Merge pull request #1696 from LLNL/task/rhornung67/v2024.07.0-RC

Merge RC branch to main for release

Author: rhornung67

.gitignore

@@ -14,3 +14,5 @@
 /install_*/
 /install-*/
 /Debug/
+*.swp
+*.orig

.gitlab-ci.yml

@@ -75,7 +75,7 @@ stages:
     include:
       - local: '.gitlab/custom-jobs-and-variables.yml'
       - project: 'radiuss/radiuss-shared-ci'
-        ref: 'v2024.04.0'
+        ref: 'v2024.06.0'
         file: 'pipelines/${CI_MACHINE}.yml'
       - artifact: '${CI_MACHINE}-jobs.yml'
         job: 'generate-job-lists'
@@ -91,20 +91,20 @@ stages:
 trigger-rajaperf:
   stage: multi-project
   rules:
-    - if: '$CI_COMMIT_BRANCH == "${MP_BRANCH}" || $MULTI_PROJECT == "ON"' #run only if ...
+    - if: $CI_COMMIT_BRANCH == $MP_BRANCH || $MULTI_PROJECT == "ON" #run only if ...
   variables:
     UPDATE_RAJA: ${MP_BRANCH}
   trigger:
     project: radiuss/rajaperf
     branch: develop
-    strategy: depend
 
 include:
+  # Sets ID tokens for every job using `default:`
   - project: 'lc-templates/id_tokens'
     file: 'id_tokens.yml'
   # [Optional] checks preliminary to running the actual CI test
   - project: 'radiuss/radiuss-shared-ci'
-    ref: 'v2024.04.0'
+    ref: 'v2024.06.0'
     file: 'utilities/preliminary-ignore-draft-pr.yml'
   # pipelines subscribed by the project
   - local: '.gitlab/subscribed-pipelines.yml'

.gitlab/custom-jobs-and-variables.yml

@@ -21,7 +21,7 @@ variables:
 # Project specific variants for ruby
   PROJECT_RUBY_VARIANTS: "~shared +openmp +vectorization +tests"
 # Project specific deps for ruby
-  PROJECT_RUBY_DEPS: ""
+  PROJECT_RUBY_DEPS: "^blt@develop "
 
 # Poodle
 # Arguments for top level allocation
@@ -31,7 +31,7 @@ variables:
 # Project specific variants for poodle
   PROJECT_POODLE_VARIANTS: "~shared +openmp +vectorization +tests"
 # Project specific deps for poodle
-  PROJECT_POODLE_DEPS: ""
+  PROJECT_POODLE_DEPS: "^blt@develop "
 
 # Corona
 # Arguments for top level allocation
@@ -70,3 +70,15 @@ variables:
   artifacts:
     reports:
       junit: junit.xml
+
+.reproducer_vars:
+  script:
+    - |
+      echo -e "
+      # Required variables \n
+      export MODULE_LIST=\"${MODULE_LIST}\" \n
+      export SPEC=\"${SPEC//\"/\\\"}\" \n
+      # Allow to set job script for debugging (only this differs from CI) \n
+      export DEBUG_MODE=true \n
+      # Using the CI build cache is optional and requires a token. Set it like so: \n
+      # export REGISTRY_TOKEN=\"<your token here>\" \n"

.gitlab/jobs/corona.yml

@@ -8,9 +8,7 @@
 # Override reproducer section to define project specific variables.
 .corona_reproducer_vars:
   script:
-    - |
-      echo -e "export MODULE_LIST=\"${MODULE_LIST}\""
-      echo -e "export SPEC=\"${SPEC//\"/\\\"}\""
+    - !reference [.reproducer_vars, script]
 
 ########################
 # Overridden shared jobs
@@ -33,3 +31,9 @@ rocmcc_5_7_0_hip_desul_atomics:
     SPEC: " ~shared +rocm ~openmp +tests +desul amdgpu_target=gfx906 %rocmcc@=5.7.0 ^hip@5.7.0 ^blt@develop"
   extends: .job_on_corona
 
+clang_19_0_0_sycl_gcc_10_3_1_rocmcc_5_7_1_hip:
+  variables:
+    SPEC: " ~shared +sycl ~openmp +tests %clang@=19.0.0 cxxflags==\"-w -fsycl -fsycl-unnamed-lambda -fsycl-targets=amdgcn-amd-amdhsa -Xsycl-target-backend --offload-arch=gfx906\" ^blt@develop"
+    MODULE_LIST: "rocm/5.7.1"
+  extends: .job_on_corona
+

.gitlab/jobs/lassen.yml

@@ -8,9 +8,7 @@
 # Override reproducer section to define project specific variables.
 .lassen_reproducer_vars:
   script:
-    - |
-      echo -e "export MODULE_LIST=\"${MODULE_LIST}\""
-      echo -e "export SPEC=\"${SPEC//\"/\\\"}\""
+    - !reference [.reproducer_vars, script]
 
 ########################
 # Overridden shared jobs
@@ -68,3 +66,13 @@ gcc_8_3_1_cuda_10_1_243_desul_atomics:
   variables:
     SPEC: " ~shared +openmp +tests +cuda +desul %gcc@=8.3.1 cuda_arch=70 ^cuda@10.1.243+allow-unsupported-compilers ^blt@develop"
   extends: .job_on_lassen
+
+# Warning: Allowed to fail temporarily
+# Deactivated due to issues with OpenMP Target and various tests and compilers.
+clang_16_0_6_ibm_omptarget:
+  variables:
+    SPEC: " ~shared +openmp +omptarget +tests %clang@=16.0.6.ibm.gcc.8.3.1 ^blt@develop"
+    ON_LASSEN: "OFF"
+  extends: .job_on_lassen
+  allow_failure: true
+

.gitlab/jobs/poodle.yml

@@ -5,38 +5,42 @@
 # SPDX-License-Identifier: (BSD-3-Clause)
 ##############################################################################
 
-# Override reproducer section to define projet specific variables.
+# Override reproducer section to define project specific variables.
 .poodle_reproducer_vars:
   script:
-    - |
-      echo -e "export MODULE_LIST=\"${MODULE_LIST}\""
-      echo -e "export SPEC=\"${SPEC//\"/\\\"}\""
+    - !reference [.reproducer_vars, script]
 
 ########################
 # Overridden shared jobs
 ########################
 # We duplicate the shared jobs description and add necessary changes for this
 # project. We keep ${PROJECT_<MACHINE>_VARIANTS} and ${PROJECT_<MACHINE>_DEPS}
-# So that the comparison with the original job is easier.
+# when possible so that the comparison with the original job is easier.
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
 clang_14_0_6:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %clang@=14.0.6 ^blt@develop"
+    SPEC: " ~shared +openmp +omptask +tests %clang@=14.0.6 ${PROJECT_POODLE_DEPS}"
   extends: .job_on_poodle
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
 gcc_10_3_1:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %gcc@=10.3.1 ^blt@develop"
+    SPEC: " ~shared +openmp +omptask +tests %gcc@=10.3.1 ${PROJECT_POODLE_DEPS}"
   extends: .job_on_poodle
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
+# Deactivated (too long on poodle)
 intel_19_1_2_gcc_10_3_1:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %intel@=19.1.2.gcc.10.3.1 ^blt@develop"
+    ON_POODLE: "OFF"
+    SPEC: " ~shared +openmp +omptask +tests %intel@=19.1.2.gcc.10.3.1 ${PROJECT_POODLE_DEPS}"
   extends: .job_on_poodle
 
+# Allowed to fail
 intel_2022_1_0:
   variables:
-    SPEC: "${PROJECT_POODLE_VARIANTS} %intel@=2022.1.0 ${PROJECT_POODLE_DEPS} ^blt@develop"
+    SPEC: "${PROJECT_POODLE_VARIANTS} %intel@=2022.1.0 ${PROJECT_POODLE_DEPS}"
   allow_failure: true
   extends: .job_on_poodle
 

.gitlab/jobs/ruby.yml

@@ -8,35 +8,37 @@
 # Override reproducer section to define project specific variables.
 .ruby_reproducer_vars:
   script:
-    - |
-      echo -e "export MODULE_LIST=\"${MODULE_LIST}\""
-      echo -e "export SPEC=\"${SPEC//\"/\\\"}\""
+    - !reference [.reproducer_vars, script]
 
 ########################
 # Overridden shared jobs
 ########################
 # We duplicate the shared jobs description and add necessary changes for this
 # project. We keep ${PROJECT_<MACHINE>_VARIANTS} and ${PROJECT_<MACHINE>_DEPS}
-# So that the comparison with the original job is easier.
+# when possible so that the comparison with the original job is easier.
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
 clang_14_0_6:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %clang@=14.0.6 ^blt@develop"
+    SPEC: " ~shared +openmp +omptask +tests %clang@=14.0.6 ${PROJECT_RUBY_DEPS}"
   extends: .job_on_ruby
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
 gcc_10_3_1:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %gcc@=10.3.1 ^blt@develop"
+    SPEC: " ~shared +openmp +omptask +tests %gcc@=10.3.1 ${PROJECT_RUBY_DEPS}"
   extends: .job_on_ruby
 
+# Identical to shared job, but use OpenMP tasks and no vectorization
 intel_19_1_2_gcc_10_3_1:
   variables:
-    SPEC: " ~shared +openmp +omptask +tests %intel@=19.1.2.gcc.10.3.1 ^blt@develop"
+    SPEC: " ~shared +openmp +omptask +tests %intel@=19.1.2.gcc.10.3.1 ${PROJECT_RUBY_DEPS}"
   extends: .job_on_ruby
 
+# Allowed to fail
 intel_2022_1_0:
   variables:
-    SPEC: "${PROJECT_RUBY_VARIANTS} %intel@=2022.1.0 ${PROJECT_RUBY_DEPS} ^blt@develop"
+    SPEC: "${PROJECT_RUBY_VARIANTS} %intel@=2022.1.0 ${PROJECT_RUBY_DEPS}"
   allow_failure: true
   extends: .job_on_ruby
 

.gitlab/jobs/tioga.yml

@@ -8,9 +8,7 @@
 # Override reproducer section to define project specific variables.
 .tioga_reproducer_vars:
   script:
-    - |
-      echo -e "export MODULE_LIST=\"${MODULE_LIST}\""
-      echo -e "export SPEC=\"${SPEC//\"/\\\"}\""
+    - !reference [.reproducer_vars, script]
 
 ########################
 # Overridden shared jobs
@@ -28,12 +26,12 @@
 # ${PROJECT_<MACHINE>_DEPS} in the extra jobs. There is no reason not to fully
 # describe the spec here.
 
-rocmcc_5_7_1_hip_desul_atomics:
+rocmcc_6_1_1_hip_desul_atomics:
   variables:
-    SPEC: "~shared +rocm ~openmp +desul +tests amdgpu_target=gfx90a %rocmcc@=5.7.1 ^hip@5.7.1 ^blt@develop"
+    SPEC: "~shared +rocm ~openmp +desul +tests amdgpu_target=gfx90a %rocmcc@=6.1.1 ^hip@6.1.1 ^blt@develop"
   extends: .job_on_tioga
 
-rocmcc_5_7_1_hip_openmp:
+rocmcc_6_1_1_hip_openmp:
   variables:
-    SPEC: "~shared +rocm +openmp +omptask +tests amdgpu_target=gfx90a %rocmcc@=5.7.1 ^hip@5.7.1 ^blt@develop"
+    SPEC: "~shared +rocm +openmp +omptask +tests amdgpu_target=gfx90a %rocmcc@=6.1.1 ^hip@6.1.1 ^blt@develop"
   extends: .job_on_tioga

.uberenv_config.json

@@ -4,7 +4,7 @@
 "package_final_phase" : "initconfig",
 "package_source_dir" : "../..",
 "spack_url": "https://github.com/spack/spack.git",
-"spack_branch": "develop-2024-02-18",
+"spack_branch": "develop-2024-07-07",
 "spack_activate" : {},
 "spack_configs_path": "scripts/radiuss-spack-configs",
 "spack_packages_path": "scripts/radiuss-spack-configs/packages",

CMakeLists.txt

@@ -15,8 +15,8 @@ include(CMakeDependentOption)
 
 # Set version number
 set(RAJA_VERSION_MAJOR 2024)
-set(RAJA_VERSION_MINOR 02)
-set(RAJA_VERSION_PATCHLEVEL 2)
+set(RAJA_VERSION_MINOR 07)
+set(RAJA_VERSION_PATCHLEVEL 0)
 
 if (RAJA_LOADED AND (NOT RAJA_LOADED STREQUAL "${RAJA_VERSION_MAJOR}.${RAJA_VERSION_MINOR}.${RAJA_VERSION_PATCHLEVEL}"))
   message(FATAL_ERROR "You are mixing RAJA versions. Loaded is ${RAJA_LOADED}, expected ${RAJA_VERSION_MAJOR}.${RAJA_VERSION_MINOR}.${RAJA_VERSION_PATCHLEVEL}")
@@ -44,11 +44,7 @@ set(CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake/thirdparty" ${CMAKE_MODULE_PA
 
 include(cmake/SetupRajaOptions.cmake)
 
-if (ENABLE_HIP)
-  cmake_minimum_required(VERSION 3.23)
-else()
-  cmake_minimum_required(VERSION 3.20)
-endif()
+cmake_minimum_required(VERSION 3.23)
 
 # Detect C++ standard and add appropriate flag _before_ loading BLT
 set(COMPILERS_KNOWN_TO_CMAKE33 AppleClang Clang GNU MSVC)

RELEASE_NOTES.md

@@ -20,6 +20,57 @@ Notable changes include:
   * Bug fixes/improvements:
 
 
+Version 2024.07.0 -- Release date 2024-07-24
+============================================
+
+This release contains new features, improvements, and bugfixes.
+
+Notable changes include:
+
+  * New features / API changes:
+     * Added support for a "multi-reduction" operation which allows users to
+       perform a run time-defined number of reduction operations in a kernel.
+       Please see the RAJA User Guide for details and examples.
+     * Added first couple of sections for a "RAJA Cookbook" in the RAJA User
+       Guide. The goal is to provide users with more detailed guidance about
+       using RAJA features, choosing execution policies, etc. Additional
+       content will be provided in future releases.
+     * Added atomicLoad and atomicStore routines for correctness in some
+       use cases.
+     * Added OpenMP 5.1 implementations for atomicMin and atomicMax.
+     * Add SYCL reduction support in RAJA::launch
+
+  * Build changes/improvements:
+     * Update camp submodule to v2024.07.0 release. This will be a version
+       constraint for this release in RAJA Spack package.
+     * Minimum required CMake version bumped to 3.23.
+
+  * Bug fixes/improvements:
+     * Fix CMake issue for case when RAJA is used as a submodule dependency.
+     * Various fixes and improvements to builtin atomic support.
+     * Fixes and improvements to other atomic operations:
+        * Modified HIP and CUDA generic atomic compare and swap algorithms
+          to use atomic loads instead of relying on volatile.
+        * Re-implemented atomic loads in terms of builtin atomics for CUDA
+          and HIP so that the generic compare and swap functions can use it.
+        * Removes volatile qualifier in atomic function signatures.
+        * Use cuda::atomic_ref in newer versions of CUDA to back 
+          atomicLoad/atomicStore.
+        * Use atomicAdd as a fallback for atomicSub in CUDA.
+        * Removed checks where __CUDA_ARCH__ is less than 350 since RAJA 
+          requires that as the minimum supported architecture (CMake check).
+     * Fixed issues with naming RAJA forall::kernels when using CUDA.
+     * Fixes in SYCL back-end for RAJA::launch.
+     * Fixed some issues in examples.
+     * Bugfixes and cleanup in parts of the SYCL back-end needed to
+       support a bunch of new SYCL kernels that will appear in 
+       RAJA Performance Suite release.
+     * Fix type naming issue that was exposed with a new version of the
+       Intel oneAPI compiler.
+     * Fix issue in User Guide documentation for configuring a project
+       using RAJA CMake configuration.
+
+
 Version 2024.02.2 -- Release date 2024-05-08
 ============================================
 

docs/conf.py

@@ -86,9 +86,9 @@
 # built documents.
 #
 # The short X.Y version.
-version = u'2024.02'
+version = u'2024.07'
 # The full version, including alpha/beta/rc tags.
-release = u'2024.02.2'
+release = u'2024.07.0'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.