Merge remote-tracking branch 'origin/master' into print-config-env

eth-cscs · Oct 28, 2024 · ca4b097 · ca4b097
2 parents 8b97538 + 9fa2db4
commit ca4b097
Show file tree

Hide file tree

Showing 50 changed files with 439 additions and 444 deletions.
diff --git a/ci/common-ci.yml b/ci/common-ci.yml
@@ -24,22 +24,20 @@ stages:
     - TAG_REPO=`find $SPACK_DLAF_REPO -type f -exec sha256sum {} \; | sha256sum - | head -c 16`
     - TAG_ENVIRONMENT=`cat $SPACK_ENVIRONMENT $COMMON_SPACK_ENVIRONMENT | sha256sum | head -c 16`
     - TAG=${TAG_IMAGE}-${TAG_APTGET}-${TAG_COMPILER}-MKL${USE_MKL}-${TAG_DOCKERFILE}-${TAG_SPACK}-${TAG_REPO}-${TAG_ENVIRONMENT}
-    - export PERSIST_IMAGE_NAME=$BUILD_IMAGE:$TAG
-    - echo "BUILD_IMAGE=$PERSIST_IMAGE_NAME" > build.env
+    - export PERSIST_IMAGE_NAME=$DEPS_IMAGE:$TAG
+    - echo "DEPS_IMAGE=$PERSIST_IMAGE_NAME" > build.env
     - echo "USE_MKL=$USE_MKL" >> build.env
-    - echo "USE_ROCBLAS=$USE_ROCBLAS" >> build.env
     - echo "USE_CODECOV=$USE_CODECOV" >> build.env
     - 'echo "INFO: Building image $PERSIST_IMAGE_NAME"'
     - 'echo "INFO: Using NUM_CORES_BUILD_DEPS=$NUM_CORES_BUILD_DEPS"'
   artifacts:
     reports:
       dotenv: build.env
   variables:
-    SPACK_SHA: 0905edf592752742eb4ddab3a528d3aee8f92930
+    SPACK_SHA: develop-2024-10-06
     SPACK_DLAF_REPO: ./spack
     DOCKER_BUILD_ARGS: '[
         "BASE_IMAGE",
-        "BUILDKIT_INLINE_CACHE=1",
         "SPACK_SHA",
         "EXTRA_APTGET",
         "COMPILER",
@@ -57,7 +55,6 @@ stages:
     EXTRA_APTGET: ""
     CXXSTD: 17
     USE_MKL: "OFF"
-    USE_ROCBLAS: "OFF"
     COMMON_SPACK_ENVIRONMENT: ci/docker/common.yaml
     USE_CODECOV: "false"
 
@@ -69,23 +66,17 @@ stages:
     - 'echo $DOCKERHUB_TOKEN | podman login docker.io -u $DOCKERHUB_USERNAME --password-stdin'
     - 'echo "INFO: Using NUM_CORES_BUILD_DLAF=$NUM_CORES_BUILD_DLAF"'
   after_script:
-    - podman run -v $PWD/ci/ctest_to_gitlab.sh:/ctest_to_gitlab.sh $DEPLOY_IMAGE /ctest_to_gitlab.sh "$DEPLOY_IMAGE" "$USE_CODECOV" "$THREADS_PER_NODE" "$SLURM_CONSTRAINT" > pipeline.yml
+    - podman run -v $PWD/ci/ctest_to_gitlab.sh:/ctest_to_gitlab.sh $DLAF_IMAGE /ctest_to_gitlab.sh "$DLAF_IMAGE" "$USE_CODECOV" "$THREADS_PER_NODE" "$SLURM_CONSTRAINT" > pipeline.yml
   variables:
-    PERSIST_IMAGE_NAME: $DEPLOY_IMAGE
+    PERSIST_IMAGE_NAME: $DLAF_IMAGE
     DOCKER_BUILD_ARGS: '[
-        "BUILD_IMAGE",
-        "DEPLOY_BASE_IMAGE",
-        "EXTRA_APTGET_DEPLOY",
+        "DEPS_IMAGE",
         "PIP_OPTS",
-        "USE_MKL",
-        "USE_ROCBLAS",
         "NUM_PROCS=$NUM_CORES_BUILD_DLAF"
       ]'
     # default configuration variables
     # can be overwritten in the configuration as needed
     DOCKERFILE: ci/docker/deploy.Dockerfile
-    DEPLOY_BASE_IMAGE: docker.io/ubuntu:24.04
-    EXTRA_APTGET_DEPLOY: ""
     PIP_OPTS: ""
   artifacts:
     paths:

diff --git a/ci/cpu/asan_ubsan_lsan.yml b/ci/cpu/asan_ubsan_lsan.yml
@@ -6,9 +6,9 @@ cpu asan ubsan lsan deps:
   variables:
     EXTRA_APTGET: "clang-18 libclang-rt-18-dev libomp-18-dev"
     COMPILER: clang@18
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/asan-ubsan-lsan.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-asan-ubsan-lsan/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-asan-ubsan-lsan/deps
 
 cpu asan ubsan lsan build:
   extends:
@@ -17,9 +17,7 @@ cpu asan ubsan lsan build:
   needs:
     - cpu asan ubsan lsan deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-asan-ubsan-lsan/deploy:$CI_COMMIT_SHA
-    # For symbolizing stacktraces with llvm-symbolizer
-    EXTRA_APTGET_DEPLOY: "llvm-18"
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-asan-ubsan-lsan/dlaf:$CI_COMMIT_SHA
 
 cpu asan ubsan lsan test:
   extends: .run_common

diff --git a/ci/cpu/clang15_release.yml b/ci/cpu/clang15_release.yml
@@ -6,9 +6,9 @@ cpu clang15 release deps:
   variables:
     EXTRA_APTGET: "clang-15"
     COMPILER: clang@15
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu-serial.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-release/deps
 
 cpu clang15 release build:
   extends:
@@ -17,7 +17,7 @@ cpu clang15 release build:
   needs:
     - cpu clang15 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-release/dlaf:$CI_COMMIT_SHA
 
 cpu clang15 release test:
   extends: .run_common

diff --git a/ci/cpu/clang15_release_cxx20.yml b/ci/cpu/clang15_release_cxx20.yml
@@ -9,7 +9,7 @@ cpu clang15 cxx20 release deps:
     CXXSTD: 20
     SPACK_ENVIRONMENT: ci/docker/release-cpu.yaml
     USE_MKL: "ON"
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-20-release/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-20-release/deps
 
 cpu clang15 cxx20 release build:
   extends:
@@ -18,7 +18,7 @@ cpu clang15 cxx20 release build:
   needs:
     - cpu clang15 cxx20 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-cxx20-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-cxx20-release/dlaf:$CI_COMMIT_SHA
 
 cpu clang15 cxx20 release test:
   extends: .run_common

diff --git a/ci/cpu/clang15_release_stdexec.yml b/ci/cpu/clang15_release_stdexec.yml
@@ -7,9 +7,9 @@ cpu clang15 stdexec release deps:
     EXTRA_APTGET: "clang-15 libomp-15-dev"
     COMPILER: clang@15
     CXXSTD: 20
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu-stdexec.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-stdexec-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-stdexec-release/deps
 
 cpu clang15 stdexec release build:
   extends:
@@ -18,7 +18,7 @@ cpu clang15 stdexec release build:
   needs:
     - cpu clang15 stdexec release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-stdexec-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang15-stdexec-release/dlaf:$CI_COMMIT_SHA
 
 cpu clang15 stdexec release test:
   extends: .run_common

diff --git a/ci/cpu/clang16_release.yml b/ci/cpu/clang16_release.yml
@@ -6,9 +6,9 @@ cpu clang16 release deps:
   variables:
     EXTRA_APTGET: "clang-16 libomp-16-dev"
     COMPILER: clang@16
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang16-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang16-release/deps
 
 cpu clang16 release build:
   extends:
@@ -17,7 +17,7 @@ cpu clang16 release build:
   needs:
     - cpu clang16 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang16-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang16-release/dlaf:$CI_COMMIT_SHA
 
 cpu clang16 release test:
   extends: .run_common

diff --git a/ci/cpu/clang18_release.yml b/ci/cpu/clang18_release.yml
@@ -6,9 +6,9 @@ cpu clang18 release deps:
   variables:
     EXTRA_APTGET: "clang-18 libomp-18-dev"
     COMPILER: clang@18
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang18-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang18-release/deps
 
 cpu clang18 release build:
   extends:
@@ -17,7 +17,7 @@ cpu clang18 release build:
   needs:
     - cpu clang18 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang18-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-clang18-release/dlaf:$CI_COMMIT_SHA
 
 cpu clang18 release test:
   extends: .run_common

diff --git a/ci/cpu/gcc11_debug_stdexec.yml b/ci/cpu/gcc11_debug_stdexec.yml
@@ -9,7 +9,7 @@ cpu gcc11 stdexec debug deps:
     CXXSTD: 20
     SPACK_ENVIRONMENT: ci/docker/debug-cpu-stdexec.yaml
     USE_MKL: "ON"
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-debug/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-debug/deps
 
 cpu gcc11 stdexec debug build:
   extends:
@@ -18,7 +18,7 @@ cpu gcc11 stdexec debug build:
   needs:
     - cpu gcc11 stdexec debug deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-debug/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-debug/dlaf:$CI_COMMIT_SHA
 
 cpu gcc11 stdexec debug test:
   extends: .run_common

diff --git a/ci/cpu/gcc11_release_stdexec.yml b/ci/cpu/gcc11_release_stdexec.yml
@@ -7,9 +7,9 @@ cpu gcc11 stdexec release deps:
     EXTRA_APTGET: "gcc-11 g++-11 gfortran-11"
     COMPILER: gcc@11
     CXXSTD: 20
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu-stdexec.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-release/deps
 
 cpu gcc11 stdexec release build:
   extends:
@@ -18,7 +18,7 @@ cpu gcc11 stdexec release build:
   needs:
     - cpu gcc11 stdexec release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc11-stdexec-release/dlaf:$CI_COMMIT_SHA
 
 cpu gcc11 stdexec release test:
   extends: .run_common

diff --git a/ci/cpu/gcc12_release_cxx20.yml b/ci/cpu/gcc12_release_cxx20.yml
@@ -9,7 +9,7 @@ cpu gcc12 cxx20 release deps:
     CXXSTD: 20
     SPACK_ENVIRONMENT: ci/docker/release-cpu.yaml
     USE_MKL: "ON"
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc12-release/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc12-release/deps
 
 cpu gcc12 cxx20 release build:
   extends:
@@ -18,7 +18,7 @@ cpu gcc12 cxx20 release build:
   needs:
     - cpu gcc12 cxx20 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc12-cxx20-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc12-cxx20-release/dlaf:$CI_COMMIT_SHA
 
 cpu gcc12 cxx20 release test:
   extends: .run_common

diff --git a/ci/cpu/gcc13_codecov.yml b/ci/cpu/gcc13_codecov.yml
@@ -6,7 +6,7 @@ cpu gcc13 codecov deps:
   variables:
     COMPILER: gcc@13
     SPACK_ENVIRONMENT: ci/docker/debug-cpu.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-codecov/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-codecov/deps
     USE_CODECOV: "true"
 
 cpu gcc13 codecov build:
@@ -17,7 +17,7 @@ cpu gcc13 codecov build:
     - cpu gcc13 codecov deps
   variables:
     DOCKERFILE: ci/docker/codecov.Dockerfile
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-codecov/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-codecov/dlaf:$CI_COMMIT_SHA
     PIP_OPTS: "--break-system-packages"
 
 cpu gcc13 codecov test:

diff --git a/ci/cpu/gcc13_release.yml b/ci/cpu/gcc13_release.yml
@@ -5,9 +5,9 @@ cpu gcc13 release deps:
   extends: .build_deps_common
   variables:
     COMPILER: gcc@13
-    USE_MKL: "ON"
     SPACK_ENVIRONMENT: ci/docker/release-cpu.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-release/build
+    USE_MKL: "ON"
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-release/deps
 
 cpu gcc13 release build:
   extends:
@@ -16,7 +16,7 @@ cpu gcc13 release build:
   needs:
     - cpu gcc13 release deps
   variables:
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cpu-gcc13-release/dlaf:$CI_COMMIT_SHA
 
 cpu gcc13 release test:
   extends: .run_common

diff --git a/ci/ctest_to_gitlab.sh b/ci/ctest_to_gitlab.sh
@@ -16,6 +16,7 @@ THREADS_PER_NODE="$3"
 SLURM_CONSTRAINT="$4"
 
 if [ "$USE_CODECOV" = true ]; then
+# CRAY_CUDA_MPS set to 0 to avoid test hanging on daint (See PR #1197)
 BASE_TEMPLATE="
 include:
   - remote: 'https://gitlab.com/cscs-ci/recipes/-/raw/master/templates/v2/.cscs.yml'
@@ -30,7 +31,7 @@ variables:
   SLURM_EXCLUSIVE: ''
   SLURM_EXACT: ''
   SLURM_CONSTRAINT: $SLURM_CONSTRAINT
-  CRAY_CUDA_MPS: 1
+  CRAY_CUDA_MPS: 0
   MPICH_MAX_THREAD_SAFETY: multiple
 
 {{JOBS}}
@@ -65,6 +66,7 @@ JOB_TEMPLATE="
     paths:
       - codecov-reports/"
 else
+# CRAY_CUDA_MPS set to 0 to avoid test hanging on daint (See PR #1197)
 BASE_TEMPLATE="
 include:
   - remote: 'https://gitlab.com/cscs-ci/recipes/-/raw/master/templates/v2/.cscs.yml'
@@ -78,7 +80,7 @@ variables:
   SLURM_EXCLUSIVE: ''
   SLURM_EXACT: ''
   SLURM_CONSTRAINT: $SLURM_CONSTRAINT
-  CRAY_CUDA_MPS: 1
+  CRAY_CUDA_MPS: 0
   MPICH_MAX_THREAD_SAFETY: multiple
 
 {{JOBS}}

diff --git a/ci/cuda/gcc11_codecov.yml b/ci/cuda/gcc11_codecov.yml
@@ -7,7 +7,7 @@ cuda gcc11 codecov deps:
     BASE_IMAGE: docker.io/nvidia/cuda:11.7.1-devel-ubuntu22.04
     COMPILER: gcc@11
     SPACK_ENVIRONMENT: ci/docker/debug-cuda.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-codecov/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-codecov/deps
     USE_CODECOV: "true"
 
 cuda gcc11 codecov build:
@@ -18,8 +18,7 @@ cuda gcc11 codecov build:
     - cuda gcc11 codecov deps
   variables:
     DOCKERFILE: ci/docker/codecov.Dockerfile
-    DEPLOY_BASE_IMAGE: docker.io/ubuntu:22.04
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-codecov/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-codecov/dlaf:$CI_COMMIT_SHA
 
 cuda gcc11 codecov test:
   extends: .run_common

diff --git a/ci/cuda/gcc11_debug_scalapack.yml b/ci/cuda/gcc11_debug_scalapack.yml
@@ -7,7 +7,7 @@ cuda gcc11 debug scalapack deps:
     BASE_IMAGE: docker.io/nvidia/cuda:11.7.1-devel-ubuntu22.04
     COMPILER: gcc@11
     SPACK_ENVIRONMENT: ci/docker/debug-cuda-scalapack.yaml
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-debug/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-debug/deps
 
 cuda gcc11 debug scalapack build:
   extends:
@@ -16,8 +16,7 @@ cuda gcc11 debug scalapack build:
   needs:
     - cuda gcc11 debug scalapack deps
   variables:
-    DEPLOY_BASE_IMAGE: docker.io/ubuntu:22.04
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-debug/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-debug/dlaf:$CI_COMMIT_SHA
 
 cuda gcc11 debug scalapack test:
   extends: .run_common

diff --git a/ci/cuda/gcc11_release.yml b/ci/cuda/gcc11_release.yml
@@ -6,10 +6,9 @@ cuda gcc11 release deps:
   variables:
     BASE_IMAGE: docker.io/nvidia/cuda:11.7.1-devel-ubuntu22.04
     COMPILER: gcc@11
-    CXXSTD: 17
     SPACK_ENVIRONMENT: ci/docker/release-cuda.yaml
     USE_MKL: "ON"
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-release/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-release/deps
 
 cuda gcc11 release build:
   extends:
@@ -18,8 +17,7 @@ cuda gcc11 release build:
   needs:
     - cuda gcc11 release deps
   variables:
-    DEPLOY_BASE_IMAGE: docker.io/ubuntu:22.04
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-release/dlaf:$CI_COMMIT_SHA
 
 cuda gcc11 release test:
   extends: .run_common

diff --git a/ci/cuda/gcc11_release_scalapack.yml b/ci/cuda/gcc11_release_scalapack.yml
@@ -8,7 +8,7 @@ cuda gcc11 release scalapack deps:
     COMPILER: gcc@11
     SPACK_ENVIRONMENT: ci/docker/release-cuda-scalapack.yaml
     USE_MKL: "ON"
-    BUILD_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-release/build
+    DEPS_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-release/deps
 
 cuda gcc11 release scalapack build:
   extends:
@@ -17,8 +17,7 @@ cuda gcc11 release scalapack build:
   needs:
     - cuda gcc11 release scalapack deps
   variables:
-    DEPLOY_BASE_IMAGE: docker.io/ubuntu:22.04
-    DEPLOY_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-release/deploy:$CI_COMMIT_SHA
+    DLAF_IMAGE: $CSCS_REGISTRY_PATH/cuda-gcc11-scalapack-release/dlaf:$CI_COMMIT_SHA
 
 cuda gcc11 release scalapack test:
   extends: .run_common