Merge remote-tracking branch 'origin/develop' into feature/burmark1/g…

…pu_stream

.gitignore

@@ -1,5 +1,16 @@
+.cache
+.idea
+*.sync-conflict-*
 *.pyc
 *.o
+*.a
 *.exe
-build*/
-install*/
+*.gch
+/*.sublime-*
+/build/
+/build_*/
+/build-*/
+/install/
+/install_*/
+/install-*/
+/Debug/

.gitlab-ci.yml

@@ -24,6 +24,8 @@
 
 # We define the following GitLab pipeline variables:
 variables:
+  GITHUB_PROJECT_NAME: "RAJAPerf"
+  GITHUB_PROJECT_ORG: "LLNL"
 # Use the umdev service user to run CI. This prevents from running pipelines as
 # an actual user.
   LLNL_SERVICE_USER: umdev
@@ -35,6 +37,11 @@ variables:
 # We build the projects in the CI clone directory.
 # TODO: add a clean-up mechanism
   BUILD_ROOT: ${CI_PROJECT_DIR}
+# Set the build-and-test command.
+  BUILD_AND_TEST_CMD: "./scripts/gitlab/build_and_test.sh"
+# Override the list of branch that will skip the "draft PR test".
+# Add protected branches here. Defaults to "develop main master".
+#  ALWAYS_RUN_LIST: "develop main"
 
 # We organize the CI on Gitlab in sub-pipelines. Each sub-pipeline corresponds
 # to a test phase on a given machine.
@@ -50,13 +57,17 @@ stages:
     include:
       - local: '.gitlab/custom-jobs-and-variables.yml'
       - project: 'radiuss/radiuss-shared-ci'
-        ref: v2022.09.0
+        ref: v2023.03.0rc
         file: '${CI_MACHINE}-build-and-test.yml'
       - local: '.gitlab/${CI_MACHINE}-build-and-test-extra.yml'
     strategy: depend
     forward:
       pipeline_variables: true
 
-# pipelines subscribed by the project
 include:
-  - local: .gitlab/subscribed-pipelines.yml
+  # checks preliminary to running the actual CI test (optional)
+  #- project: 'radiuss/radiuss-shared-ci'
+  #  ref: v2023.03.0rc
+  #  file: 'preliminary-ignore-draft-pr.yml'
+  # pipelines subscribed by the project
+  - local: '.gitlab/subscribed-pipelines.yml'

.gitlab/corona-build-and-test-extra.yml

@@ -27,7 +27,7 @@ variables:
   INCLUDED_FILE_CANNOT_BE_EMPTY: "True"
 
 # INFO: This job is activated in RAJA CI, but we don't use desul atomics here
-#rocm_5_1_1_clang_13_0_0_desul_atomics:
+#rocmcc_5_1_1_hip_desul_atomics:
 #  variables:
-#    SPEC: "+rocm~openmp +desul amdgpu_target=gfx906 %clang@13.0.0 ^blt@develop ^hip@5.1.1"
+#    SPEC: "+rocm~openmp +desul amdgpu_target=gfx906 %rocmcc@5.1.1 ^blt@develop ^hip@5.1.1"
 #  extends: .build_and_test_on_corona

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

@@ -26,14 +26,24 @@ variables:
 
 # Corona
 # Arguments for top level allocation
-  CORONA_BUILD_AND_TEST_SHARED_ALLOC: "--time-limit=60m --nodes=1"
+  CORONA_BUILD_AND_TEST_SHARED_ALLOC: "--exclusive --time-limit=60m --nodes=1"
 # Arguments for job level allocation
   CORONA_BUILD_AND_TEST_JOB_ALLOC: "--time-limit=30m --nodes=1"
 # Project specific variants for corona
   PROJECT_CORONA_VARIANTS: "~openmp "
 # Project specific deps for corona
   PROJECT_CORONA_DEPS: "^blt@develop "
 
+# Tioga
+# Arguments for top level allocation
+  TIOGA_BUILD_AND_TEST_SHARED_ALLOC: "--exclusive --time-limit=60m --nodes=1"
+# Arguments for job level allocation
+  TIOGA_BUILD_AND_TEST_JOB_ALLOC: "--time-limit=45m --nodes=1"
+# Project specific variants for corona
+  PROJECT_TIOGA_VARIANTS: "~openmp"
+# Project specific deps for corona
+  PROJECT_TIOGA_DEPS: "^blt@develop "
+
 # Lassen and Butte use a different job scheduler (spectrum lsf) that does not
 # allow pre-allocation the same way slurm does.
 # Arguments for job level allocation

.gitlab/lassen-build-and-test-extra.yml

@@ -27,6 +27,14 @@ ibm_clang_9_0_0_gcc_8_3_1:
   extends: .build_and_test_on_lassen
   allow_failure: true
 
+# Overriding shared spec: Longer allocation + Allow failures
+ibm_clang_9_0_0_gcc_8_3_1_cuda_10_1_168:
+  variables:
+    SPEC: " ${PROJECT_LASSEN_VARIANTS} +cuda %clang@ibm.9.0.0 cxxflags=\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1\" cflags=\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1\" ^cuda@10.1.168 ${PROJECT_LASSEN_DEPS}"
+    LASSEN_BUILD_AND_TEST_JOB_ALLOC: "1 -W 60"
+  extends: .build_and_test_on_lassen
+  allow_failure: true
+
 # Overriding shared spec: Extra flags
 gcc_8_3_1:
   variables:
@@ -55,28 +63,32 @@ xl_16_1_1_12_gcc_8_3_1:
     LASSEN_BUILD_AND_TEST_JOB_ALLOC: "1 -W 50"
   extends: .build_and_test_on_lassen
 
-# Overriding shared spec: Longer allocation + Allow failures
-ibm_clang_9_0_0_gcc_8_3_1_cuda_10_1_168:
+# Overriding shared spec: Longer allocation + Extra flags
+xl_16_1_1_12_gcc_7_3_1_cuda_10_1_168:
   variables:
-    SPEC: " ${PROJECT_LASSEN_VARIANTS} +cuda %clang@ibm.9.0.0 cxxflags=\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1\" cflags=\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1\" ^cuda@10.1.168 ${PROJECT_LASSEN_DEPS}"
+    SPEC: "${PROJECT_LASSEN_VARIANTS} +cuda %xl@16.1.1.12 cxxflags==\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-7.3.1 -qthreaded -std=c++14 -O3 -qstrict -qxlcompatmacros -qlanglvl=extended0x -qalias=noansi -qhot -qsmp=omp -qnoeh -qsuppress=1500-029 -qsuppress=1500-036\" cflags==\"--gcc-toolchain=/usr/tce/packages/gcc/gcc-7.3.1\" cuda_arch=70 ^cuda@10.1.168 ${PROJECT_LASSEN_DEPS}"
     LASSEN_BUILD_AND_TEST_JOB_ALLOC: "1 -W 60"
   extends: .build_and_test_on_lassen
+  script:
+    - |
+      echo -e "\e[31mDeactivated spec !\e[0m"
+      echo -e "\e[31m${SPEC}\e[0m"
+      echo -e "\e[31mRAJA won’t build with Cuda < 11.1.0 due to a known issue.\e[0m"
+    - exit 1
   allow_failure: true
 
 # Overriding shared spec: Extra flags + Longer allocation + Allow failure
 xl_16_1_1_12_cuda_11_1_0:
   variables:
     SPEC: " ${PROJECT_LASSEN_VARIANTS} +cuda %xl@16.1.1.12 cxxflags=\"-qthreaded -std=c++14 -O3 -qstrict -qxlcompatmacros -qlanglvl=extended0x -qalias=noansi -qhot -qpic -qsmp=omp -qsuppress=1500-029 -qsuppress=1500-036\" cuda_arch=70 ^cuda@11.1.0 ${PROJECT_LASSEN_DEPS}"
     LASSEN_BUILD_AND_TEST_JOB_ALLOC: "1 -W 60"
-  allow_failure: true
   extends: .build_and_test_on_lassen
 
 # Overriding shared spec: Extra flags + Longer allocation + Allow failure
 xl_16_1_1_12_gcc_8_3_1_cuda_11_1_0:
   variables:
     SPEC: " ${PROJECT_LASSEN_VARIANTS} +cuda %xl@16.1.1.12 cxxflags\"=--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1 -qthreaded -std=c++14 -O3 -qstrict -qxlcompatmacros -qlanglvl=extended0x -qalias=noansi -qhot -qpic -qsmp=omp -qsuppress=1500-029 -qsuppress=1500-036\" cflags=--gcc-toolchain=/usr/tce/packages/gcc/gcc-8.3.1 cuda_arch=70 ^cuda@11.1.0 ${PROJECT_LASSEN_DEPS}"
     LASSEN_BUILD_AND_TEST_JOB_ALLOC: "1 -W 60"
-  allow_failure: true
   extends: .build_and_test_on_lassen
 
 

.gitlab/subscribed-pipelines.yml

@@ -19,6 +19,11 @@ corona-build-and-test:
     CI_MACHINE: "corona"
   extends: [.build-and-test]
 
+tioga-build-and-test:
+  variables:
+    CI_MACHINE: "tioga"
+  extends: [.build-and-test]
+
 lassen-build-and-test:
   variables:
     CI_MACHINE: "lassen"

.gitlab/tioga-build-and-test-extra.yml

@@ -0,0 +1,39 @@
+###############################################################################
+# Copyright (c) 2017-23, Lawrence Livermore National Security, LLC
+# and RAJA Performance Suite project contributors.
+# See the RAJAPerf/LICENSE file for details.
+#
+# SPDX-License-Identifier: (BSD-3-Clause)
+#############################################################################
+
+########################
+# Overridden shared jobs
+########################
+# We duplicate the shared jobs description and add necessary changes for RAJA.
+# We keep ${PROJECT_<MACHINE>_VARIANTS} and ${PROJECT_<MACHINE>_DEPS} So that
+# the comparison with the original job is easier.
+
+# No overridden jobs so far.
+
+############
+# Extra jobs
+############
+# We do not recommend using ${PROJECT_<MACHINE>_VARIANTS} and
+# ${PROJECT_<MACHINE>_DEPS} in the extra jobs. There is no reason not to fully
+# describe the spec here.
+
+# With GitLab CI, included files cannot be empty.
+variables:
+  INCLUDED_FILE_CANNOT_BE_EMPTY: "True"
+
+# INFO: This job is activated in RAJA CI, but we don't use desul atomics here
+#rocmcc_5_2_3_hip_desul_atomics:
+#  variables:
+#    SPEC: "~shared +rocm ~openmp +desul +tests amdgpu_target=gfx90a %rocmcc@5.2.3 ^hip@5.2.3 ^blt@develop"
+#  extends: .build_and_test_on_tioga
+#
+# INFO: This job is activated in RAJA CI, but we may not want to run openmp here
+#rocmcc_5_2_3_hip_openmp:
+#  variables:
+#    SPEC: "~shared +rocm +openmp +tests amdgpu_target=gfx90a %rocmcc@5.2.3 ^hip@5.2.3 ^blt@develop"
+#  extends: .build_and_test_on_tioga

.readthedocs.yml

@@ -0,0 +1,39 @@
+# .readthedocs.yml
+# Read the Docs configuration file
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+# Required
+version: 2
+
+# Set the version of Python and other tools you might need
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.11"
+
+##
+## Code for debugging directory structure in readthedocs 
+## to fix broken links, etc.
+##
+# Print tree
+#build:
+#  os: ubuntu-20.04
+#  apt_packages:
+#    - tree
+#  tools:
+#    python: "3.7"
+#  jobs:
+#    post_build:
+#      - tree -J
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/conf.py
+
+# Optionally build docs in add'l formats such as PDF and ePub
+#formats: all
+
+# Set requirements needed to build the docs
+python:
+  install:
+    - requirements: docs/requirements.txt

.uberenv_config.json

@@ -7,7 +7,7 @@
 "spack_branch": "v0.19.0",
 "spack_activate" : {},
 "spack_configs_path": "tpl/RAJA/scripts/radiuss-spack-configs",
-"spack_packages_path": "scripts/spack_packages",
+"spack_packages_path": "tpl/RAJA/scripts/radiuss-spack-configs/packages",
 "spack_concretizer": "clingo",
 "spack_setup_clingo": false
 }

CMakeLists.txt

@@ -200,3 +200,7 @@ add_subdirectory(src)
 if (RAJA_PERFSUITE_ENABLE_TESTS)
   add_subdirectory(test)
 endif()
+
+if (RAJA_PERFSUITE_ENABLE_DOCUMENTATION)
+  add_subdirectory(docs)
+endif ()