From dc342eb623808a1fbea9fca3e59820b41eae4f31 Mon Sep 17 00:00:00 2001
From: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Date: Mon, 11 Aug 2025 21:13:40 +0530
Subject: [PATCH 1/4] Cleanup CI

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
---
 .github/codecov.yml                           |  8 +++
 .github/workflows/_wait_for_checks.yml        | 29 ++++++++++
 .github/workflows/gpu_tests.yml               | 19 +++++-
 .../workflows/multi_version_unit_tests.yml    | 46 ---------------
 .github/workflows/partial_unit_tests.yml      | 32 ----------
 .github/workflows/unit_tests.yml              | 58 ++++++++++++++++++-
 6 files changed, 111 insertions(+), 81 deletions(-)
 create mode 100644 .github/codecov.yml
 create mode 100644 .github/workflows/_wait_for_checks.yml
 delete mode 100644 .github/workflows/multi_version_unit_tests.yml
 delete mode 100644 .github/workflows/partial_unit_tests.yml

diff --git a/.github/codecov.yml b/.github/codecov.yml
new file mode 100644
index 000000000..fec90abb6
--- /dev/null
+++ b/.github/codecov.yml
@@ -0,0 +1,8 @@
+# Allow atmost 5% coverage drop from main branch.
+coverage:
+  status:
+    project:
+      default:
+        target: auto
+        threshold: 5%
+    patch: false
diff --git a/.github/workflows/_wait_for_checks.yml b/.github/workflows/_wait_for_checks.yml
new file mode 100644
index 000000000..9e28fcaa2
--- /dev/null
+++ b/.github/workflows/_wait_for_checks.yml
@@ -0,0 +1,29 @@
+name: Wait for checks
+
+on:
+  workflow_call:
+    inputs:
+      match_pattern:
+        required: true
+        type: string
+      delay:
+        required: false
+        type: string
+        default: 10s
+
+jobs:
+  wait:
+    runs-on: ubuntu-latest
+    permissions:
+      checks: read
+    steps:
+      - name: Wait for checks (PRs only)
+        if: github.event_name == 'pull_request' || startsWith(github.ref, 'refs/heads/pull-request/')
+        uses: poseidon/wait-for-status-checks@v0.6.0
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          match_pattern: ${{ inputs.match_pattern }}
+          delay: ${{ inputs.delay }}
+      - name: No-op for non-PR events
+        if: github.event_name != 'pull_request' && !startsWith(github.ref, 'refs/heads/pull-request/')
+        run: echo "Not a pull_request event"
diff --git a/.github/workflows/gpu_tests.yml b/.github/workflows/gpu_tests.yml
index 2665a6006..c77a43396 100644
--- a/.github/workflows/gpu_tests.yml
+++ b/.github/workflows/gpu_tests.yml
@@ -14,11 +14,28 @@ on:
 
 # Cancel previous runs if new commit is pushed to the same PR
 concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number }}
+  group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
 
 jobs:
+  wait-dco:
+    uses: ./.github/workflows/_wait_for_checks.yml
+    permissions:
+      checks: read
+    secrets: inherit
+    with:
+      match_pattern: '^DCO$'
+  wait-unit-tests:
+    needs: [wait-dco]
+    uses: ./.github/workflows/_wait_for_checks.yml
+    permissions:
+      checks: read
+    secrets: inherit
+    with:
+      match_pattern: '^linux$' # Unit tests / linux
+      delay: 60s
   gpu-tests:
+    needs: [wait-unit-tests]
     # Runner list at https://github.com/nv-gha-runners/enterprise-runner-configuration/blob/main/docs/runner-groups.md
     runs-on: linux-amd64-gpu-h100-latest-1
     timeout-minutes: 60
diff --git a/.github/workflows/multi_version_unit_tests.yml b/.github/workflows/multi_version_unit_tests.yml
deleted file mode 100644
index def4b392b..000000000
--- a/.github/workflows/multi_version_unit_tests.yml
+++ /dev/null
@@ -1,46 +0,0 @@
-# Run unit tests with older supported Python and Torch versions
-name: Multi version tests
-
-on:
-  pull_request:
-    branches: [main, release/*]
-    paths:
-      - ".github/workflows/multi_version_unit_tests.yml"
-      - "modelopt/**"
-      - "tests/unit/**"
-      - "pyproject.toml"
-      - "setup.py"
-      - "tox.ini"
-
-# Cancel previous runs if new commit is pushed to the same PR
-concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number }}
-  cancel-in-progress: true
-
-jobs:
-  multi-py-unit:
-    runs-on: ubuntu-latest
-    timeout-minutes: 30
-    strategy:
-      matrix:
-        py: [10, 11]
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.${{ matrix.py }}"
-      - name: Run unit tests
-        run: pip install tox && tox -e py3${{ matrix.py }}-torch28-unit
-  multi-torch-unit:
-    runs-on: ubuntu-latest
-    timeout-minutes: 30
-    strategy:
-      matrix:
-        torch: [25, 26, 27]
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-      - name: Run unit tests
-        run: pip install tox && tox -e py312-torch${{ matrix.torch }}-unit
diff --git a/.github/workflows/partial_unit_tests.yml b/.github/workflows/partial_unit_tests.yml
deleted file mode 100644
index c3cb5ce18..000000000
--- a/.github/workflows/partial_unit_tests.yml
+++ /dev/null
@@ -1,32 +0,0 @@
-name: Partial install unit tests
-
-on:
-  pull_request:
-    branches: [main, release/*]
-    paths:
-      - ".github/workflows/partial_unit_tests.yml"
-      - "modelopt/**"
-      - "tests/unit/**"
-      - "pyproject.toml"
-      - "setup.py"
-      - "tox.ini"
-
-# Cancel previous runs if new commit is pushed to the same PR
-concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number }}
-  cancel-in-progress: true
-
-jobs:
-  partial-unit:
-    runs-on: ubuntu-latest
-    timeout-minutes: 30
-    strategy:
-      matrix:
-        test-env: [onnx, torch]
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-      - name: Run unit tests
-        run: pip install tox && tox -e py312-partial-unit-${{ matrix.test-env }}
diff --git a/.github/workflows/unit_tests.yml b/.github/workflows/unit_tests.yml
index 9fcc6c87b..355baab54 100644
--- a/.github/workflows/unit_tests.yml
+++ b/.github/workflows/unit_tests.yml
@@ -26,7 +26,15 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  unit-tests:
+  wait-dco:
+    uses: ./.github/workflows/_wait_for_checks.yml
+    permissions:
+      checks: read
+    secrets: inherit
+    with:
+      match_pattern: '^DCO$'
+  linux:
+    needs: [wait-dco]
     runs-on: ubuntu-latest
     timeout-minutes: 30
     steps:
@@ -42,8 +50,9 @@ jobs:
           token: ${{ secrets.CODECOV_TOKEN }}
           fail_ci_if_error: true
           verbose: true
-  unit-tests-windows:
+  windows:
     if: github.event_name == 'pull_request'
+    needs: [linux]
     runs-on: windows-latest
     timeout-minutes: 30
     steps:
@@ -53,3 +62,48 @@ jobs:
           python-version: "3.12"
       - name: Run unit tests (without coverage)
         run: pip install tox && tox -e py312-torch28-unit
+  multi-py:
+    if: github.event_name == 'pull_request'
+    needs: [linux]
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    strategy:
+      matrix:
+        py: [10, 11]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.${{ matrix.py }}"
+      - name: Run unit tests
+        run: pip install tox && tox -e py3${{ matrix.py }}-torch28-unit
+  multi-torch:
+    if: github.event_name == 'pull_request'
+    needs: [linux]
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    strategy:
+      matrix:
+        torch: [25, 26, 27]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Run unit tests
+        run: pip install tox && tox -e py312-torch${{ matrix.torch }}-unit
+  partial-install:
+    if: github.event_name == 'pull_request'
+    needs: [linux]
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    strategy:
+      matrix:
+        test-env: [onnx, torch]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Run unit tests
+        run: pip install tox && tox -e py312-partial-unit-${{ matrix.test-env }}

From db23e00325d1724992f1a57cfd9d73f18e304d08 Mon Sep 17 00:00:00 2001
From: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Date: Mon, 11 Aug 2025 22:04:01 +0530
Subject: [PATCH 2/4] Run gpu tests if relevant files are changed

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
---
 .github/workflows/gpu_tests.yml  | 37 +++++++++++++++++++++-----------
 .github/workflows/pages.yml      |  2 +-
 .github/workflows/unit_tests.yml |  2 +-
 3 files changed, 27 insertions(+), 14 deletions(-)

diff --git a/.github/workflows/gpu_tests.yml b/.github/workflows/gpu_tests.yml
index c77a43396..3eca9cc5f 100644
--- a/.github/workflows/gpu_tests.yml
+++ b/.github/workflows/gpu_tests.yml
@@ -3,18 +3,11 @@ name: GPU tests
 on:
   push:
     branches: ["pull-request/[0-9]+"]
-    # TODO: paths cannot be used since push happens to copied PR and only latest commit to PR is used
-    # paths:
-    #   - ".github/workflows/gpu_tests.yml"
-    #   - "modelopt/**"
-    #   - "tests/gpu/**"
-    #   - "pyproject.toml"
-    #   - "setup.py"
-    #   - "tox.ini"
+    # NOTE: paths cannot be used since push happens to copied PR and only latest commit to PR is used
 
 # Cancel previous runs if new commit is pushed to the same PR
 concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
+  group: ${{ github.workflow }}-${{ github.sha }}
   cancel-in-progress: true
 
 jobs:
@@ -25,6 +18,26 @@ jobs:
     secrets: inherit
     with:
       match_pattern: '^DCO$'
+  check-changes:
+    needs: [wait-dco]
+    runs-on: ubuntu-latest
+    outputs:
+      any_changed: ${{ steps.changed-tests.outputs.any_changed }}
+    steps:
+      - id: get-pr-info
+        uses: nv-gha-runners/get-pr-info@main
+      - name: Check for changes in test-relevant directories
+        id: changed-tests
+        uses: step-security/changed-files@v46.0.5
+        with:
+          files: |
+            .github/workflows/gpu_tests.yml
+            modelopt/**
+            tests/gpu/**
+            tox.ini
+            pyproject.toml
+            setup.py
+          base_sha: ${{ fromJSON(steps.get-pr-info.outputs.pr-info).base.ref }}
   wait-unit-tests:
     needs: [wait-dco]
     uses: ./.github/workflows/_wait_for_checks.yml
@@ -35,7 +48,8 @@ jobs:
       match_pattern: '^linux$' # Unit tests / linux
       delay: 60s
   gpu-tests:
-    needs: [wait-unit-tests]
+    needs: [check-changes, wait-unit-tests]
+    if: needs.check-changes.outputs.any_changed == 'true'
     # Runner list at https://github.com/nv-gha-runners/enterprise-runner-configuration/blob/main/docs/runner-groups.md
     runs-on: linux-amd64-gpu-h100-latest-1
     timeout-minutes: 60
@@ -47,7 +61,6 @@ jobs:
         PIP_CONSTRAINT: "" # Disable pip constraint for upgrading packages
     steps:
       - uses: actions/checkout@v4
-      - name: Setup proxy cache
-        uses: nv-gha-runners/setup-proxy-cache@main
+      - uses: nv-gha-runners/setup-proxy-cache@main
       - name: Run gpu tests
         run: pip install tox-current-env && tox -e py312-cuda12-gpu --current-env
diff --git a/.github/workflows/pages.yml b/.github/workflows/pages.yml
index 7b76bf75c..ba7355c57 100644
--- a/.github/workflows/pages.yml
+++ b/.github/workflows/pages.yml
@@ -8,7 +8,7 @@ on:
 
 # Cancel previous runs if new commit is pushed
 concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}
   cancel-in-progress: true
 
 # Sets permissions of the GITHUB_TOKEN to allow deployment to GitHub Pages
diff --git a/.github/workflows/unit_tests.yml b/.github/workflows/unit_tests.yml
index 355baab54..34f0124c7 100644
--- a/.github/workflows/unit_tests.yml
+++ b/.github/workflows/unit_tests.yml
@@ -22,7 +22,7 @@ on:
 
 # Cancel previous runs if new commit is pushed
 concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}
   cancel-in-progress: true
 
 jobs:

From 9834826aa6ade3ff5601729a0d483a6eea1b7288 Mon Sep 17 00:00:00 2001
From: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Date: Mon, 11 Aug 2025 22:15:39 +0530
Subject: [PATCH 3/4] temp

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
---
 .github/workflows/gpu_tests.yml | 20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)

diff --git a/.github/workflows/gpu_tests.yml b/.github/workflows/gpu_tests.yml
index 3eca9cc5f..c3726609f 100644
--- a/.github/workflows/gpu_tests.yml
+++ b/.github/workflows/gpu_tests.yml
@@ -11,19 +11,12 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  wait-dco:
-    uses: ./.github/workflows/_wait_for_checks.yml
-    permissions:
-      checks: read
-    secrets: inherit
-    with:
-      match_pattern: '^DCO$'
   check-changes:
-    needs: [wait-dco]
     runs-on: ubuntu-latest
     outputs:
       any_changed: ${{ steps.changed-tests.outputs.any_changed }}
     steps:
+      - uses: actions/checkout@v4
       - id: get-pr-info
         uses: nv-gha-runners/get-pr-info@main
       - name: Check for changes in test-relevant directories
@@ -38,17 +31,18 @@ jobs:
             pyproject.toml
             setup.py
           base_sha: ${{ fromJSON(steps.get-pr-info.outputs.pr-info).base.ref }}
-  wait-unit-tests:
-    needs: [wait-dco]
+  wait-checks:
+    needs: [check-changes]
+    if: needs.check-changes.outputs.any_changed == 'true'
     uses: ./.github/workflows/_wait_for_checks.yml
     permissions:
       checks: read
     secrets: inherit
     with:
-      match_pattern: '^linux$' # Unit tests / linux
-      delay: 60s
+      match_pattern: '^DCO$|^linux$' # Wait for DCO and Unit tests / linux to pass
+      delay: 300s
   gpu-tests:
-    needs: [check-changes, wait-unit-tests]
+    needs: [check-changes, wait-checks]
     if: needs.check-changes.outputs.any_changed == 'true'
     # Runner list at https://github.com/nv-gha-runners/enterprise-runner-configuration/blob/main/docs/runner-groups.md
     runs-on: linux-amd64-gpu-h100-latest-1

From 660c3d9d7067ce4151dc48636f50ac805f1ceb0d Mon Sep 17 00:00:00 2001
From: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Date: Mon, 11 Aug 2025 22:51:16 +0530
Subject: [PATCH 4/4] temp

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
---
 .github/workflows/gpu_tests.yml  | 10 +++++-----
 .github/workflows/unit_tests.yml |  4 ++--
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/gpu_tests.yml b/.github/workflows/gpu_tests.yml
index c3726609f..b9b2ce540 100644
--- a/.github/workflows/gpu_tests.yml
+++ b/.github/workflows/gpu_tests.yml
@@ -11,7 +11,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  check-changes:
+  check-file-changes:
     runs-on: ubuntu-latest
     outputs:
       any_changed: ${{ steps.changed-tests.outputs.any_changed }}
@@ -32,8 +32,8 @@ jobs:
             setup.py
           base_sha: ${{ fromJSON(steps.get-pr-info.outputs.pr-info).base.ref }}
   wait-checks:
-    needs: [check-changes]
-    if: needs.check-changes.outputs.any_changed == 'true'
+    needs: [check-file-changes]
+    if: needs.check-file-changes.outputs.any_changed == 'true'
     uses: ./.github/workflows/_wait_for_checks.yml
     permissions:
       checks: read
@@ -42,8 +42,8 @@ jobs:
       match_pattern: '^DCO$|^linux$' # Wait for DCO and Unit tests / linux to pass
       delay: 300s
   gpu-tests:
-    needs: [check-changes, wait-checks]
-    if: needs.check-changes.outputs.any_changed == 'true'
+    needs: [check-file-changes, wait-checks]
+    if: needs.check-file-changes.outputs.any_changed == 'true'
     # Runner list at https://github.com/nv-gha-runners/enterprise-runner-configuration/blob/main/docs/runner-groups.md
     runs-on: linux-amd64-gpu-h100-latest-1
     timeout-minutes: 60
diff --git a/.github/workflows/unit_tests.yml b/.github/workflows/unit_tests.yml
index 34f0124c7..de3568158 100644
--- a/.github/workflows/unit_tests.yml
+++ b/.github/workflows/unit_tests.yml
@@ -26,7 +26,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  wait-dco:
+  check-dco:
     uses: ./.github/workflows/_wait_for_checks.yml
     permissions:
       checks: read
@@ -34,7 +34,7 @@ jobs:
     with:
       match_pattern: '^DCO$'
   linux:
-    needs: [wait-dco]
+    needs: [check-dco]
     runs-on: ubuntu-latest
     timeout-minutes: 30
     steps: