Enable MPS CI runners (#252)

* Test MPS CI runners

* Cherry pick remaining files

* Enable lintrunner:

* Change lint  runner

* Retrigger checks

* Retrigger checks #2

* Retrigger checks #3

* Retrigger checks #4

* Retrigger checks #5

* Retrigger checks #5

* Retrigger checks #7

* Retrigger checks #8

* Retrigger checks #9

* Retrigger checks #9 (change arch to arm)

* Retrigger checks #10

* Retrigger checks #11

* Retrigger checks #12

* Retrigger checks #13

* Retrigger checks #14

* Retrigger checks #14

* Retrigger checks #15

* Retrigger checks #16

* Retrigger checks #16

* Retrigger checks #17

* Retrigger checks #19

* Retrigger checks #20

* Retrigger checks #21

* Fix lintrunner

* Fix lintrunner

* Remove lint.json

.github/actionlint.yaml

@@ -16,6 +16,7 @@ self-hosted-runner:
     - bm-runner
     - linux.rocm.gpu
     - macos-m1-12
+    - macos-m1-13
     - macos-12-xl
     - macos-12
     - macos12.3-m1

.github/workflows/_mac-build.yml

@@ -63,8 +63,8 @@ on:
 
 jobs:
   build:
-    # Don't run on forked repos.
-    if: github.repository_owner == 'pytorch'
+    # # Don't run on forked repos.
+    # if: github.repository_owner == 'pytorch'
     runs-on: ${{ inputs.runner-type }}
     env:
       # For sccache access (only on non-forked PRs)

.github/workflows/_mac-test-mps.yml

@@ -14,11 +14,16 @@ on:
         description: |
           If this is set, our linter will use this to make sure that every other
           job with the same `sync-tag` is identical.
+      runs-on:
+        required: false
+        type: string
+        default: "macos-m1-12"
+        description: Hardware to run tests on
 
 jobs:
   run_mps_test:
     name: "Run MPS tests"
-    runs-on: macos-m1-12
+    runs-on: ${{ inputs.runs-on }}
     steps:
       - name: Checkout PyTorch
         uses: malfet/checkout@silent-checkout
@@ -77,6 +82,12 @@ jobs:
 
           ${CONDA_RUN} python3 test/run_test.py --mps --verbose
 
+      - name: Print remaining test logs
+        shell: bash
+        if: always()
+        run: |
+          cat test/**/*.log || true
+
       - name: Get workflow job id
         id: get-job-id
         uses: ./.github/actions/get-workflow-job-id
@@ -86,7 +97,7 @@ jobs:
 
       - name: Upload test artifacts
         uses: ./.github/actions/upload-test-artifacts
-        if: always() && (steps.test.conclusion == 'success' || steps.test.conclusion == 'failure')
+        if: always() && steps.test.conclusion && steps.test.conclusion != 'skipped'
         with:
           use-gha: true
           file-suffix: ${{ github.job }}-mps-1-1-macos-m1-12_${{ steps.get-job-id.outputs.job-id }}

.github/workflows/_mac-test.yml

@@ -106,6 +106,7 @@ jobs:
 
       - name: Start monitoring script
         id: monitor-script
+        continue-on-error: true
         run: |
           ${CONDA_RUN} python3 -m tools.stats.monitor > usage_log.txt 2>&1 &
           echo "monitor-script-pid=${!}" >> "${GITHUB_OUTPUT}"
@@ -150,6 +151,12 @@ jobs:
           ${CONDA_RUN} python3 -mpip install --no-index --no-deps $(echo dist/*.whl)
           ${CONDA_RUN} .jenkins/pytorch/macos-test.sh
 
+      - name: Print remaining test logs
+        shell: bash
+        if: always()
+        run: |
+          cat test/**/*.log || true
+
       - name: Get workflow job id
         id: get-job-id
         uses: ./.github/actions/get-workflow-job-id
@@ -167,7 +174,7 @@ jobs:
 
       - name: Upload test artifacts
         uses: ./.github/actions/upload-test-artifacts
-        if: always() && (steps.test.conclusion == 'success' || steps.test.conclusion == 'failure')
+        if: always() && steps.test.conclusion && steps.test.conclusion != 'skipped'
         with:
           use-gha: true
           file-suffix: ${{ github.job }}-${{ matrix.config }}-${{ matrix.shard }}-${{ matrix.num_shards }}-${{ matrix.runner }}_${{ steps.get-job-id.outputs.job-id }}