Merge remote-tracking branch 'scream/master' into jgfouca/final_screa…

…m_downstream_2024_11_21 * scream/master: (145 commits) Mergify: dismiss reviews when new commits are pushed EAMxx: Add aerosols and gases outputs to ensure the microscopy baseline test is sensitive to changes in the interface. fix standard names for low/med/hgh cloud area fraction types Add standard name metadata to most EAMxx output variables EAMxx: Update mam4xx submodule to fix CUDA test. EAMxx: Modify microphysics interface to include sethet (washout rates) computations. Bump DavidAnson/markdownlint-cli2-action from 17 to 18 EAMxx: Moves fences into the interface and remove extra comments EAMxx: Adds missing namelist entries for multi process test EAMxx: Remove all diffs from microphysics cpp file ff mam4xx to current main Revert "EAMxx: Clang format" EAMxx:Inits constituent fluxes to zero fix cuda compile warnings EAMxx: Moves online emiss and constituent init to a new function hpp file EAMxx: Fixed some comments paths and other minor cleanup EAMxx: Clang format EAMxx: Fixes a gpu bug requiring a particular format for dstflx update mam4xx submodule to fix gpu compilation error EAMxx:Removes accidently added file ...
E3SM-Project · Nov 21, 2024 · 28474c3 · 28474c3
2 parents de2142e + 767c2a0
commit 28474c3
Show file tree

Hide file tree

Showing 124 changed files with 4,010 additions and 3,452 deletions.
diff --git a/.github/workflows/e3sm-gh-md-linter.yml b/.github/workflows/e3sm-gh-md-linter.yml
@@ -27,7 +27,7 @@ jobs:
       with:
         files: '**/*.md'
         separator: ","
-    - uses: DavidAnson/markdownlint-cli2-action@v17
+    - uses: DavidAnson/markdownlint-cli2-action@v18
       if: steps.changed-files.outputs.any_changed == 'true'
       with:
         config: 'docs/.markdownlint.json'

diff --git a/.github/workflows/eamxx-sa-coverage.yml b/.github/workflows/eamxx-sa-coverage.yml
@@ -2,6 +2,11 @@ name: eamxx-sa-coverage
 
 on:
   workflow_dispatch:
+    inputs:
+      submit:
+        description: 'Force cdash submission'
+        required: true
+        type: boolean
 
   # Add schedule trigger for nightly runs at midnight MT (Standard Time)
   schedule:
@@ -13,7 +18,8 @@ concurrency:
   cancel-in-progress: true
 
 env:
-  submit: ${{ github.event_name == 'schedule' && 'true' || 'false' }}  # Submit to cdash only for nightlies
+  # Submit to cdash only for nightlies or if the user explicitly forced a submission via workflow dispatch
+  submit: ${{ github.event_name == 'schedule' || (github.event_name == 'workflow_dispatch' && inputs.submit) }}
 
 jobs:
   gcc-openmp:
@@ -48,11 +54,40 @@ jobs:
           submodules: recursive
       - name: Show action trigger
         uses: ./.github/actions/show-workflow-trigger
+      - name: Get CUDA Arch
+        run: |
+          # Ensure nvidia-smi is available
+          if ! command -v nvidia-smi &> /dev/null; then
+              echo "nvidia-smi could not be found. Please ensure you have Nvidia drivers installed."
+              exit 1
+          fi
+
+          # Get the GPU model from nvidia-smi, and set env for next step
+          gpu_model=$(nvidia-smi --query-gpu=name --format=csv,noheader | head -n 1)
+          case "$gpu_model" in
+              *"H100"*)
+                  echo "Hopper=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=90" >> $GITHUB_ENV
+                  ARCH=90
+                  ;;
+              *"A100"*)
+                  echo "Ampere=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=80" >> $GITHUB_ENV
+                  ;;
+              *"V100"*)
+                  echo "Volta=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=70" >> $GITHUB_ENV
+                  ;;
+              *)
+                  echo "Unsupported GPU model: $gpu_model"
+                  exit 1
+                  ;;
+          esac
       - name: Run tests
         uses: ./.github/actions/test-all-scream
         with:
           build_type: cov
           machine: ghci-snl-cuda
           generate: false
           submit: ${{ env.submit }}
-          cmake-configs: Kokkos_ARCH_VOLTA70=ON;CMAKE_CUDA_ARCHITECTURES=70
+          cmake-configs: Kokkos_ARCH_HOPPER90=${{ env.Hopper }};Kokkos_ARCH_AMPERE80=${{ env.Ampere }};Kokkos_ARCH_VOLTA70=${{ env.Volta }};CMAKE_CUDA_ARCHITECTURES=${{ env.CUDA_ARCH }}
diff --git a/.github/workflows/eamxx-sa-sanitizer.yml b/.github/workflows/eamxx-sa-sanitizer.yml
@@ -2,6 +2,11 @@ name: eamxx-sa-sanitizer
 
 on:
   workflow_dispatch:
+    inputs:
+      submit:
+        description: 'Force cdash submission'
+        required: true
+        type: boolean
 
   # Add schedule trigger for nightly runs at midnight MT (Standard Time)
   schedule:
@@ -13,12 +18,13 @@ concurrency:
   cancel-in-progress: true
 
 env:
-  submit: ${{ github.event_name == 'schedule' && 'true' || 'false' }}  # Submit to cdash only for nightlies
+  # Submit to cdash only for nightlies or if the user explicitly forced a submission via workflow dispatch
+  submit: ${{ github.event_name == 'schedule' || (github.event_name == 'workflow_dispatch' && inputs.submit) }}
 
 jobs:
   gcc-openmp:
     runs-on:  [self-hosted, ghci-snl-cpu, gcc]
-    name: gcc-openmp / cov
+    name: gcc-openmp / valg
     steps:
       - name: Check out the repository
         uses: actions/checkout@v4
@@ -52,11 +58,40 @@ jobs:
           submodules: recursive
       - name: Show action trigger
         uses: ./.github/actions/show-workflow-trigger
+      - name: Get CUDA Arch
+        run: |
+          # Ensure nvidia-smi is available
+          if ! command -v nvidia-smi &> /dev/null; then
+              echo "nvidia-smi could not be found. Please ensure you have Nvidia drivers installed."
+              exit 1
+          fi
+
+          # Get the GPU model from nvidia-smi, and set env for next step
+          gpu_model=$(nvidia-smi --query-gpu=name --format=csv,noheader | head -n 1)
+          case "$gpu_model" in
+              *"H100"*)
+                  echo "Hopper=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=90" >> $GITHUB_ENV
+                  ARCH=90
+                  ;;
+              *"A100"*)
+                  echo "Ampere=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=80" >> $GITHUB_ENV
+                  ;;
+              *"V100"*)
+                  echo "Volta=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=70" >> $GITHUB_ENV
+                  ;;
+              *)
+                  echo "Unsupported GPU model: $gpu_model"
+                  exit 1
+                  ;;
+          esac
       - name: Run tests
         uses: ./.github/actions/test-all-scream
         with:
           build_type: ${{ matrix.build_type }}
           machine: ghci-snl-cuda
           generate: false
           submit: ${{ env.submit }}
-          cmake-configs: Kokkos_ARCH_VOLTA70=ON;CMAKE_CUDA_ARCHITECTURES=70
+          cmake-configs: Kokkos_ARCH_HOPPER90=${{ env.Hopper }};Kokkos_ARCH_AMPERE80=${{ env.Ampere }};Kokkos_ARCH_VOLTA70=${{ env.Volta }};CMAKE_CUDA_ARCHITECTURES=${{ env.CUDA_ARCH }}
diff --git a/.github/workflows/eamxx-sa-testing.yml b/.github/workflows/eamxx-sa-testing.yml
@@ -21,6 +21,10 @@ on:
         description: 'Generate baselines'
         required: true
         type: boolean
+      submit:
+        description: 'Force cdash submission'
+        required: true
+        type: boolean
 
   # Add schedule trigger for nightly runs at midnight MT (Standard Time)
   schedule:
@@ -34,7 +38,8 @@ concurrency:
   cancel-in-progress: true
 
 env:
-  submit: ${{ github.event_name == 'schedule' && 'true' || 'false' }}  # Submit to cdash only for nightlies
+  # Submit to cdash only for nightlies or if the user explicitly forced a submission via workflow dispatch
+  submit: ${{ github.event_name == 'schedule' || (github.event_name == 'workflow_dispatch' && inputs.submit) }}
 
 jobs:
   pre_process_pr:
@@ -44,7 +49,8 @@ jobs:
       relevant_paths: ${{ steps.check_paths.outputs.value }}
       labels: ${{ steps.get_labels.outputs.labels }}
     steps:
-      - id: check_paths
+      - name: Check files modified by PR
+        id: check_paths
         run: |
           paths=(
             components/eamxx
@@ -61,9 +67,22 @@ jobs:
           pattern=$(IFS=\|; echo "${paths[*]}")
 
           # Use the GitHub API to get the list of changed files
-          response=$(curl -s -H "Authorization: token ${{ secrets.GITHUB_TOKEN }}" \
-            "https://api.github.com/repos/${{ github.repository }}/pulls/${{ github.event.number }}/files")
-          changed_files=$(echo "$response" | grep -o '"filename": *"[^"]*"' | sed 's/"filename": *//; s/"//g')
+          # There are page size limits, so do it in chunks
+          page=1
+          while true; do
+            response=$(curl -s -H "Authorization: token ${{ secrets.GITHUB_TOKEN }}" \
+              "https://api.github.com/repos/E3SM-Project/scream/pulls/${{ github.event.number }}/files?per_page=100&page=$page")
+
+            # Check if the response is empty, and break if it is
+            [ -z "$response" ] && break
+
+            changed_files+=$(echo "$response" | grep -o '"filename": *"[^"]*"' | sed 's/"filename": *//; s/"//g')$'\n'
+
+            # Check if there are more pages, and quite if there aren't
+            [[ $(echo "$response" | jq '. | length') -lt 100 ]] && break
+
+            page=$((page + 1))
+          done
 
           # Check for matches and echo the matching files (or "" if none)
           matching_files=$(echo "$changed_files" | grep -E "^($pattern)" || echo "")
@@ -74,25 +93,29 @@ jobs:
             echo "No relevant files touched by this PR."
             echo "value=false" >> $GITHUB_OUTPUT
           fi
-      - id: get_labels
+      - name: Retrieve PR labels
+        id: get_labels
         run: |
           labels="${{ join(github.event.pull_request.labels.*.name, ',') }}"
           echo "labels=${labels}" >> $GITHUB_OUTPUT
   gcc-openmp:
     needs: [pre_process_pr]
     if: |
-      github.event_name == 'schedule' ||
+      !failure() && !cancelled() &&
       (
-        github.event_name == 'pull_request' &&
-        needs.pre_process_pr.outputs.relevant_paths=='true' &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip gcc') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip openmp') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-sa') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-all')
-      ) || (
-        github.event_name == 'workflow_dispatch' &&
-        github.event.inputs.job_to_run == 'gcc-openmp' || 
-        github.event.inputs.job_to_run == 'all'
+        github.event_name == 'schedule' ||
+        (
+          github.event_name == 'pull_request' &&
+          needs.pre_process_pr.outputs.relevant_paths=='true' &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip gcc') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip openmp') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-sa') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-all')
+        ) || (
+          github.event_name == 'workflow_dispatch' &&
+          github.event.inputs.job_to_run == 'gcc-openmp' || 
+          github.event.inputs.job_to_run == 'all'
+        )
       )
     runs-on:  [self-hosted, ghci-snl-cpu, gcc]
     strategy:
@@ -128,18 +151,21 @@ jobs:
   gcc-cuda:
     needs: [pre_process_pr]
     if: |
-      github.event_name == 'schedule' ||
+      !failure() && !cancelled() &&
       (
-        github.event_name == 'pull_request' &&
-        needs.pre_process_pr.outputs.relevant_paths=='true' &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip gcc') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip cuda') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-sa') &&
-        !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-all')
-      ) || (
-        github.event_name == 'workflow_dispatch' &&
-        github.event.inputs.job_to_run == 'gcc-cuda' ||
-        github.event.inputs.job_to_run == 'all'
+        github.event_name == 'schedule' ||
+        (
+          github.event_name == 'pull_request' &&
+          needs.pre_process_pr.outputs.relevant_paths=='true' &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip gcc') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip cuda') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-sa') &&
+          !contains(needs.pre_process_pr.outputs.labels,'CI: skip eamxx-all')
+        ) || (
+          github.event_name == 'workflow_dispatch' &&
+          github.event.inputs.job_to_run == 'gcc-cuda' ||
+          github.event.inputs.job_to_run == 'all'
+        )
       )
     runs-on:  [self-hosted, ghci-snl-cuda, cuda, gcc]
     strategy:
@@ -164,11 +190,40 @@ jobs:
               echo "generate=true" >> $GITHUB_ENV
             fi
           fi
+      - name: Get CUDA Arch
+        run: |
+          # Ensure nvidia-smi is available
+          if ! command -v nvidia-smi &> /dev/null; then
+              echo "nvidia-smi could not be found. Please ensure you have Nvidia drivers installed."
+              exit 1
+          fi
+
+          # Get the GPU model from nvidia-smi, and set env for next step
+          gpu_model=$(nvidia-smi --query-gpu=name --format=csv,noheader | head -n 1)
+          case "$gpu_model" in
+              *"H100"*)
+                  echo "Hopper=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=90" >> $GITHUB_ENV
+                  ARCH=90
+                  ;;
+              *"A100"*)
+                  echo "Ampere=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=80" >> $GITHUB_ENV
+                  ;;
+              *"V100"*)
+                  echo "Volta=ON" >> $GITHUB_ENV
+                  echo "CUDA_ARCH=70" >> $GITHUB_ENV
+                  ;;
+              *)
+                  echo "Unsupported GPU model: $gpu_model"
+                  exit 1
+                  ;;
+          esac
       - name: Run tests
         uses: ./.github/actions/test-all-scream
         with:
           build_type: ${{ matrix.build_type }}
           machine: ghci-snl-cuda
           generate: ${{ env.generate }}
           submit: ${{ env.submit }}
-          cmake-configs: Kokkos_ARCH_VOLTA70=ON;CMAKE_CUDA_ARCHITECTURES=70
+          cmake-configs: Kokkos_ARCH_HOPPER90=${{ env.Hopper }};Kokkos_ARCH_AMPERE80=${{ env.Ampere }};Kokkos_ARCH_VOLTA70=${{ env.Volta }};CMAKE_CUDA_ARCHITECTURES=${{ env.CUDA_ARCH }}