Improve lintrunner job runtime

.github/workflows/_get-changed-files.yml

@@ -0,0 +1,43 @@
+name: Get Changed Files
+
+on:
+  workflow_call:
+    outputs:
+      changed-files:
+        description: "List of changed files (space-separated) or '*' if not in a PR"
+        value: ${{ jobs.get-changed-files.outputs.changed-files }}
+
+jobs:
+  get-changed-files:
+    runs-on: ubuntu-latest
+    outputs:
+      changed-files: ${{ steps.get-files.outputs.changed-files }}
+
+    steps:
+      - name: Get changed files
+        id: get-files
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: |
+          # Check if we're in a pull request context
+          if [ "${{ github.event_name }}" = "pull_request" ] || [ "${{ github.event_name }}" = "pull_request_target" ]; then
+            echo "Running in PR context"
+
+            # Get the PR number from the github context
+            PR_NUMBER="${{ github.event.number }}"
+
+            # Use gh CLI to get changed files in the PR with explicit repo
+            CHANGED_FILES=$(gh api repos/${{ github.repository }}/pulls/$PR_NUMBER/files --paginate --jq '.[] | select(.status != "removed") | .filename' | tr '\n' ' ' | sed 's/ $//')
+
+            if [ -z "$CHANGED_FILES" ]; then
+              echo "No changed files found, setting to '*'"
+              CHANGED_FILES="*"
+            fi
+
+            echo "Changed files: $CHANGED_FILES"
+            echo "changed-files=$CHANGED_FILES" >> "$GITHUB_OUTPUT"
+
+          else
+            echo "Not in PR context, setting changed files to '*'"
+            echo "changed-files=*" >> "$GITHUB_OUTPUT"
+          fi

.github/workflows/lint.yml

@@ -15,15 +15,27 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  lintrunner:
+  get-changed-files:
+    if: github.repository_owner == 'pytorch'
+    name: Get changed files
+    uses: ./.github/workflows/_get-changed-files.yml
+
+  lintrunner-mypy:
     uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
+    needs: [get-changed-files]
     permissions:
       id-token: write
       contents: read
+    if: |
+      github.repository_owner == 'pytorch' && (
+        needs.get-changed-files.outputs.changed-files == '*' ||
+        contains(needs.get-changed-files.outputs.changed-files, '.py') ||
+        contains(needs.get-changed-files.outputs.changed-files, '.pyi')
+      )
     with:
       runner: linux.2xlarge
       docker-image: ci-image:executorch-ubuntu-22.04-linter
-      submodules: 'recursive'
+      submodules: true
       fetch-depth: 0
       ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
       timeout: 90
@@ -45,14 +57,64 @@ jobs:
           cp -r "${CACHE_DIRECTORY}" . || true
         fi
 
-        # This has already been cached in the docker image
-        lintrunner init
+        RC=0
+        # Run lintrunner on all files. pytorch/pytorch notes that mypy
+        # in particular needs this.
+        if ! lintrunner --force-color --all-files --take MYPY --tee-json=lint.json 2> /dev/null; then
+          echo ""
+          echo -e "\e[1m\e[36mYou can reproduce these results locally by using \`lintrunner --take MYPY\`. (If you don't get the same results, run \'lintrunner init\' to update your local linter)\e[0m"
+          echo -e "\e[1m\e[36mSee https://github.com/pytorch/pytorch/wiki/lintrunner for setup instructions.\e[0m"
+          RC=1
+        fi
+
+        # Use jq to massage the JSON lint output into GitHub Actions workflow commands.
+        jq --raw-output \
+          '"::\(if .severity == "advice" or .severity == "disabled" then "warning" else .severity end) file=\(.path),line=\(.line),col=\(.char),title=\(.code) \(.name)::" + (.description | gsub("\\n"; "%0A"))' \
+          lint.json || true
+
+        exit $RC
+
+  lintrunner:
+    uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
+    needs: [get-changed-files]
+    permissions:
+      id-token: write
+      contents: read
+    with:
+      runner: linux.2xlarge
+      docker-image: ci-image:executorch-ubuntu-22.04-linter
+      submodules: false
+      fetch-depth: 0
+      ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
+      timeout: 90
+      script: |
+        # The generic Linux job chooses to use base env, not the one setup by the image
+        CONDA_ENV=$(conda env list --json | jq -r ".envs | .[-1]")
+        conda activate "${CONDA_ENV}"
+
+        # Not sure why this isn't set up in the docker
+        # image. lintrunner-mypy seems to work becaus setup-linux.sh
+        # does this as part of install_executorch.
+        pip install -r requirements-dev.txt
+
+        CACHE_DIRECTORY="/tmp/.lintbin"
+        # Try to recover the cached binaries
+        if [[ -d "${CACHE_DIRECTORY}" ]]; then
+          # It's ok to fail this as lintrunner init would download these binaries
+          # again if they do not exist
+          cp -r "${CACHE_DIRECTORY}" . || true
+        fi
 
         RC=0
-        # Run lintrunner on all files
-        if ! lintrunner --force-color --all-files --tee-json=lint.json 2> /dev/null; then
+        CHANGED_FILES="${{ needs.get-changed-files.outputs.changed-files }}"
+        if [ "$CHANGED_FILES" = '*' ]; then
+          LINTRUNNER_FILES="--all-files"
+        else
+          LINTRUNNER_FILES="${CHANGED_FILES}"
+        fi
+        if ! lintrunner --force-color ${LINTRUNNER_FILES} --skip MYPY --tee-json=lint.json 2> /dev/null; then
           echo ""
-          echo -e "\e[1m\e[36mYou can reproduce these results locally by using \`lintrunner\`. (If you don't get the same results, run \'lintrunner init\' to update your local linter)\e[0m"
+          echo -e "\e[1m\e[36mYou can reproduce these results locally by using \`lintrunner --skip MYPY\`. (If you don't get the same results, run \'lintrunner init\' to update your local linter)\e[0m"
           echo -e "\e[1m\e[36mSee https://github.com/pytorch/pytorch/wiki/lintrunner for setup instructions.\e[0m"
           RC=1
         fi