1.2) Let ${TRILINOS} be the absolute path to the Trilinos directory.
-
-2) Given that the Kokkos build & test is clean and
-   changes are committed to the Kokkos clone.
-
-3) Snapshot the current commit in the Kokkos clone into the Trilinos clone.
-   This overwrites ${TRILINOS}/packages/kokkos with the content of ${KOKKOS}:
-	${KOKKOS}/scripts/snapshot.py --verbose ${KOKKOS} ${TRILINOS}/packages
-
-4) Verify the snapshot commit happened as expected
-	cd ${TRILINOS}/packages/kokkos
-	git log -1 --name-only
-
-5) Modify, build, and test Trilinos with the Kokkos snapshot.
-
-6) Given that that the Trilinos build & test is clean and
-   changes are committed to the Trilinos clone.
-
-7) Attempt push to the Kokkos repository.
-   If push fails then you must 'remove the Kokkos snapshot'
-   from your Trilinos clone.
-   See below.
-
-8) Attempt to push to the Trilinos repository.
-   If updating for a failed push requires you to change Kokkos you must
-   'remove the Kokkos snapshot' from your Trilinos clone.
-   See below.
-
-------------------------------------------------------------------------
-*** WHEN SOMETHING GOES WRONG AND YOU MUST              ***
-*** REMOVE THE KOKKOS SNAPSHOT FROM YOUR TRILINOS CLONE ***
-
-1) Query the Trilinos clone commit log.
-	git log --oneline
-
-2) Note the <SHA1> of the commit to the Trillinos clone
-   immediately BEFORE the Kokkos snapshot commit.
-   Copy this <SHA1> for use in the next command.
-
-3) IF more than one outstanding commit then you can remove just the
-   Kokkos snapshot commit with 'git rebase -i'.  Edit the rebase file.
-   Remove or comment out the Kokkos snapshot commit entry.
-	git rebase -i <SHA1>
-
-4) IF the Kokkos snapshot commit is the one and only
-   outstanding commit then remove just than commit.
-	git reset --hard HEAD~1
-
-------------------------------------------------------------------------
-*** REGARDING 'snapshot.py' TOOL ***
-
-The 'snapshot.py' tool is developed and maintained by the
-Center for Computing Research (CCR)
-Software Engineering, Maintenance, and Support (SEMS) team.
-
-Contact Brent Perschbacher <bmpersc@sandia.gov> for questions>
-
-------------------------------------------------------------------------
-

lib/kokkos/Makefile.kokkos

@@ -1,6 +1,6 @@
 # Default settings common options.
 
-#LAMMPS specific settings:
+#SPARTA specific settings:
 ifndef KOKKOS_PATH
   KOKKOS_PATH=../../lib/kokkos
 endif
@@ -11,7 +11,7 @@ CXXFLAGS += $(SHFLAGS)
 endif
 
 KOKKOS_VERSION_MAJOR = 4
-KOKKOS_VERSION_MINOR = 4
+KOKKOS_VERSION_MINOR = 5
 KOKKOS_VERSION_PATCH = 1
 KOKKOS_VERSION = $(shell echo $(KOKKOS_VERSION_MAJOR)*10000+$(KOKKOS_VERSION_MINOR)*100+$(KOKKOS_VERSION_PATCH) | bc)
 
@@ -23,7 +23,7 @@ KOKKOS_DEVICES ?= "OpenMP"
 # 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,ARMv9-Grace
 # IBM:      Power8,Power9
-# AMD-GPUS: AMD_GFX906,AMD_GFX908,AMD_GFX90A,AMD_GFX940,AMD_GFX942,AMD_GFX1030,AMD_GFX1100,AMD_GFX1103
+# AMD-GPUS: AMD_GFX906,AMD_GFX908,AMD_GFX90A,AMD_GFX940,AMD_GFX942,AMD_GFX942_APU,AMD_GFX1030,AMD_GFX1100,AMD_GFX1103
 # AMD-CPUS: AMDAVX,Zen,Zen2,Zen3
 # Intel-GPUs: Intel_Gen,Intel_Gen9,Intel_Gen11,Intel_Gen12LP,Intel_DG1,Intel_XeHP,Intel_PVC
 KOKKOS_ARCH ?= ""
@@ -40,16 +40,19 @@ KOKKOS_TRIBITS ?= "no"
 KOKKOS_STANDALONE_CMAKE ?= "no"
 
 # Default settings specific options.
-# Options: force_uvm,use_ldg,rdc,enable_lambda,enable_constexpr,disable_malloc_async
+# Options: force_uvm,use_ldg,rdc,enable_lambda,enable_constexpr,enable_malloc_async
-KOKKOS_CUDA_OPTIONS ?= "disable_malloc_async"
+KOKKOS_CUDA_OPTIONS ?= ""
 
-# Options: rdc
+# Options: rdc,enable_malloc_async
 KOKKOS_HIP_OPTIONS ?= ""
 
 # Default settings specific options.
 # Options: enable_async_dispatch
 KOKKOS_HPX_OPTIONS ?= ""
 
+#Options : force_host_as_device
+KOKKOS_OPENACC_OPTIONS ?= ""
+
 # Helper functions for conversion to upper case
 uppercase_TABLE:=a,A b,B c,C d,D e,E f,F g,G h,H i,I j,J k,K l,L m,M n,N o,O p,P q,Q r,R s,S t,T u,U v,V w,W x,X y,Y z,Z
 uppercase_internal=$(if $1,$$(subst $(firstword $1),$(call uppercase_internal,$(wordlist 2,$(words $1),$1),$2)),$2)
@@ -92,7 +95,7 @@ KOKKOS_INTERNAL_CUDA_USE_UVM := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),
 KOKKOS_INTERNAL_CUDA_USE_RELOC := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),rdc)
 KOKKOS_INTERNAL_CUDA_USE_LAMBDA := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),enable_lambda)
 KOKKOS_INTERNAL_CUDA_USE_CONSTEXPR := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),enable_constexpr)
-KOKKOS_INTERNAL_CUDA_DISABLE_MALLOC_ASYNC := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),disable_malloc_async)
+KOKKOS_INTERNAL_CUDA_ENABLE_MALLOC_ASYNC := $(call kokkos_has_string,$(KOKKOS_CUDA_OPTIONS),enable_malloc_async)
 KOKKOS_INTERNAL_HPX_ENABLE_ASYNC_DISPATCH := $(call kokkos_has_string,$(KOKKOS_HPX_OPTIONS),enable_async_dispatch)
 # deprecated
 KOKKOS_INTERNAL_ENABLE_DESUL_ATOMICS := $(call kokkos_has_string,$(KOKKOS_OPTIONS),enable_desul_atomics)
@@ -103,6 +106,8 @@ KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE := $(call kokkos_has_string,$(KOKKOS_OPT
 KOKKOS_INTERNAL_ENABLE_DEPRECATION_WARNINGS := $(call kokkos_has_string,$(KOKKOS_OPTIONS),enable_deprecation_warnings)
 
 KOKKOS_INTERNAL_HIP_USE_RELOC := $(call kokkos_has_string,$(KOKKOS_HIP_OPTIONS),rdc)
+KOKKOS_INTERNAL_HIP_ENABLE_MALLOC_ASYNC := $(call kokkos_has_string,$(KOKKOS_HIP_OPTIONS),enable_malloc_async)
+KOKKOS_INTERNAL_OPENACC_FORCE_HOST_AS_DEVICE := $(call kokkos_has_string,$(KOKKOS_OPENACC_OPTIONS),force_host_as_device)
 
 # Check for Kokkos Host Execution Spaces one of which must be on.
 KOKKOS_INTERNAL_USE_OPENMP := $(call kokkos_has_string,$(subst OpenMPTarget,,$(KOKKOS_DEVICES)),OpenMP)
@@ -178,7 +183,7 @@ KOKKOS_INTERNAL_COMPILER_CRAY        := $(strip $(shell $(CXX) -craype-verbose 2
 KOKKOS_INTERNAL_COMPILER_NVCC        := $(strip $(shell echo "$(shell export OMPI_CXX=$(OMPI_CXX); export MPICH_CXX=$(MPICH_CXX); $(CXX) --version 2>&1 | grep -c nvcc)>0" | bc))
 KOKKOS_INTERNAL_COMPILER_NVHPC       := $(strip $(shell $(CXX) --version 2>&1 | grep -c "nvc++"))
 KOKKOS_INTERNAL_COMPILER_CLANG       := $(call kokkos_has_string,$(KOKKOS_CXX_VERSION),clang)
-KOKKOS_INTERNAL_COMPILER_CRAY_CLANG  := $(strip $(shell $(CXX) -craype-verbose 2>&1 | grep -c "clang++"))
+KOKKOS_INTERNAL_COMPILER_CRAY_CLANG  := $(strip $(shell $(CXX) -craype-verbose 2>&1 | grep -v "error:" | grep -c "clang++"))
 KOKKOS_INTERNAL_COMPILER_INTEL_CLANG := $(call kokkos_has_string,$(KOKKOS_CXX_VERSION),oneAPI)
 KOKKOS_INTERNAL_COMPILER_APPLE_CLANG := $(call kokkos_has_string,$(KOKKOS_CXX_VERSION),Apple clang)
 KOKKOS_INTERNAL_COMPILER_HCC         := $(call kokkos_has_string,$(KOKKOS_CXX_VERSION),HCC)
@@ -292,6 +297,8 @@ ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
   # Set OpenACC flags.
   ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 1)
     KOKKOS_INTERNAL_OPENACC_FLAG := -acc
+  else ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
+    KOKKOS_INTERNAL_OPENACC_FLAG := -fopenacc -fopenacc-fake-async-wait -fopenacc-implicit-worker=vector -Wno-openacc-and-cxx -Wno-openmp-mapping -Wno-unknown-cuda-version -Wno-pass-failed
   else
     $(error Makefile.kokkos: OpenACC is enabled but the compiler must be NVHPC (got version string $(KOKKOS_CXX_VERSION)))
   endif
@@ -411,8 +418,8 @@ ifeq ($(KOKKOS_INTERNAL_USE_ARCH_NVIDIA), 0)
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_NVIDIA), 1)
+  KOKKOS_INTERNAL_NVCC_PATH := $(shell which nvcc)
   ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
-    KOKKOS_INTERNAL_NVCC_PATH := $(shell which nvcc)
     CUDA_PATH ?= $(KOKKOS_INTERNAL_NVCC_PATH:/bin/nvcc=)
     ifeq ($(KOKKOS_INTERNAL_USE_OPENMPTARGET), 1)
       KOKKOS_INTERNAL_OPENMPTARGET_FLAG := $(KOKKOS_INTERNAL_OPENMPTARGET_FLAG) --cuda-path=$(CUDA_PATH)
@@ -457,6 +464,7 @@ ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX90A), 0)
 endif
 KOKKOS_INTERNAL_USE_ARCH_AMD_GFX940 := $(call kokkos_has_string,$(KOKKOS_ARCH),AMD_GFX940)
 KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942 := $(call kokkos_has_string,$(KOKKOS_ARCH),AMD_GFX942)
+KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942_APU := $(call kokkos_has_string,$(KOKKOS_ARCH),AMD_GFX942_APU)
 KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1030 := $(call kokkos_has_string,$(KOKKOS_ARCH),AMD_GFX1030)
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1030), 0)
   KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1030 := $(call kokkos_has_string,$(KOKKOS_ARCH),NAVI1030)
@@ -466,6 +474,15 @@ ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1100), 0)
   KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1100 := $(call kokkos_has_string,$(KOKKOS_ARCH),NAVI1100)
 endif
 KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1103 := $(call kokkos_has_string,$(KOKKOS_ARCH),AMD_GFX1103)
+KOKKOS_INTERNAL_USE_ARCH_AMD := $(shell expr $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX906)  \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX908)  \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX90A)  \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX940)  \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942) \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942_APU) \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1030) \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1100) \
+                                              + $(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1103))
 
 # Any AVX?
 KOKKOS_INTERNAL_USE_ARCH_AVX        := $(shell expr $(KOKKOS_INTERNAL_USE_ARCH_SNB) + $(KOKKOS_INTERNAL_USE_ARCH_AMDAVX))
@@ -561,6 +578,9 @@ endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_OPENACC")
+  ifeq ($(KOKKOS_INTERNAL_OPENACC_FORCE_HOST_AS_DEVICE), 1)
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_OPENACC_FORCE_HOST_AS_DEVICE")
+  endif
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENMP), 1)
@@ -733,7 +753,7 @@ ifeq ($(KOKKOS_INTERNAL_USE_CUDA), 1)
     endif
   endif
 
-  ifeq ($(KOKKOS_INTERNAL_CUDA_DISABLE_MALLOC_ASYNC), 0)
+  ifeq ($(KOKKOS_INTERNAL_CUDA_ENABLE_MALLOC_ASYNC), 1)
     tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_IMPL_CUDA_MALLOC_ASYNC")
   else
     tmp := $(call kokkos_append_header,"/* $H""undef KOKKOS_ENABLE_IMPL_CUDA_MALLOC_ASYNC */")
@@ -1024,86 +1044,122 @@ ifeq ($(KOKKOS_INTERNAL_USE_OPENMPTARGET), 1)
   endif
 endif
 
+ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
+  ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
+      KOKKOS_INTERNAL_CUDA_ARCH_FLAG=--offload-arch
+  endif
+endif
+
 # Do not add this flag if its the cray compiler or the nvhpc compiler.
 ifeq ($(KOKKOS_INTERNAL_COMPILER_CRAY_CLANG), 0)
-  ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
+  # Lets start with adding architecture defines
-    # Lets start with adding architecture defines
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER30), 1)
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER30), 1)
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER30")
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER30")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_30
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER32), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER32), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER32")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER32")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_32
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER35), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER35), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER35")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER35")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_35
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER37), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_KEPLER37), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER37")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_KEPLER37")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_37
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL50), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL50), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL50")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL50")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_50
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL52), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL52), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL52")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL52")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_52
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL53), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_MAXWELL53), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL53")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_MAXWELL53")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_53
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_PASCAL60), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_PASCAL60), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL60")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL60")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_60
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_PASCAL61), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_PASCAL61), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL61")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_PASCAL61")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_61
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_VOLTA70), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_VOLTA70), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA70")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA70")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_70
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_VOLTA72), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_VOLTA72), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA72")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_VOLTA72")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_72
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_TURING75), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_TURING75")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_TURING75), 1)
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_TURING75")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_75
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMPERE80), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMPERE80), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE80")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE80")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_80
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMPERE86), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMPERE86), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE86")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMPERE86")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_86
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_ADA89), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_ADA89")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_ADA89), 1)
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_ADA89")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_89
     endif
-    ifeq ($(KOKKOS_INTERNAL_USE_ARCH_HOPPER90), 1)
+  endif
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_HOPPER")
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_HOPPER90), 1)
-      tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_HOPPER90")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_HOPPER")
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_HOPPER90")
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 0)
       KOKKOS_INTERNAL_CUDA_ARCH_FLAG := $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)=sm_90
     endif
   endif
@@ -1119,6 +1175,9 @@ ifneq ($(KOKKOS_INTERNAL_USE_ARCH_NVIDIA), 0)
     ifeq ($(KOKKOS_INTERNAL_USE_OPENMPTARGET), 1)
       KOKKOS_LDFLAGS += $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)
     endif
+    ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
+      KOKKOS_LDFLAGS += $(KOKKOS_INTERNAL_CUDA_ARCH_FLAG)
+    endif
   endif
 endif
 
@@ -1126,43 +1185,48 @@ endif
 # Figure out the architecture flag for ROCm.
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX906), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX906")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx906\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx906
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx906
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX908), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX908")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx908\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx908
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx908
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX90A), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX90A")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx90A\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx90a
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx90a
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX940), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX940")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx940\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx940
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx940
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX942")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx942\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx942
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx942
+endif
+ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX942_APU), 1)
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX942_APU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx942\"")
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx942
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1030), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX1030")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx1030\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx1030
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx1030
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1100), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX1100")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx1100\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx1100
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx1100
 endif
 ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD_GFX1103), 1)
   tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GFX1103")
-  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU")
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ARCH_AMD_GPU \"gfx1103\"")
-  KOKKOS_INTERNAL_HIP_ARCH_FLAG := --offload-arch=gfx1103
+  KOKKOS_INTERNAL_AMD_ARCH_FLAG := --offload-arch=gfx1103
 endif
 
 
@@ -1171,8 +1235,8 @@ ifeq ($(KOKKOS_INTERNAL_USE_HIP), 1)
   KOKKOS_SRC += $(KOKKOS_PATH)/tpls/desul/src/Lock_Array_HIP.cpp
   KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/core/src/HIP/*.hpp)
 
-  KOKKOS_CXXFLAGS+=$(KOKKOS_INTERNAL_HIP_ARCH_FLAG)
+  KOKKOS_CXXFLAGS+=$(KOKKOS_INTERNAL_AMD_ARCH_FLAG)
-  KOKKOS_LDFLAGS+=$(KOKKOS_INTERNAL_HIP_ARCH_FLAG)
+  KOKKOS_LDFLAGS+=$(KOKKOS_INTERNAL_AMD_ARCH_FLAG)
 
   ifeq ($(KOKKOS_INTERNAL_HIP_USE_RELOC), 1)
     tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_HIP_RELOCATABLE_DEVICE_CODE")
@@ -1182,6 +1246,21 @@ ifeq ($(KOKKOS_INTERNAL_USE_HIP), 1)
     KOKKOS_CXXFLAGS+=-fno-gpu-rdc
     KOKKOS_LDFLAGS+=-fno-gpu-rdc
   endif
+
+  ifeq ($(KOKKOS_INTERNAL_HIP_ENABLE_MALLOC_ASYNC), 1)
+    tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_IMPL_HIP_MALLOC_ASYNC")
+  else
+    tmp := $(call kokkos_append_header,"/* $H""undef KOKKOS_ENABLE_IMPL_HIP_MALLOC_ASYNC */")
+  endif
+endif
+
+ifneq ($(KOKKOS_INTERNAL_USE_ARCH_AMD), 0)
+  ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
+    ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
+      KOKKOS_CXXFLAGS += $(KOKKOS_INTERNAL_AMD_ARCH_FLAG)
+      KOKKOS_LDFLAGS += $(KOKKOS_INTERNAL_AMD_ARCH_FLAG)
+    endif
+  endif
 endif
 
 # Figure out Intel architecture flags.
@@ -1235,6 +1314,8 @@ ifeq ($(KOKKOS_INTERNAL_USE_SYCL), 1)
   KOKKOS_CXXFLAGS+=$(KOKKOS_INTERNAL_INTEL_ARCH_FLAG)
   KOKKOS_LDFLAGS+=-fsycl
   KOKKOS_LDFLAGS+=$(KOKKOS_INTERNAL_INTEL_ARCH_FLAG)
+
+  tmp := $(call kokkos_append_header,"$H""define KOKKOS_ENABLE_SYCL_RELOCATABLE_DEVICE_CODE")
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENMPTARGET), 1)
@@ -1322,6 +1403,8 @@ ifneq ($(KOKKOS_INTERNAL_NEW_CONFIG), 0)
 endif
 
 KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/core/src/*.hpp)
+KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/core/src/View/*.hpp)
+KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/core/src/View/MDSpan/*.hpp)
 KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/core/src/impl/*.hpp)
 KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/containers/src/*.hpp)
 KOKKOS_HEADERS += $(wildcard $(KOKKOS_PATH)/containers/src/impl/*.hpp)
@@ -1374,6 +1457,48 @@ ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
   KOKKOS_CXXFLAGS += $(KOKKOS_INTERNAL_OPENACC_FLAG)
   KOKKOS_LDFLAGS += $(KOKKOS_INTERNAL_OPENACC_FLAG)
   KOKKOS_LIBS += $(KOKKOS_INTERNAL_OPENACC_LIB)
+  ifeq ($(KOKKOS_INTERNAL_USE_ARCH_NVIDIA), 1)
+    ifneq ($(CUDA_PATH),)
+      ifeq ($(call kokkos_path_exists,$(CUDA_PATH)/lib), 1)
+        CUDA_PATH := $(CUDA_PATH:/compilers=/cuda)
+      endif
+    endif
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
+      ifneq ($(CUDA_PATH),)
+        KOKKOS_LDFLAGS += -L$(CUDA_PATH)/lib64
+      endif
+      KOKKOS_LIBS += -lcudart
+    endif
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 1)
+      KOKKOS_LIBS += -cuda
+    endif
+    ifeq ($(KOKKOS_INTERNAL_OPENACC_FORCE_HOST_AS_DEVICE), 1)
+      $(error If a GPU architecture is specified,  KOKKOS_OPENACC_OPTIONS = force_host_as_device cannot be used. Disable the force_host_as_device option)
+    endif
+  else ifeq ($(KOKKOS_INTERNAL_USE_ARCH_AMD), 1)
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_CLANG), 1)
+      ifneq ($(ROCM_PATH),)
+        KOKKOS_CPPFLAGS += -I$(ROCM_PATH)/include
+        KOKKOS_LDFLAGS += -L$(ROCM_PATH)/lib
+      endif
+      KOKKOS_LIBS += -lamdhip64
+    endif
+    ifeq ($(KOKKOS_INTERNAL_OPENACC_FORCE_HOST_AS_DEVICE), 1)
+      $(error If a GPU architecture is specified,  KOKKOS_OPENACC_OPTIONS = force_host_as_device cannot be used. Disable the force_host_as_device option)
+    endif
+  else ifeq ($(KOKKOS_INTERNAL_OPENACC_FORCE_HOST_AS_DEVICE), 1)
+    # Compile for kernel execution on the host. In that case, 
+    # memory is shared between the OpenACC space and the host space.
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 1)
+      KOKKOS_CXXFLAGS += -acc=multicore
+    endif
+  else
+    # Automatic fallback mode; try to offload any available GPU, and fall back
+    # to the host CPU if no available GPU is found.
+    ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 1)
+      KOKKOS_CXXFLAGS += -acc=gpu,multicore
+    endif
+  endif
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENMP), 1)
@@ -1484,7 +1609,11 @@ else
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)
-  tmp := $(call desul_append_header,"$H""define DESUL_ATOMICS_ENABLE_OPENACC")
+  ifeq ($(KOKKOS_INTERNAL_COMPILER_NVHPC), 1)
+    tmp := $(call desul_append_header,"$H""define DESUL_ATOMICS_ENABLE_OPENACC")
+  else
+    tmp := $(call desul_append_header,"/* $H""undef DESUL_ATOMICS_ENABLE_OPENACC */")
+  endif
 else
   tmp := $(call desul_append_header,"/* $H""undef DESUL_ATOMICS_ENABLE_OPENACC */")
 endif
@@ -1512,6 +1641,12 @@ $(DESUL_CONFIG_HEADER):
 
 KOKKOS_CPP_DEPENDS := $(DESUL_CONFIG_HEADER) KokkosCore_config.h $(KOKKOS_HEADERS)
 
+# Tasking is deprecated
+ifeq ($(KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE), 1)
+  TMP_KOKKOS_SRC := $(KOKKOS_SRC)
+  KOKKOS_SRC = $(patsubst %Task.cpp,, $(TMP_KOKKOS_SRC))
+endif
+
 KOKKOS_OBJ = $(KOKKOS_SRC:.cpp=.o)
 KOKKOS_OBJ_LINK = $(notdir $(KOKKOS_OBJ))
 

lib/kokkos/Makefile.targets

@@ -16,8 +16,6 @@ Kokkos_HostSpace.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_Ho
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/impl/Kokkos_HostSpace.cpp
 Kokkos_hwloc.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_hwloc.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/impl/Kokkos_hwloc.cpp
-Kokkos_TaskQueue.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_TaskQueue.cpp
-	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/impl/Kokkos_TaskQueue.cpp
 Kokkos_HostThreadTeam.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_HostThreadTeam.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/impl/Kokkos_HostThreadTeam.cpp
 Kokkos_HostBarrier.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_HostBarrier.cpp
@@ -38,17 +36,21 @@ Kokkos_Abort.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/impl/Kokkos_Abort.
 ifeq ($(KOKKOS_INTERNAL_USE_SERIAL), 1)
 Kokkos_Serial.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Serial/Kokkos_Serial.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Serial/Kokkos_Serial.cpp
+ifneq ($(KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE), 1)
 Kokkos_Serial_Task.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Serial/Kokkos_Serial_Task.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Serial/Kokkos_Serial_Task.cpp
 endif
+endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_CUDA), 1)
 Kokkos_Cuda_Instance.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Cuda/Kokkos_Cuda_Instance.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Cuda/Kokkos_Cuda_Instance.cpp
 Kokkos_CudaSpace.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Cuda/Kokkos_CudaSpace.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Cuda/Kokkos_CudaSpace.cpp
+ifneq ($(KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE), 1)
 Kokkos_Cuda_Task.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/Cuda/Kokkos_Cuda_Task.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/Cuda/Kokkos_Cuda_Task.cpp
+endif
 Lock_Array_CUDA.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/tpls/desul/src/Lock_Array_CUDA.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/tpls/desul/src/Lock_Array_CUDA.cpp
 endif
@@ -73,6 +75,8 @@ Kokkos_HIP_Space.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP_Space.cpp
 Kokkos_HIP_Instance.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP_Instance.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP_Instance.cpp
+Kokkos_HIP_ZeroMemset.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP_ZeroMemset.cpp
+	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/HIP/Kokkos_HIP_ZeroMemset.cpp
 Lock_Array_HIP.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/tpls/desul/src/Lock_Array_HIP.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/tpls/desul/src/Lock_Array_HIP.cpp
 endif
@@ -89,26 +93,26 @@ Kokkos_OpenMP.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_Ope
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_OpenMP.cpp
 Kokkos_OpenMP_Instance.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_OpenMP_Instance.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_OpenMP_Instance.cpp
+ifneq ($(KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE), 1)
 Kokkos_OpenMP_Task.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_OpenMP_Task.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMP/Kokkos_OpenMP_Task.cpp
 endif
+endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_HPX), 1)
 Kokkos_HPX.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/HPX/Kokkos_HPX.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/HPX/Kokkos_HPX.cpp
+ifneq ($(KOKKOS_INTERNAL_DISABLE_DEPRECATED_CODE), 1)
 Kokkos_HPX_Task.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/HPX/Kokkos_HPX_Task.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/HPX/Kokkos_HPX_Task.cpp
 endif
+endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENMPTARGET), 1)
-Kokkos_OpenMPTarget_Exec.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Exec.cpp
-	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Exec.cpp
 Kokkos_OpenMPTarget_Instance.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Instance.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Instance.cpp
 Kokkos_OpenMPTargetSpace.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTargetSpace.cpp
 	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTargetSpace.cpp
-Kokkos_OpenMPTarget_Task.o: $(KOKKOS_CPP_DEPENDS) $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Task.cpp
-	$(CXX) $(KOKKOS_CPPFLAGS) $(KOKKOS_CXXFLAGS) $(CXXFLAGS) -c $(KOKKOS_PATH)/core/src/OpenMPTarget/Kokkos_OpenMPTarget_Task.cpp
 endif
 
 ifeq ($(KOKKOS_INTERNAL_USE_OPENACC), 1)

lib/kokkos/README.md

@@ -30,12 +30,12 @@ To start learning about 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).
+The current release is [4.5.01](https://github.com/kokkos/kokkos/releases/tag/4.5.01).
 
 ```bash
-curl -OJ -L https://github.com/kokkos/kokkos/archive/refs/tags/4.3.01.tar.gz
+curl -OJ -L https://github.com/kokkos/kokkos/releases/download/4.5.01/kokkos-4.5.01.tar.gz
 # Or with wget
-wget https://github.com/kokkos/kokkos/archive/refs/tags/4.3.01.tar.gz
+wget https://github.com/kokkos/kokkos/releases/download/4.5.01/kokkos-4.5.01.tar.gz
 ```
 
 To clone the latest development version of Kokkos from GitHub:

lib/kokkos/algorithms/CMakeLists.txt

@@ -1,7 +1,7 @@
-IF (NOT Kokkos_INSTALL_TESTING)
+if(NOT Kokkos_INSTALL_TESTING)
-  ADD_SUBDIRECTORY(src)
+  add_subdirectory(src)
-ENDIF()
+endif()
 # FIXME_OPENACC: temporarily disabled due to unimplemented features
-IF(NOT ((KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC) OR KOKKOS_ENABLE_OPENACC))
+if(NOT ((KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL NVHPC) OR KOKKOS_ENABLE_OPENACC))
-  KOKKOS_ADD_TEST_DIRECTORIES(unit_tests)
+  kokkos_add_test_directories(unit_tests)
-ENDIF()
+endif()

lib/kokkos/algorithms/src/CMakeLists.txt

@@ -1,34 +1,29 @@
 #I have to leave these here for tribits
-KOKKOS_INCLUDE_DIRECTORIES(${CMAKE_CURRENT_BINARY_DIR})
+kokkos_include_directories(${CMAKE_CURRENT_BINARY_DIR})
-KOKKOS_INCLUDE_DIRECTORIES(${CMAKE_CURRENT_SOURCE_DIR})
+kokkos_include_directories(${CMAKE_CURRENT_SOURCE_DIR})
 
 #-----------------------------------------------------------------------------
 
-FILE(GLOB ALGO_HEADERS *.hpp)
+file(GLOB ALGO_HEADERS *.hpp)
-FILE(GLOB ALGO_SOURCES *.cpp)
+file(GLOB ALGO_SOURCES *.cpp)
-APPEND_GLOB(ALGO_HEADERS ${CMAKE_CURRENT_SOURCE_DIR}/std_algorithms/*.hpp)
+append_glob(ALGO_HEADERS ${CMAKE_CURRENT_SOURCE_DIR}/std_algorithms/*.hpp)
-APPEND_GLOB(ALGO_HEADERS ${CMAKE_CURRENT_SOURCE_DIR}/std_algorithms/impl/*.hpp)
+append_glob(ALGO_HEADERS ${CMAKE_CURRENT_SOURCE_DIR}/std_algorithms/impl/*.hpp)
 
-INSTALL (
+install(
   DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}/"
   DESTINATION ${KOKKOS_HEADER_DIR}
-  FILES_MATCHING PATTERN "*.hpp"
+  FILES_MATCHING
+  PATTERN "*.hpp"
 )
 
 #-----------------------------------------------------------------------------
 
 # We have to pass the sources in here for Tribits
 # These will get ignored for standalone CMake and a true interface library made
-KOKKOS_ADD_INTERFACE_LIBRARY(
+kokkos_add_interface_library(kokkosalgorithms NOINSTALLHEADERS ${ALGO_HEADERS} SOURCES ${ALGO_SOURCES})
-  kokkosalgorithms
+kokkos_lib_include_directories(
-  NOINSTALLHEADERS ${ALGO_HEADERS}
+  kokkosalgorithms ${KOKKOS_TOP_BUILD_DIR} ${CMAKE_CURRENT_BINARY_DIR} ${CMAKE_CURRENT_SOURCE_DIR}
-  SOURCES ${ALGO_SOURCES}
-)
-KOKKOS_LIB_INCLUDE_DIRECTORIES(kokkosalgorithms
-  ${KOKKOS_TOP_BUILD_DIR}
-  ${CMAKE_CURRENT_BINARY_DIR}
-  ${CMAKE_CURRENT_SOURCE_DIR}
 )
 
-KOKKOS_LINK_TPL(kokkoscontainers PUBLIC ROCTHRUST)
+kokkos_link_tpl(kokkoscontainers PUBLIC ROCTHRUST)
-KOKKOS_LINK_TPL(kokkoscore PUBLIC ONEDPL)
+kokkos_link_tpl(kokkoscore PUBLIC ONEDPL)

lib/kokkos/algorithms/src/Kokkos_Random.hpp

@@ -615,7 +615,7 @@ template <class DeviceType>
 struct Random_UniqueIndex {
   using locks_view_type = View<int**, DeviceType>;
   KOKKOS_FUNCTION
-  static int get_state_idx(const locks_view_type) {
+  static int get_state_idx(const locks_view_type&) {
     KOKKOS_IF_ON_HOST(
         (return DeviceType::execution_space::impl_hardware_thread_id();))
 
@@ -665,17 +665,16 @@ struct Random_UniqueIndex<
 
 #ifdef KOKKOS_ENABLE_SYCL
 template <class MemorySpace>
-struct Random_UniqueIndex<
+struct Random_UniqueIndex<Kokkos::Device<Kokkos::SYCL, MemorySpace>> {
-    Kokkos::Device<Kokkos::Experimental::SYCL, MemorySpace>> {
   using locks_view_type =
-      View<int**, Kokkos::Device<Kokkos::Experimental::SYCL, MemorySpace>>;
+      View<int**, Kokkos::Device<Kokkos::SYCL, MemorySpace>>;
   KOKKOS_FUNCTION
   static int get_state_idx(const locks_view_type& locks_) {
     auto item = sycl::ext::oneapi::experimental::this_nd_item<3>();
     std::size_t threadIdx[3] = {item.get_local_id(2), item.get_local_id(1),
                                 item.get_local_id(0)};
     std::size_t blockIdx[3]  = {item.get_group(2), item.get_group(1),
-                               item.get_group(0)};
+                                item.get_group(0)};
     std::size_t blockDim[3] = {item.get_local_range(2), item.get_local_range(1),
                                item.get_local_range(0)};
     std::size_t gridDim[3]  = {
@@ -1121,7 +1120,7 @@ class Random_XorShift1024_Pool {
   using execution_space = typename device_type::execution_space;
   using locks_type      = View<int**, device_type>;
   using int_view_type   = View<int**, device_type>;
-  using state_data_type = View<uint64_t * [16], device_type>;
+  using state_data_type = View<uint64_t* [16], device_type>;
 
   locks_type locks_      = {};
   state_data_type state_ = {};

lib/kokkos/algorithms/src/sorting/Kokkos_BinOpsPublicAPI.hpp

@@ -35,11 +35,11 @@ struct BinOp1D {
 #endif
 
   // Construct BinOp with number of bins, minimum value and maximum value
-  BinOp1D(int max_bins__, typename KeyViewType::const_value_type min,
+  BinOp1D(int max_bins, typename KeyViewType::const_value_type min,
           typename KeyViewType::const_value_type max)
-      : max_bins_(max_bins__ + 1),
+      : max_bins_(max_bins + 1),
         // Cast to double to avoid possible overflow when using integer
-        mul_(static_cast<double>(max_bins__) /
+        mul_(static_cast<double>(max_bins) /
              (static_cast<double>(max) - static_cast<double>(min))),
         min_(static_cast<double>(min)) {
     // For integral types the number of bins may be larger than the range
@@ -47,7 +47,7 @@ struct BinOp1D {
     // and then don't need to sort bins.
     if (std::is_integral<typename KeyViewType::const_value_type>::value &&
         (static_cast<double>(max) - static_cast<double>(min)) <=
-            static_cast<double>(max_bins__)) {
+            static_cast<double>(max_bins)) {
       mul_ = 1.;
     }
   }
@@ -82,16 +82,16 @@ struct BinOp3D {
   BinOp3D() = delete;
 #endif
 
-  BinOp3D(int max_bins__[], typename KeyViewType::const_value_type min[],
+  BinOp3D(int max_bins[], typename KeyViewType::const_value_type min[],
           typename KeyViewType::const_value_type max[]) {
-    max_bins_[0] = max_bins__[0];
+    max_bins_[0] = max_bins[0];
-    max_bins_[1] = max_bins__[1];
+    max_bins_[1] = max_bins[1];
-    max_bins_[2] = max_bins__[2];
+    max_bins_[2] = max_bins[2];
-    mul_[0]      = static_cast<double>(max_bins__[0]) /
+    mul_[0]      = static_cast<double>(max_bins[0]) /
               (static_cast<double>(max[0]) - static_cast<double>(min[0]));
-    mul_[1] = static_cast<double>(max_bins__[1]) /
+    mul_[1] = static_cast<double>(max_bins[1]) /
               (static_cast<double>(max[1]) - static_cast<double>(min[1]));
-    mul_[2] = static_cast<double>(max_bins__[2]) /
+    mul_[2] = static_cast<double>(max_bins[2]) /
               (static_cast<double>(max[2]) - static_cast<double>(min[2]));
     min_[0] = static_cast<double>(min[0]);
     min_[1] = static_cast<double>(min[1]);

lib/kokkos/algorithms/src/sorting/Kokkos_BinSortPublicAPI.hpp

@@ -388,7 +388,8 @@ class BinSort {
     // reasonable experimentally.
     if (use_std_sort && bin_size > 10) {
       KOKKOS_IF_ON_HOST(
-          (std::sort(&sort_order(lower_bound), &sort_order(upper_bound),
+          (std::sort(sort_order.data() + lower_bound,
+                     sort_order.data() + upper_bound,
                      [this](int p, int q) { return bin_op(keys_rnd, p, q); });))
     } else {
       for (int k = lower_bound + 1; k < upper_bound; ++k) {

lib/kokkos/algorithms/src/sorting/Kokkos_SortPublicAPI.hpp

@@ -53,9 +53,13 @@ void sort(const ExecutionSpace& exec,
 
   if constexpr (Impl::better_off_calling_std_sort_v<ExecutionSpace>) {
     exec.fence("Kokkos::sort without comparator use std::sort");
-    auto first = ::Kokkos::Experimental::begin(view);
+    if (view.span_is_contiguous()) {
-    auto last  = ::Kokkos::Experimental::end(view);
+      std::sort(view.data(), view.data() + view.size());
-    std::sort(first, last);
+    } else {
+      auto first = ::Kokkos::Experimental::begin(view);
+      auto last  = ::Kokkos::Experimental::end(view);
+      std::sort(first, last);
+    }
   } else {
     Impl::sort_device_view_without_comparator(exec, view);
   }
@@ -107,9 +111,13 @@ void sort(const ExecutionSpace& exec,
 
   if constexpr (Impl::better_off_calling_std_sort_v<ExecutionSpace>) {
     exec.fence("Kokkos::sort with comparator use std::sort");
-    auto first = ::Kokkos::Experimental::begin(view);
+    if (view.span_is_contiguous()) {
-    auto last  = ::Kokkos::Experimental::end(view);
+      std::sort(view.data(), view.data() + view.size(), comparator);
-    std::sort(first, last, comparator);
+    } else {
+      auto first = ::Kokkos::Experimental::begin(view);
+      auto last  = ::Kokkos::Experimental::end(view);
+      std::sort(first, last, comparator);
+    }
   } else {
     Impl::sort_device_view_with_comparator(exec, view, comparator);
   }

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

@@ -30,6 +30,7 @@
 
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored "-Wshadow"
+#pragma GCC diagnostic ignored "-Wsuggest-override"
 
 #if defined(KOKKOS_COMPILER_CLANG)
 // Some versions of Clang fail to compile Thrust, failing with errors like
@@ -76,13 +77,10 @@ namespace Kokkos::Impl {
 
 template <typename T>
 constexpr inline bool is_admissible_to_kokkos_sort_by_key =
-    ::Kokkos::is_view<T>::value&& T::rank() == 1 &&
+    ::Kokkos::is_view<T>::value && T::rank() == 1 &&
-    (std::is_same<typename T::traits::array_layout,
+    (std::is_same_v<typename T::traits::array_layout, Kokkos::LayoutLeft> ||
-                  Kokkos::LayoutLeft>::value ||
+     std::is_same_v<typename T::traits::array_layout, Kokkos::LayoutRight> ||
-     std::is_same<typename T::traits::array_layout,
+     std::is_same_v<typename T::traits::array_layout, Kokkos::LayoutStride>);
-                  Kokkos::LayoutRight>::value ||
-     std::is_same<typename T::traits::array_layout,
-                  Kokkos::LayoutStride>::value);
 
 template <class ViewType>
 KOKKOS_INLINE_FUNCTION constexpr void
@@ -144,7 +142,7 @@ void sort_by_key_rocthrust(
 
 #if defined(KOKKOS_ENABLE_ONEDPL)
 template <class Layout>
-inline constexpr bool sort_on_device_v<Kokkos::Experimental::SYCL, Layout> =
+inline constexpr bool sort_on_device_v<Kokkos::SYCL, Layout> =
     std::is_same_v<Layout, Kokkos::LayoutLeft> ||
     std::is_same_v<Layout, Kokkos::LayoutRight>;
 
@@ -152,7 +150,7 @@ inline constexpr bool sort_on_device_v<Kokkos::Experimental::SYCL, Layout> =
 template <class KeysDataType, class... KeysProperties, class ValuesDataType,
           class... ValuesProperties, class... MaybeComparator>
 void sort_by_key_onedpl(
-    const Kokkos::Experimental::SYCL& exec,
+    const Kokkos::SYCL& exec,
     const Kokkos::View<KeysDataType, KeysProperties...>& keys,
     const Kokkos::View<ValuesDataType, ValuesProperties...>& values,
     MaybeComparator&&... maybeComparator) {
@@ -176,7 +174,7 @@ template <typename ExecutionSpace, typename PermutationView, typename ViewType>
 void applyPermutation(const ExecutionSpace& space,
                       const PermutationView& permutation,
                       const ViewType& view) {
-  static_assert(std::is_integral<typename PermutationView::value_type>::value);
+  static_assert(std::is_integral_v<typename PermutationView::value_type>);
 
   auto view_copy = Kokkos::create_mirror(
       Kokkos::view_alloc(space, typename ExecutionSpace::memory_space{},
@@ -335,7 +333,7 @@ void sort_by_key_device_view_without_comparator(
 template <class KeysDataType, class... KeysProperties, class ValuesDataType,
           class... ValuesProperties>
 void sort_by_key_device_view_without_comparator(
-    const Kokkos::Experimental::SYCL& exec,
+    const Kokkos::SYCL& exec,
     const Kokkos::View<KeysDataType, KeysProperties...>& keys,
     const Kokkos::View<ValuesDataType, ValuesProperties...>& values) {
 #ifdef KOKKOS_ONEDPL_HAS_SORT_BY_KEY
@@ -392,7 +390,7 @@ void sort_by_key_device_view_with_comparator(
 template <class ComparatorType, class KeysDataType, class... KeysProperties,
           class ValuesDataType, class... ValuesProperties>
 void sort_by_key_device_view_with_comparator(
-    const Kokkos::Experimental::SYCL& exec,
+    const Kokkos::SYCL& exec,
     const Kokkos::View<KeysDataType, KeysProperties...>& keys,
     const Kokkos::View<ValuesDataType, ValuesProperties...>& values,
     const ComparatorType& comparator) {

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

@@ -34,6 +34,7 @@
 
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored "-Wshadow"
+#pragma GCC diagnostic ignored "-Wsuggest-override"
 
 #if defined(KOKKOS_COMPILER_CLANG)
 // Some versions of Clang fail to compile Thrust, failing with errors like
@@ -146,7 +147,7 @@ void sort_via_binsort(const ExecutionSpace& exec,
   bool sort_in_bins = true;
   // TODO: figure out better max_bins then this ...
   int64_t max_bins = view.extent(0) / 2;
-  if (std::is_integral<typename ViewType::non_const_value_type>::value) {
+  if (std::is_integral_v<typename ViewType::non_const_value_type>) {
     // Cast to double to avoid possible overflow when using integer
     auto const max_val = static_cast<double>(result.max_val);
     auto const min_val = static_cast<double>(result.min_val);
@@ -157,7 +158,7 @@ void sort_via_binsort(const ExecutionSpace& exec,
       sort_in_bins = false;
     }
   }
-  if (std::is_floating_point<typename ViewType::non_const_value_type>::value) {
+  if (std::is_floating_point_v<typename ViewType::non_const_value_type>) {
     KOKKOS_ASSERT(std::isfinite(static_cast<double>(result.max_val) -
                                 static_cast<double>(result.min_val)));
   }
@@ -211,11 +212,11 @@ void sort_rocthrust(const HIP& space,
 
 #if defined(KOKKOS_ENABLE_ONEDPL)
 template <class DataType, class... Properties, class... MaybeComparator>
-void sort_onedpl(const Kokkos::Experimental::SYCL& space,
+void sort_onedpl(const Kokkos::SYCL& space,
                  const Kokkos::View<DataType, Properties...>& view,
                  MaybeComparator&&... maybeComparator) {
   using ViewType = Kokkos::View<DataType, Properties...>;
-  static_assert(SpaceAccessibility<Kokkos::Experimental::SYCL,
+  static_assert(SpaceAccessibility<Kokkos::SYCL,
                                    typename ViewType::memory_space>::accessible,
                 "SYCL execution space is not able to access the memory space "
                 "of the View argument!");
@@ -268,19 +269,29 @@ void copy_to_host_run_stdsort_copy_back(
     KE::copy(exec, view, view_dc);
 
     // run sort on the mirror of view_dc
-    auto mv_h  = create_mirror_view_and_copy(Kokkos::HostSpace(), view_dc);
+    auto mv_h = create_mirror_view_and_copy(Kokkos::HostSpace(), view_dc);
-    auto first = KE::begin(mv_h);
+    if (view.span_is_contiguous()) {
-    auto last  = KE::end(mv_h);
+      std::sort(mv_h.data(), mv_h.data() + mv_h.size(),
-    std::sort(first, last, std::forward<MaybeComparator>(maybeComparator)...);
+                std::forward<MaybeComparator>(maybeComparator)...);
+    } else {
+      auto first = KE::begin(mv_h);
+      auto last  = KE::end(mv_h);
+      std::sort(first, last, std::forward<MaybeComparator>(maybeComparator)...);
+    }
     Kokkos::deep_copy(exec, view_dc, mv_h);
 
     // copy back to argument view
     KE::copy(exec, KE::cbegin(view_dc), KE::cend(view_dc), KE::begin(view));
   } else {
     auto view_h = create_mirror_view_and_copy(Kokkos::HostSpace(), view);
-    auto first  = KE::begin(view_h);
+    if (view.span_is_contiguous()) {
-    auto last   = KE::end(view_h);
+      std::sort(view_h.data(), view_h.data() + view_h.size(),
-    std::sort(first, last, std::forward<MaybeComparator>(maybeComparator)...);
+                std::forward<MaybeComparator>(maybeComparator)...);
+    } else {
+      auto first = KE::begin(view_h);
+      auto last  = KE::end(view_h);
+      std::sort(first, last, std::forward<MaybeComparator>(maybeComparator)...);
+    }
     Kokkos::deep_copy(exec, view, view_h);
   }
 }
@@ -310,7 +321,7 @@ void sort_device_view_without_comparator(
 #if defined(KOKKOS_ENABLE_ONEDPL)
 template <class DataType, class... Properties>
 void sort_device_view_without_comparator(
-    const Kokkos::Experimental::SYCL& exec,
+    const Kokkos::SYCL& exec,
     const Kokkos::View<DataType, Properties...>& view) {
   using ViewType = Kokkos::View<DataType, Properties...>;
   static_assert(
@@ -365,8 +376,7 @@ void sort_device_view_with_comparator(
 #if defined(KOKKOS_ENABLE_ONEDPL)
 template <class ComparatorType, class DataType, class... Properties>
 void sort_device_view_with_comparator(
-    const Kokkos::Experimental::SYCL& exec,
+    const Kokkos::SYCL& exec, const Kokkos::View<DataType, Properties...>& view,
-    const Kokkos::View<DataType, Properties...>& view,
     const ComparatorType& comparator) {
   using ViewType = Kokkos::View<DataType, Properties...>;
   static_assert(
@@ -397,12 +407,12 @@ sort_device_view_with_comparator(
   // and then copies data back. Potentially, this can later be changed
   // with a better solution like our own quicksort on device or similar.
 
-  using ViewType = Kokkos::View<DataType, Properties...>;
-  using MemSpace = typename ViewType::memory_space;
 // Note with HIP unified memory this code path is still the right thing to do
 // if we end up here when RocThrust is not enabled.
 // The create_mirror_view_and_copy will do the right thing (no copy).
-#ifndef KOKKOS_ENABLE_IMPL_HIP_UNIFIED_MEMORY
+#ifndef KOKKOS_IMPL_HIP_UNIFIED_MEMORY
+  using ViewType = Kokkos::View<DataType, Properties...>;
+  using MemSpace = typename ViewType::memory_space;
   static_assert(!SpaceAccessibility<HostSpace, MemSpace>::accessible,
                 "Impl::sort_device_view_with_comparator: should not be called "
                 "on a view that is already accessible on the host");

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

@@ -91,7 +91,7 @@ template <typename ExecutionSpace, typename IteratorType, typename ValueType,
                            int> = 0>
 ValueType reduce(const ExecutionSpace& ex, IteratorType first,
                  IteratorType last, ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_default_functors_exespace_impl(
@@ -105,7 +105,7 @@ template <typename ExecutionSpace, typename IteratorType, typename ValueType,
 ValueType reduce(const std::string& label, const ExecutionSpace& ex,
                  IteratorType first, IteratorType last,
                  ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_default_functors_exespace_impl(label, ex, first, last,
@@ -119,7 +119,7 @@ template <typename ExecutionSpace, typename DataType, typename... Properties,
 ValueType reduce(const ExecutionSpace& ex,
                  const ::Kokkos::View<DataType, Properties...>& view,
                  ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;
@@ -137,7 +137,7 @@ template <typename ExecutionSpace, typename DataType, typename... Properties,
 ValueType reduce(const std::string& label, const ExecutionSpace& ex,
                  const ::Kokkos::View<DataType, Properties...>& view,
                  ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;
@@ -157,7 +157,7 @@ template <typename ExecutionSpace, typename IteratorType, typename ValueType,
 ValueType reduce(const ExecutionSpace& ex, IteratorType first,
                  IteratorType last, ValueType init_reduction_value,
                  BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_custom_functors_exespace_impl(
@@ -172,7 +172,7 @@ template <typename ExecutionSpace, typename IteratorType, typename ValueType,
 ValueType reduce(const std::string& label, const ExecutionSpace& ex,
                  IteratorType first, IteratorType last,
                  ValueType init_reduction_value, BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_custom_functors_exespace_impl(
@@ -186,7 +186,7 @@ template <typename ExecutionSpace, typename DataType, typename... Properties,
 ValueType reduce(const ExecutionSpace& ex,
                  const ::Kokkos::View<DataType, Properties...>& view,
                  ValueType init_reduction_value, BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;
@@ -204,7 +204,7 @@ template <typename ExecutionSpace, typename DataType, typename... Properties,
 ValueType reduce(const std::string& label, const ExecutionSpace& ex,
                  const ::Kokkos::View<DataType, Properties...>& view,
                  ValueType init_reduction_value, BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;
@@ -258,7 +258,7 @@ template <
 KOKKOS_FUNCTION ValueType reduce(const TeamHandleType& teamHandle,
                                  IteratorType first, IteratorType last,
                                  ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_default_functors_team_impl(teamHandle, first, last,
@@ -273,7 +273,7 @@ KOKKOS_FUNCTION ValueType
 reduce(const TeamHandleType& teamHandle,
        const ::Kokkos::View<DataType, Properties...>& view,
        ValueType init_reduction_value) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;
@@ -294,7 +294,7 @@ KOKKOS_FUNCTION ValueType reduce(const TeamHandleType& teamHandle,
                                  IteratorType first, IteratorType last,
                                  ValueType init_reduction_value,
                                  BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::reduce_custom_functors_team_impl(teamHandle, first, last,
@@ -309,7 +309,7 @@ KOKKOS_FUNCTION ValueType
 reduce(const TeamHandleType& teamHandle,
        const ::Kokkos::View<DataType, Properties...>& view,
        ValueType init_reduction_value, BinaryOp joiner) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   namespace KE = ::Kokkos::Experimental;

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

@@ -117,7 +117,7 @@ ValueType transform_reduce(const ExecutionSpace& ex, IteratorType1 first1,
                            ValueType init_reduction_value,
                            BinaryJoinerType joiner,
                            BinaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_exespace_impl(
@@ -136,7 +136,7 @@ ValueType transform_reduce(const std::string& label, const ExecutionSpace& ex,
                            IteratorType2 first2, ValueType init_reduction_value,
                            BinaryJoinerType joiner,
                            BinaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_exespace_impl(
@@ -157,7 +157,7 @@ ValueType transform_reduce(
     ValueType init_reduction_value, BinaryJoinerType joiner,
     BinaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(first_view);
@@ -182,7 +182,7 @@ ValueType transform_reduce(
     ValueType init_reduction_value, BinaryJoinerType joiner,
     BinaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(first_view);
@@ -208,7 +208,7 @@ ValueType transform_reduce(const ExecutionSpace& ex, IteratorType first1,
                            IteratorType last1, ValueType init_reduction_value,
                            BinaryJoinerType joiner,
                            UnaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_exespace_impl(
@@ -228,7 +228,7 @@ ValueType transform_reduce(const std::string& label, const ExecutionSpace& ex,
                            ValueType init_reduction_value,
                            BinaryJoinerType joiner,
                            UnaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_exespace_impl(
@@ -248,7 +248,7 @@ ValueType transform_reduce(const ExecutionSpace& ex,
                            BinaryJoinerType joiner,
                            UnaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(view);
@@ -270,7 +270,7 @@ ValueType transform_reduce(const std::string& label, const ExecutionSpace& ex,
                            BinaryJoinerType joiner,
                            UnaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(view);
@@ -345,7 +345,7 @@ KOKKOS_FUNCTION ValueType transform_reduce(
     const TeamHandleType& teamHandle, IteratorType1 first1, IteratorType1 last1,
     IteratorType2 first2, ValueType init_reduction_value,
     BinaryJoinerType joiner, BinaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_team_impl(
@@ -366,7 +366,7 @@ transform_reduce(const TeamHandleType& teamHandle,
                  ValueType init_reduction_value, BinaryJoinerType joiner,
                  BinaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(first_view);
@@ -393,7 +393,7 @@ KOKKOS_FUNCTION ValueType transform_reduce(const TeamHandleType& teamHandle,
                                            ValueType init_reduction_value,
                                            BinaryJoinerType joiner,
                                            UnaryTransform transformer) {
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   return Impl::transform_reduce_custom_functors_team_impl(
@@ -412,7 +412,7 @@ transform_reduce(const TeamHandleType& teamHandle,
                  ValueType init_reduction_value, BinaryJoinerType joiner,
                  UnaryTransform transformer) {
   namespace KE = ::Kokkos::Experimental;
-  static_assert(std::is_move_constructible<ValueType>::value,
+  static_assert(std::is_move_constructible_v<ValueType>,
                 "ValueType must be move constructible.");
 
   Impl::static_assert_is_admissible_to_kokkos_std_algorithms(view);

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

@@ -33,12 +33,12 @@ struct is_admissible_to_kokkos_std_algorithms : std::false_type {};
 template <typename T>
 struct is_admissible_to_kokkos_std_algorithms<
     T, std::enable_if_t<::Kokkos::is_view<T>::value && T::rank() == 1 &&
-                        (std::is_same<typename T::traits::array_layout,
+                        (std::is_same_v<typename T::traits::array_layout,
-                                      Kokkos::LayoutLeft>::value ||
+                                        Kokkos::LayoutLeft> ||
-                         std::is_same<typename T::traits::array_layout,
+                         std::is_same_v<typename T::traits::array_layout,
-                                      Kokkos::LayoutRight>::value ||
+                                        Kokkos::LayoutRight> ||
-                         std::is_same<typename T::traits::array_layout,
+                         std::is_same_v<typename T::traits::array_layout,
-                                      Kokkos::LayoutStride>::value)>>
+                                        Kokkos::LayoutStride>)>>
     : std::true_type {};
 
 template <class ViewType>
@@ -102,8 +102,8 @@ struct are_random_access_iterators;
 template <class T>
 struct are_random_access_iterators<T> {
   static constexpr bool value =
-      is_iterator_v<T> && std::is_base_of<std::random_access_iterator_tag,
+      is_iterator_v<T> && std::is_base_of_v<std::random_access_iterator_tag,
-                                          typename T::iterator_category>::value;
+                                            typename T::iterator_category>;
 };
 
 template <class Head, class... Tail>
@@ -165,9 +165,8 @@ struct iterators_have_matching_difference_type<T> {
 
 template <class T1, class T2>
 struct iterators_have_matching_difference_type<T1, T2> {
-  static constexpr bool value =
+  static constexpr bool value = std::is_same_v<typename T1::difference_type,
-      std::is_same<typename T1::difference_type,
+                                               typename T2::difference_type>;
-                   typename T2::difference_type>::value;
 };
 
 template <class T1, class T2, class... Tail>

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

@@ -30,7 +30,7 @@ namespace Impl {
 template <class IteratorType1, class IteratorType2>
 struct StdMoveBackwardFunctor {
   using index_type = typename IteratorType1::difference_type;
-  static_assert(std::is_signed<index_type>::value,
+  static_assert(std::is_signed_v<index_type>,
                 "Kokkos: StdMoveBackwardFunctor requires signed index type");
 
   IteratorType1 m_last;

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

@@ -36,18 +36,18 @@ class RandomAccessIterator< ::Kokkos::View<DataType, Args...> > {
   using iterator_type = RandomAccessIterator<view_type>;
 
   using iterator_category = std::random_access_iterator_tag;
-  using value_type        = typename view_type::value_type;
+  using value_type        = typename view_type::non_const_value_type;
   using difference_type   = ptrdiff_t;
   using pointer           = typename view_type::pointer_type;
   using reference         = typename view_type::reference_type;
 
   static_assert(view_type::rank == 1 &&
-                    (std::is_same<typename view_type::traits::array_layout,
+                    (std::is_same_v<typename view_type::traits::array_layout,
-                                  Kokkos::LayoutLeft>::value ||
+                                    Kokkos::LayoutLeft> ||
-                     std::is_same<typename view_type::traits::array_layout,
+                     std::is_same_v<typename view_type::traits::array_layout,
-                                  Kokkos::LayoutRight>::value ||
+                                    Kokkos::LayoutRight> ||
-                     std::is_same<typename view_type::traits::array_layout,
+                     std::is_same_v<typename view_type::traits::array_layout,
-                                  Kokkos::LayoutStride>::value),
+                                    Kokkos::LayoutStride>),
                 "RandomAccessIterator only supports 1D Views with LayoutLeft, "
                 "LayoutRight, LayoutStride.");
 
@@ -61,9 +61,9 @@ class RandomAccessIterator< ::Kokkos::View<DataType, Args...> > {
 
 #ifndef KOKKOS_ENABLE_CXX17  // C++20 and beyond
   template <class OtherViewType>
-  requires(std::is_constructible_v<view_type, OtherViewType>) KOKKOS_FUNCTION
+    requires(std::is_constructible_v<view_type, OtherViewType>)
-      explicit(!std::is_convertible_v<OtherViewType, view_type>)
+  KOKKOS_FUNCTION explicit(!std::is_convertible_v<OtherViewType, view_type>)
-          RandomAccessIterator(const RandomAccessIterator<OtherViewType>& other)
+      RandomAccessIterator(const RandomAccessIterator<OtherViewType>& other)
       : m_view(other.m_view), m_current_index(other.m_current_index) {}
 #else
   template <

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

@@ -30,7 +30,7 @@ namespace Impl {
 template <class InputIterator>
 struct StdReverseFunctor {
   using index_type = typename InputIterator::difference_type;
-  static_assert(std::is_signed<index_type>::value,
+  static_assert(std::is_signed_v<index_type>,
                 "Kokkos: StdReverseFunctor requires signed index type");
 
   InputIterator m_first;

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

@@ -30,7 +30,7 @@ namespace Impl {
 template <class InputIterator, class OutputIterator>
 struct StdReverseCopyFunctor {
   using index_type = typename InputIterator::difference_type;
-  static_assert(std::is_signed<index_type>::value,
+  static_assert(std::is_signed_v<index_type>,
                 "Kokkos: StdReverseCopyFunctor requires signed index type");
 
   InputIterator m_last;

lib/kokkos/algorithms/unit_tests/CMakeLists.txt

@@ -1,12 +1,10 @@
-
 #Leave these here for now - I don't need transitive deps anyway
-KOKKOS_INCLUDE_DIRECTORIES(${CMAKE_CURRENT_BINARY_DIR})
+kokkos_include_directories(${CMAKE_CURRENT_BINARY_DIR})
-KOKKOS_INCLUDE_DIRECTORIES(REQUIRED_DURING_INSTALLATION_TESTING ${CMAKE_CURRENT_SOURCE_DIR})
+kokkos_include_directories(REQUIRED_DURING_INSTALLATION_TESTING ${CMAKE_CURRENT_SOURCE_DIR})
-KOKKOS_INCLUDE_DIRECTORIES(${CMAKE_CURRENT_SOURCE_DIR}/../src )
+kokkos_include_directories(${CMAKE_CURRENT_SOURCE_DIR}/../src)
-KOKKOS_INCLUDE_DIRECTORIES(${KOKKOS_SOURCE_DIR}/core/unit_test/category_files)
+kokkos_include_directories(${KOKKOS_SOURCE_DIR}/core/unit_test/category_files)
 
-
+set(ALGORITHM UnitTestMain.cpp)
-SET(ALGORITHM UnitTestMain.cpp)
 
 foreach(Tag Threads;Serial;OpenMP;Cuda;HPX;HIP;SYCL;OpenMPTarget)
   string(TOUPPER ${Tag} DEVICE)
@@ -23,21 +21,11 @@ foreach(Tag Threads;Serial;OpenMP;Cuda;HPX;HIP;SYCL;OpenMPTarget)
     # Generate a .cpp file for each one that runs it on the current backend (Tag),
     # and add this .cpp file to the sources for UnitTest_RandomAndSort.
     set(ALGO_SORT_SOURCES)
-    foreach(SOURCE_Input
+    foreach(SOURCE_Input TestSort TestSortByKey TestSortCustomComp TestBinSortA TestBinSortB TestNestedSort)
-	TestSort
-	TestSortByKey
-	TestSortCustomComp
-	TestBinSortA
-	TestBinSortB
-	TestNestedSort
-      )
       set(file ${dir}/${SOURCE_Input}.cpp)
       # Write to a temporary intermediate file and call configure_file to avoid
       # updating timestamps triggering unnecessary rebuilds on subsequent cmake runs.
-      file(WRITE ${dir}/dummy.cpp
+      file(WRITE ${dir}/dummy.cpp "#include <Test${Tag}_Category.hpp>\n" "#include <${SOURCE_Input}.hpp>\n")
-        "#include <Test${Tag}_Category.hpp>\n"
-        "#include <${SOURCE_Input}.hpp>\n"
-        )
       configure_file(${dir}/dummy.cpp ${file})
       list(APPEND ALGO_SORT_SOURCES ${file})
     endforeach()
@@ -47,14 +35,9 @@ foreach(Tag Threads;Serial;OpenMP;Cuda;HPX;HIP;SYCL;OpenMPTarget)
     # ------------------------------------------
     # do as above
     set(ALGO_RANDOM_SOURCES)
-    foreach(SOURCE_Input
+    foreach(SOURCE_Input TestRandom)
-	TestRandom
-      )
       set(file ${dir}/${SOURCE_Input}.cpp)
-      file(WRITE ${dir}/dummy.cpp
+      file(WRITE ${dir}/dummy.cpp "#include <Test${Tag}_Category.hpp>\n" "#include <${SOURCE_Input}.hpp>\n")
-        "#include <Test${Tag}_Category.hpp>\n"
-        "#include <${SOURCE_Input}.hpp>\n"
-        )
       configure_file(${dir}/dummy.cpp ${file})
       list(APPEND ALGO_RANDOM_SOURCES ${file})
     endforeach()
@@ -65,11 +48,7 @@ endforeach()
 # std set A
 # ------------------------------------------
 set(STDALGO_SOURCES_A)
-foreach(Name
+foreach(Name StdReducers StdAlgorithmsConstraints RandomAccessIterator)
-	StdReducers
-	StdAlgorithmsConstraints
-	RandomAccessIterator
-  )
   list(APPEND STDALGO_SOURCES_A Test${Name}.cpp)
 endforeach()
 
@@ -77,10 +56,7 @@ endforeach()
 # std set B
 # ------------------------------------------
 set(STDALGO_SOURCES_B)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsMinMaxElementOps)
-	StdAlgorithmsCommon
-	StdAlgorithmsMinMaxElementOps
-  )
   list(APPEND STDALGO_SOURCES_B Test${Name}.cpp)
 endforeach()
 
@@ -88,22 +64,23 @@ endforeach()
 # std set C
 # ------------------------------------------
 set(STDALGO_SOURCES_C)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsLexicographicalCompare
+  StdAlgorithmsCommon
-	StdAlgorithmsForEach
+  StdAlgorithmsLexicographicalCompare
-	StdAlgorithmsFind
+  StdAlgorithmsForEach
-	StdAlgorithmsFindFirstOf
+  StdAlgorithmsFind
-	StdAlgorithmsFindEnd
+  StdAlgorithmsFindFirstOf
-	StdAlgorithmsCount
+  StdAlgorithmsFindEnd
-	StdAlgorithmsEqual
+  StdAlgorithmsCount
-	StdAlgorithmsAllAnyNoneOf
+  StdAlgorithmsEqual
-	StdAlgorithmsAdjacentFind
+  StdAlgorithmsAllAnyNoneOf
-	StdAlgorithmsSearch
+  StdAlgorithmsAdjacentFind
-	StdAlgorithmsSearch_n
+  StdAlgorithmsSearch
-	StdAlgorithmsMismatch
+  StdAlgorithmsSearch_n
-	StdAlgorithmsMoveBackward
+  StdAlgorithmsMismatch
-  )
+  StdAlgorithmsMoveBackward
+)
   list(APPEND STDALGO_SOURCES_C Test${Name}.cpp)
 endforeach()
 
@@ -111,27 +88,28 @@ endforeach()
 # std set D
 # ------------------------------------------
 set(STDALGO_SOURCES_D)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsModOps
+  StdAlgorithmsCommon
-	StdAlgorithmsModSeqOps
+  StdAlgorithmsModOps
-	StdAlgorithmsReplace
+  StdAlgorithmsModSeqOps
-	StdAlgorithmsReplaceIf
+  StdAlgorithmsReplace
-	StdAlgorithmsReplaceCopy
+  StdAlgorithmsReplaceIf
-	StdAlgorithmsReplaceCopyIf
+  StdAlgorithmsReplaceCopy
-	StdAlgorithmsCopyIf
+  StdAlgorithmsReplaceCopyIf
-	StdAlgorithmsUnique
+  StdAlgorithmsCopyIf
-	StdAlgorithmsUniqueCopy
+  StdAlgorithmsUnique
-	StdAlgorithmsRemove
+  StdAlgorithmsUniqueCopy
-	StdAlgorithmsRemoveIf
+  StdAlgorithmsRemove
-	StdAlgorithmsRemoveCopy
+  StdAlgorithmsRemoveIf
-	StdAlgorithmsRemoveCopyIf
+  StdAlgorithmsRemoveCopy
-	StdAlgorithmsRotate
+  StdAlgorithmsRemoveCopyIf
-	StdAlgorithmsRotateCopy
+  StdAlgorithmsRotate
-	StdAlgorithmsReverse
+  StdAlgorithmsRotateCopy
-	StdAlgorithmsShiftLeft
+  StdAlgorithmsReverse
-	StdAlgorithmsShiftRight
+  StdAlgorithmsShiftLeft
-  )
+  StdAlgorithmsShiftRight
+)
   list(APPEND STDALGO_SOURCES_D Test${Name}.cpp)
 endforeach()
 
@@ -139,20 +117,21 @@ endforeach()
 # std set E
 # ------------------------------------------
 set(STDALGO_SOURCES_E)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsIsSorted
+  StdAlgorithmsCommon
-	StdAlgorithmsIsSortedUntil
+  StdAlgorithmsIsSorted
-	StdAlgorithmsPartitioningOps
+  StdAlgorithmsIsSortedUntil
-	StdAlgorithmsPartitionCopy
+  StdAlgorithmsPartitioningOps
-	StdAlgorithmsNumerics
+  StdAlgorithmsPartitionCopy
-	StdAlgorithmsAdjacentDifference
+  StdAlgorithmsNumerics
-	StdAlgorithmsExclusiveScan
+  StdAlgorithmsAdjacentDifference
-	StdAlgorithmsInclusiveScan
+  StdAlgorithmsExclusiveScan
-	StdAlgorithmsTransformUnaryOp
+  StdAlgorithmsInclusiveScan
-	StdAlgorithmsTransformExclusiveScan
+  StdAlgorithmsTransformUnaryOp
-	StdAlgorithmsTransformInclusiveScan
+  StdAlgorithmsTransformExclusiveScan
-  )
+  StdAlgorithmsTransformInclusiveScan
+)
   list(APPEND STDALGO_SOURCES_E Test${Name}.cpp)
 endforeach()
 
@@ -160,11 +139,7 @@ endforeach()
 # std team Q
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_Q)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamInclusiveScan StdAlgorithmsTeamTransformInclusiveScan)
-	StdAlgorithmsCommon
-	StdAlgorithmsTeamInclusiveScan
-	StdAlgorithmsTeamTransformInclusiveScan
-  )
   list(APPEND STDALGO_TEAM_SOURCES_Q Test${Name}.cpp)
 endforeach()
 
@@ -172,11 +147,7 @@ endforeach()
 # std team P
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_P)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamExclusiveScan StdAlgorithmsTeamTransformExclusiveScan)
-	StdAlgorithmsCommon
-	StdAlgorithmsTeamExclusiveScan
-	StdAlgorithmsTeamTransformExclusiveScan
-  )
   list(APPEND STDALGO_TEAM_SOURCES_P Test${Name}.cpp)
 endforeach()
 
@@ -184,14 +155,9 @@ endforeach()
 # std team M
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_M)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamTransformUnaryOp StdAlgorithmsTeamTransformBinaryOp
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamGenerate StdAlgorithmsTeamGenerate_n StdAlgorithmsTeamSwapRanges
-	StdAlgorithmsTeamTransformUnaryOp
+)
-	StdAlgorithmsTeamTransformBinaryOp
-	StdAlgorithmsTeamGenerate
-	StdAlgorithmsTeamGenerate_n
-	StdAlgorithmsTeamSwapRanges
-  )
   list(APPEND STDALGO_TEAM_SOURCES_M Test${Name}.cpp)
 endforeach()
 
@@ -199,14 +165,9 @@ endforeach()
 # std team L
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_L)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamIsSorted StdAlgorithmsTeamIsSortedUntil
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamIsPartitioned StdAlgorithmsTeamPartitionCopy StdAlgorithmsTeamPartitionPoint
-	StdAlgorithmsTeamIsSorted
+)
-	StdAlgorithmsTeamIsSortedUntil
-	StdAlgorithmsTeamIsPartitioned
-	StdAlgorithmsTeamPartitionCopy
-	StdAlgorithmsTeamPartitionPoint
-  )
   list(APPEND STDALGO_TEAM_SOURCES_L Test${Name}.cpp)
 endforeach()
 
@@ -214,13 +175,9 @@ endforeach()
 # std team I
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_I)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamUnique StdAlgorithmsTeamAdjacentDifference StdAlgorithmsTeamReduce
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamTransformReduce
-	StdAlgorithmsTeamUnique
+)
-	StdAlgorithmsTeamAdjacentDifference
-	StdAlgorithmsTeamReduce
-	StdAlgorithmsTeamTransformReduce
-  )
   list(APPEND STDALGO_TEAM_SOURCES_I Test${Name}.cpp)
 endforeach()
 
@@ -228,18 +185,19 @@ endforeach()
 # std team H
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_H)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsTeamCopy
+  StdAlgorithmsCommon
-	StdAlgorithmsTeamCopy_n
+  StdAlgorithmsTeamCopy
-	StdAlgorithmsTeamCopyBackward
+  StdAlgorithmsTeamCopy_n
-	StdAlgorithmsTeamCopyIf
+  StdAlgorithmsTeamCopyBackward
-	StdAlgorithmsTeamUniqueCopy
+  StdAlgorithmsTeamCopyIf
-	StdAlgorithmsTeamRemove
+  StdAlgorithmsTeamUniqueCopy
-	StdAlgorithmsTeamRemoveIf
+  StdAlgorithmsTeamRemove
-	StdAlgorithmsTeamRemoveCopy
+  StdAlgorithmsTeamRemoveIf
-	StdAlgorithmsTeamRemoveCopyIf
+  StdAlgorithmsTeamRemoveCopy
-  )
+  StdAlgorithmsTeamRemoveCopyIf
+)
   list(APPEND STDALGO_TEAM_SOURCES_H Test${Name}.cpp)
 endforeach()
 
@@ -247,13 +205,9 @@ endforeach()
 # std team G
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_G)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamMove StdAlgorithmsTeamMoveBackward StdAlgorithmsTeamShiftLeft
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamShiftRight
-	StdAlgorithmsTeamMove
+)
-	StdAlgorithmsTeamMoveBackward
-	StdAlgorithmsTeamShiftLeft
-	StdAlgorithmsTeamShiftRight
-  )
   list(APPEND STDALGO_TEAM_SOURCES_G Test${Name}.cpp)
 endforeach()
 
@@ -261,13 +215,9 @@ endforeach()
 # std team F
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_F)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamReverse StdAlgorithmsTeamReverseCopy StdAlgorithmsTeamRotate
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamRotateCopy
-	StdAlgorithmsTeamReverse
+)
-	StdAlgorithmsTeamReverseCopy
-	StdAlgorithmsTeamRotate
-	StdAlgorithmsTeamRotateCopy
-  )
   list(APPEND STDALGO_TEAM_SOURCES_F Test${Name}.cpp)
 endforeach()
 
@@ -275,15 +225,16 @@ endforeach()
 # std team E
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_E)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsTeamFill
+  StdAlgorithmsCommon
-	StdAlgorithmsTeamFill_n
+  StdAlgorithmsTeamFill
-	StdAlgorithmsTeamReplace
+  StdAlgorithmsTeamFill_n
-	StdAlgorithmsTeamReplaceIf
+  StdAlgorithmsTeamReplace
-	StdAlgorithmsTeamReplaceCopy
+  StdAlgorithmsTeamReplaceIf
-	StdAlgorithmsTeamReplaceCopyIf
+  StdAlgorithmsTeamReplaceCopy
-  )
+  StdAlgorithmsTeamReplaceCopyIf
+)
   list(APPEND STDALGO_TEAM_SOURCES_E Test${Name}.cpp)
 endforeach()
 
@@ -291,12 +242,7 @@ endforeach()
 # std team D
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_D)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamMinElement StdAlgorithmsTeamMaxElement StdAlgorithmsTeamMinMaxElement)
-	StdAlgorithmsCommon
-	StdAlgorithmsTeamMinElement
-	StdAlgorithmsTeamMaxElement
-	StdAlgorithmsTeamMinMaxElement
-  )
   list(APPEND STDALGO_TEAM_SOURCES_D Test${Name}.cpp)
 endforeach()
 
@@ -304,16 +250,17 @@ endforeach()
 # std team C
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_C)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsTeamFind
+  StdAlgorithmsCommon
-	StdAlgorithmsTeamFindIf
+  StdAlgorithmsTeamFind
-	StdAlgorithmsTeamFindIfNot
+  StdAlgorithmsTeamFindIf
-	StdAlgorithmsTeamAllOf
+  StdAlgorithmsTeamFindIfNot
-	StdAlgorithmsTeamAnyOf
+  StdAlgorithmsTeamAllOf
-	StdAlgorithmsTeamNoneOf
+  StdAlgorithmsTeamAnyOf
-	StdAlgorithmsTeamSearchN
+  StdAlgorithmsTeamNoneOf
-  )
+  StdAlgorithmsTeamSearchN
+)
   list(APPEND STDALGO_TEAM_SOURCES_C Test${Name}.cpp)
 endforeach()
 
@@ -321,13 +268,9 @@ endforeach()
 # std team B
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_B)
-foreach(Name
+foreach(Name StdAlgorithmsCommon StdAlgorithmsTeamEqual StdAlgorithmsTeamSearch StdAlgorithmsTeamFindEnd
-	StdAlgorithmsCommon
+             StdAlgorithmsTeamFindFirstOf
-	StdAlgorithmsTeamEqual
+)
-	StdAlgorithmsTeamSearch
-	StdAlgorithmsTeamFindEnd
-	StdAlgorithmsTeamFindFirstOf
-  )
   list(APPEND STDALGO_TEAM_SOURCES_B Test${Name}.cpp)
 endforeach()
 
@@ -335,34 +278,33 @@ endforeach()
 # std team A
 # ------------------------------------------
 set(STDALGO_TEAM_SOURCES_A)
-foreach(Name
+foreach(
-	StdAlgorithmsCommon
+  Name
-	StdAlgorithmsTeamAdjacentFind
+  StdAlgorithmsCommon
-	StdAlgorithmsTeamCount
+  StdAlgorithmsTeamAdjacentFind
-	StdAlgorithmsTeamCountIf
+  StdAlgorithmsTeamCount
-	StdAlgorithmsTeamForEach
+  StdAlgorithmsTeamCountIf
-	StdAlgorithmsTeamForEachN
+  StdAlgorithmsTeamForEach
-	StdAlgorithmsTeamLexicographicalCompare
+  StdAlgorithmsTeamForEachN
-	StdAlgorithmsTeamMismatch
+  StdAlgorithmsTeamLexicographicalCompare
-  )
+  StdAlgorithmsTeamMismatch
+)
   list(APPEND STDALGO_TEAM_SOURCES_A Test${Name}.cpp)
 endforeach()
 
 # FIXME_OPENMPTARGET - remove sort test as it leads to ICE with clang/16 and above at compile time.
-if(KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL "Clang" AND KOKKOS_CXX_COMPILER_VERSION VERSION_GREATER_EQUAL 16.0.0)
+if(KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL "Clang" AND KOKKOS_CXX_COMPILER_VERSION
-    list(REMOVE_ITEM ALGO_SORT_SOURCES
+                                                                              VERSION_GREATER_EQUAL 16.0.0
-    TestSort.cpp
+)
-  )
+  list(REMOVE_ITEM ALGO_SORT_SOURCES TestSort.cpp)
 endif()
 
 # FIXME_OPENMPTARGET remove tests for OpenMPTarget because in these cases
 # the impl needs to use either Kokkos or tailored reducers
 # which results in runtime memory errors.
 if(KOKKOS_ENABLE_OPENMPTARGET)
-  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_L
+  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_L TestStdAlgorithmsTeamIsPartitioned.cpp
-    TestStdAlgorithmsTeamIsPartitioned.cpp
+       TestStdAlgorithmsTeamPartitionPoint.cpp TestStdAlgorithmsTeamPartitionCopy.cpp
-    TestStdAlgorithmsTeamPartitionPoint.cpp
-    TestStdAlgorithmsTeamPartitionCopy.cpp
   )
 endif()
 
@@ -370,7 +312,9 @@ endif()
 # in these cases the impl needs to use either Kokkos or
 # tailored reducers which results in runtime memory errors.
 if(KOKKOS_ENABLE_OPENMPTARGET)
-  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_C
+  list(
+    REMOVE_ITEM
+    STDALGO_TEAM_SOURCES_C
     TestStdAlgorithmsTeamFind.cpp
     TestStdAlgorithmsTeamFindIf.cpp
     TestStdAlgorithmsTeamFindIfNot.cpp
@@ -386,35 +330,20 @@ endif()
 # FRIZZI: 04/26/2023: not sure if the compilation error is still applicable
 # but we conservatively leave this guard on
 if(NOT (KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL IntelLLVM))
-  KOKKOS_ADD_EXECUTABLE_AND_TEST(
+  kokkos_add_executable_and_test(
-    UnitTest_Sort
+    UnitTest_Sort SOURCES UnitTestMain.cpp TestStdAlgorithmsCommon.cpp ${ALGO_SORT_SOURCES}
-    SOURCES
-    UnitTestMain.cpp
-    TestStdAlgorithmsCommon.cpp
-    ${ALGO_SORT_SOURCES}
   )
 
-  KOKKOS_ADD_EXECUTABLE_AND_TEST(
+  kokkos_add_executable_and_test(UnitTest_Random SOURCES UnitTestMain.cpp ${ALGO_RANDOM_SOURCES})
-    UnitTest_Random
-    SOURCES
-    UnitTestMain.cpp
-    ${ALGO_RANDOM_SOURCES}
-  )
 endif()
 
 # FIXME_OPENMPTARGET: These tests cause internal compiler errors as of 09/01/22
 # when compiling for Intel's Xe-HP GPUs.
 if(KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL IntelLLVM)
-  list(REMOVE_ITEM STDALGO_SOURCES_D
+  list(REMOVE_ITEM STDALGO_SOURCES_D TestStdAlgorithmsCopyIf.cpp TestStdAlgorithmsRemoveCopy.cpp
-    TestStdAlgorithmsCopyIf.cpp
+       TestStdAlgorithmsUnique.cpp TestStdAlgorithmsUniqueCopy.cpp
-    TestStdAlgorithmsRemoveCopy.cpp
-    TestStdAlgorithmsUnique.cpp
-    TestStdAlgorithmsUniqueCopy.cpp
-  )
-  list(REMOVE_ITEM STDALGO_SOURCES_E
-    TestStdAlgorithmsExclusiveScan.cpp
-    TestStdAlgorithmsInclusiveScan.cpp
   )
+  list(REMOVE_ITEM STDALGO_SOURCES_E TestStdAlgorithmsExclusiveScan.cpp TestStdAlgorithmsInclusiveScan.cpp)
 endif()
 
 # FIXME_OPENMPTARGET remove tests for OpenMPTarget
@@ -422,48 +351,31 @@ endif()
 if(KOKKOS_ENABLE_OPENMPTARGET)
   # the following use either Kokkos or tailored reducers
   # which results in runtime memory errors.
-  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_B
+  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_B TestStdAlgorithmsTeamFindEnd.cpp TestStdAlgorithmsTeamFindFirstOf.cpp
-    TestStdAlgorithmsTeamFindEnd.cpp
+       TestStdAlgorithmsTeamSearch.cpp
-    TestStdAlgorithmsTeamFindFirstOf.cpp
-    TestStdAlgorithmsTeamSearch.cpp
   )
 
-  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_A
+  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_A TestStdAlgorithmsTeamAdjacentFind.cpp
-    TestStdAlgorithmsTeamAdjacentFind.cpp
+       TestStdAlgorithmsTeamLexicographicalCompare.cpp TestStdAlgorithmsTeamMismatch.cpp
-    TestStdAlgorithmsTeamLexicographicalCompare.cpp
-    TestStdAlgorithmsTeamMismatch.cpp
   )
 
   # this causes an illegal memory access if team_members_have_matching_result
   # is called
-  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_M
+  list(REMOVE_ITEM STDALGO_TEAM_SOURCES_M TestStdAlgorithmsTeamTransformBinaryOp.cpp)
-    TestStdAlgorithmsTeamTransformBinaryOp.cpp
-  )
 endif()
 
 foreach(ID A;B;C;D;E)
-  KOKKOS_ADD_EXECUTABLE_AND_TEST(
+  kokkos_add_executable_and_test(AlgorithmsUnitTest_StdSet_${ID} SOURCES UnitTestMain.cpp ${STDALGO_SOURCES_${ID}})
-    AlgorithmsUnitTest_StdSet_${ID}
-    SOURCES
-    UnitTestMain.cpp
-    ${STDALGO_SOURCES_${ID}}
-    )
 endforeach()
 
 foreach(ID A;B;C;D;E;F;G;H;I;L;M;P;Q)
-  KOKKOS_ADD_EXECUTABLE_AND_TEST(
+  kokkos_add_executable_and_test(
-    AlgorithmsUnitTest_StdSet_Team_${ID}
+    AlgorithmsUnitTest_StdSet_Team_${ID} SOURCES UnitTestMain.cpp ${STDALGO_TEAM_SOURCES_${ID}}
-    SOURCES
+  )
-    UnitTestMain.cpp
-    ${STDALGO_TEAM_SOURCES_${ID}}
-    )
 endforeach()
 
 # FIXME_OPENMPTARGET This test causes internal compiler errors as of 09/01/22
 # when compiling for Intel's Xe-HP GPUs.
 if(NOT (KOKKOS_ENABLE_OPENMPTARGET AND KOKKOS_CXX_COMPILER_ID STREQUAL IntelLLVM))
-  KOKKOS_ADD_EXECUTABLE(
+  kokkos_add_executable(AlgorithmsUnitTest_StdAlgoCompileOnly SOURCES TestStdAlgorithmsCompileOnly.cpp)
-    AlgorithmsUnitTest_StdAlgoCompileOnly
-    SOURCES TestStdAlgorithmsCompileOnly.cpp
-  )
 endif()

lib/kokkos/algorithms/unit_tests/TestBinSortA.hpp

@@ -31,13 +31,13 @@ struct bin3d_is_sorted_struct {
   using value_type      = unsigned int;
   using execution_space = ExecutionSpace;
 
-  Kokkos::View<Scalar * [3], ExecutionSpace> keys;
+  Kokkos::View<Scalar* [3], ExecutionSpace> keys;
 
   int max_bins;
   Scalar min;
   Scalar max;
 
-  bin3d_is_sorted_struct(Kokkos::View<Scalar * [3], ExecutionSpace> keys_,
+  bin3d_is_sorted_struct(Kokkos::View<Scalar* [3], ExecutionSpace> keys_,
                          int max_bins_, Scalar min_, Scalar max_)
       : keys(keys_), max_bins(max_bins_), min(min_), max(max_) {}
   KOKKOS_INLINE_FUNCTION
@@ -65,9 +65,9 @@ struct sum3D {
   using value_type      = double;
   using execution_space = ExecutionSpace;
 
-  Kokkos::View<Scalar * [3], ExecutionSpace> keys;
+  Kokkos::View<Scalar* [3], ExecutionSpace> keys;
 
-  sum3D(Kokkos::View<Scalar * [3], ExecutionSpace> keys_) : keys(keys_) {}
+  sum3D(Kokkos::View<Scalar* [3], ExecutionSpace> keys_) : keys(keys_) {}
   KOKKOS_INLINE_FUNCTION
   void operator()(int i, double& count) const {
     count += keys(i, 0);
@@ -77,8 +77,8 @@ struct sum3D {
 };
 
 template <class ExecutionSpace, typename KeyType>
-void test_3D_sort_impl(unsigned int n) {
+void test_3D_sort_impl(size_t n) {
-  using KeyViewType = Kokkos::View<KeyType * [3], ExecutionSpace>;
+  using KeyViewType = Kokkos::View<KeyType* [3], ExecutionSpace>;
 
   KeyViewType keys("Keys", n * n * n);
 
@@ -207,7 +207,7 @@ void test_sort_integer_overflow() {
   // array with two extrema in reverse order to expose integer overflow bug in
   // bin calculation
   T a[2]  = {Kokkos::Experimental::finite_max<T>::value,
-            Kokkos::Experimental::finite_min<T>::value};
+             Kokkos::Experimental::finite_min<T>::value};
   auto vd = Kokkos::create_mirror_view_and_copy(
       ExecutionSpace(), Kokkos::View<T[2], Kokkos::HostSpace>(a));
   Kokkos::sort(vd);
@@ -219,6 +219,10 @@ void test_sort_integer_overflow() {
 }  // namespace BinSortSetA
 
 TEST(TEST_CATEGORY, BinSortGenericTests) {
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
   using ExecutionSpace = TEST_EXECSPACE;
   using key_type       = unsigned;
   constexpr int N      = 171;
@@ -246,11 +250,11 @@ TEST(TEST_CATEGORY, BinSortEmptyView) {
   // does not matter if we use int or something else
   Kokkos::View<int*, ExecutionSpace> v("v", 0);
 
-  // test all exposed public sort methods
+  // test all exposed public sort methods are callable and do not throw
-  ASSERT_NO_THROW(Sorter.sort(ExecutionSpace(), v, 0, 0));
+  Sorter.sort(ExecutionSpace(), v, 0, 0);
-  ASSERT_NO_THROW(Sorter.sort(v, 0, 0));
+  Sorter.sort(v, 0, 0);
-  ASSERT_NO_THROW(Sorter.sort(ExecutionSpace(), v));
+  Sorter.sort(ExecutionSpace(), v);
-  ASSERT_NO_THROW(Sorter.sort(v));
+  Sorter.sort(v);
 }
 
 TEST(TEST_CATEGORY, BinSortEmptyKeysView) {
@@ -263,7 +267,26 @@ TEST(TEST_CATEGORY, BinSortEmptyKeysView) {
   BinOp_t binOp(5, 0, 10);
   Kokkos::BinSort<KeyViewType, BinOp_t> Sorter(ExecutionSpace{}, kv, binOp);
 
-  ASSERT_NO_THROW(Sorter.create_permute_vector(ExecutionSpace{}));
+  Sorter.create_permute_vector(ExecutionSpace{});  // does not throw
+}
+
+// BinSort may delegate sorting within bins to std::sort when running on host
+// and having a sufficiently large number of items within a single bin (10 by
+// default). Test that this is done without undefined behavior when accessing
+// the boundaries of the bin. Should be used in conjunction with a memory
+// sanitizer or bounds check.
+TEST(TEST_CATEGORY, BinSort_issue_7221) {
+  using ExecutionSpace = TEST_EXECSPACE;
+
+  using KeyViewType = Kokkos::View<int*, ExecutionSpace>;
+  KeyViewType kv("kv", 11);
+
+  using BinOp_t = Kokkos::BinOp1D<KeyViewType>;
+  BinOp_t binOp(1, -10, 10);
+  Kokkos::BinSort<KeyViewType, BinOp_t> Sorter(ExecutionSpace{}, kv, binOp,
+                                               /*sort_within_bins*/ true);
+
+  Sorter.create_permute_vector(ExecutionSpace{});  // does not throw
 }
 
 }  // namespace Test

lib/kokkos/algorithms/unit_tests/TestBinSortB.hpp

@@ -185,6 +185,10 @@ void run_for_rank2() {
 }  // namespace BinSortSetB
 
 TEST(TEST_CATEGORY, BinSortUnsignedKeyLayoutStrideValues) {
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
   using ExeSpace = TEST_EXECSPACE;
   using key_type = unsigned;
   BinSortSetB::run_for_rank1<ExeSpace, key_type, int>();

lib/kokkos/algorithms/unit_tests/TestNestedSort.hpp

@@ -386,6 +386,11 @@ void test_nested_sort_by_key(unsigned int N, KeyType minKey, KeyType maxKey,
 }  // namespace NestedSortImpl
 
 TEST(TEST_CATEGORY, NestedSort) {
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
+
   using ExecutionSpace = TEST_EXECSPACE;
   NestedSortImpl::test_nested_sort<ExecutionSpace, unsigned>(171, 0U, UINT_MAX);
   NestedSortImpl::test_nested_sort<ExecutionSpace, float>(42, -1e6f, 1e6f);
@@ -394,6 +399,11 @@ TEST(TEST_CATEGORY, NestedSort) {
 }
 
 TEST(TEST_CATEGORY, NestedSortByKey) {
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
+
   using ExecutionSpace = TEST_EXECSPACE;
 
   // Second/third template arguments are key and value respectively.

lib/kokkos/algorithms/unit_tests/TestRandom.hpp

@@ -542,6 +542,11 @@ void test_duplicate_stream() {
 }  // namespace AlgoRandomImpl
 
 TEST(TEST_CATEGORY, Random_XorShift64) {
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
+
   using ExecutionSpace = TEST_EXECSPACE;
 
 #if defined(KOKKOS_ENABLE_SYCL) || defined(KOKKOS_ENABLE_CUDA) || \
@@ -562,6 +567,10 @@ TEST(TEST_CATEGORY, Random_XorShift64) {
 
 TEST(TEST_CATEGORY, Random_XorShift1024_0) {
   using ExecutionSpace = TEST_EXECSPACE;
+  // FIXME_OPENMPTARGET - causes runtime failure with CrayClang compiler
+#if defined(KOKKOS_COMPILER_CRAY_LLVM) && defined(KOKKOS_ENABLE_OPENMPTARGET)
+  GTEST_SKIP() << "known to fail with OpenMPTarget+Cray LLVM";
+#endif
 
 #if defined(KOKKOS_ENABLE_SYCL) || defined(KOKKOS_ENABLE_CUDA) || \
     defined(KOKKOS_ENABLE_HIP)
@@ -589,7 +598,7 @@ TEST(TEST_CATEGORY, Multi_streams) {
 #endif
 
 #if defined(KOKKOS_ENABLE_SYCL) && defined(KOKKOS_IMPL_ARCH_NVIDIA_GPU)
-  if constexpr (std::is_same_v<ExecutionSpace, Kokkos::Experimental::SYCL>) {
+  if constexpr (std::is_same_v<ExecutionSpace, Kokkos::SYCL>) {
     GTEST_SKIP() << "Failing on NVIDIA GPUs";  // FIXME_SYCL
   }
 #endif

lib/kokkos/algorithms/unit_tests/TestRandomAccessIterator.cpp

@@ -23,7 +23,7 @@ namespace stdalgos {
 
 struct random_access_iterator_test : std_algorithms_test {
  public:
-  virtual void SetUp() {
+  void SetUp() override {
     Kokkos::parallel_for(m_static_view.extent(0),
                          AssignIndexFunctor<static_view_t>(m_static_view));
 
@@ -264,6 +264,37 @@ TEST_F(random_access_iterator_test, traits_helpers) {
   static_assert(KE::Impl::are_iterators_v<T1_t, T2_t, T3_t>);
   static_assert(KE::Impl::are_random_access_iterators_v<T1_t, T2_t, T3_t>);
   static_assert(!KE::Impl::are_iterators_v<int, T2_t, T3_t>);
+
+  static_assert(std::is_same_v<decltype(KE::begin(m_static_view))::value_type,
+                               value_type>);
+  static_assert(std::is_same_v<decltype(KE::begin(m_dynamic_view))::value_type,
+                               value_type>);
+  static_assert(std::is_same_v<decltype(KE::begin(m_strided_view))::value_type,
+                               value_type>);
+
+  static_assert(
+      std::is_same_v<decltype(KE::end(m_static_view))::value_type, value_type>);
+  static_assert(std::is_same_v<decltype(KE::end(m_dynamic_view))::value_type,
+                               value_type>);
+  static_assert(std::is_same_v<decltype(KE::end(m_strided_view))::value_type,
+                               value_type>);
+
+  static_assert(
+      std::is_same_v<decltype(KE::begin(m_static_view))::value_type,
+                     decltype(KE::cbegin(m_static_view))::value_type>);
+  static_assert(
+      std::is_same_v<decltype(KE::begin(m_dynamic_view))::value_type,
+                     decltype(KE::cbegin(m_dynamic_view))::value_type>);
+  static_assert(
+      std::is_same_v<decltype(KE::begin(m_strided_view))::value_type,
+                     decltype(KE::cbegin(m_strided_view))::value_type>);
+
+  static_assert(std::is_same_v<decltype(KE::end(m_static_view))::value_type,
+                               decltype(KE::cend(m_static_view))::value_type>);
+  static_assert(std::is_same_v<decltype(KE::end(m_dynamic_view))::value_type,
+                               decltype(KE::cend(m_dynamic_view))::value_type>);
+  static_assert(std::is_same_v<decltype(KE::end(m_strided_view))::value_type,
+                               decltype(KE::cend(m_strided_view))::value_type>);
 }
 
 }  // namespace stdalgos