Release prep

.github/workflows/release.yml

@@ -0,0 +1,110 @@
+name: Release
+
+# Publishes the pure-Python nvsubquadratic wheel + sdist to PyPI on a
+# `vX.Y.Z` tag push (or manual dispatch with an existing tag).
+#
+# Routing:
+#   - prerelease tags containing "rc" (e.g. v0.1.0rc1) -> TestPyPI
+#   - final tags (e.g. v0.1.0)                         -> PyPI
+#
+# Auth is OIDC "trusted publisher" — no API token secret. One-time setup
+# required before the first run:
+#   1. On pypi.org and test.pypi.org, add a trusted publisher for this repo
+#      pointing at workflow `release.yml` and the matching environment below.
+#   2. Create GitHub environments named `pypi` and `testpypi`
+#      (Settings -> Environments).
+# The package version is read from the VERSION file; the build job fails fast
+# if the tag does not match it, so bump VERSION before tagging.
+
+on:
+  push:
+    tags:
+      - "v*.*.*"
+  workflow_dispatch:
+    inputs:
+      tag:
+        description: "Existing tag to build & publish (e.g. v0.1.0 or v0.1.0rc1)"
+        required: true
+
+permissions:
+  contents: read
+
+concurrency:
+  group: release-${{ github.event.inputs.tag || github.ref_name }}
+  cancel-in-progress: false
+
+jobs:
+  build:
+    name: Build sdist + wheel
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.inputs.tag || github.ref }}
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
+      - name: Verify tag matches VERSION
+        run: |
+          TAG="${{ github.event.inputs.tag || github.ref_name }}"
+          TAG_VER="${TAG#v}"
+          FILE_VER="$(tr -d '[:space:]' < VERSION)"
+          if [ "$TAG_VER" != "$FILE_VER" ]; then
+            echo "::error::Tag $TAG (version $TAG_VER) does not match VERSION file ($FILE_VER). Bump VERSION before tagging."
+            exit 1
+          fi
+          echo "Tag $TAG matches VERSION $FILE_VER"
+
+      - name: Build
+        run: |
+          python -m pip install --upgrade pip build twine
+          python -m build
+          twine check dist/*
+
+      - name: Upload dist artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+  publish-testpypi:
+    name: Publish to TestPyPI (prerelease)
+    needs: build
+    if: contains(github.event.inputs.tag || github.ref_name, 'rc')
+    runs-on: ubuntu-latest
+    environment: testpypi
+    permissions:
+      id-token: write
+    steps:
+      - name: Download dist
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: Publish to TestPyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          repository-url: https://test.pypi.org/legacy/
+
+  publish-pypi:
+    name: Publish to PyPI (final release)
+    needs: build
+    if: ${{ !contains(github.event.inputs.tag || github.ref_name, 'rc') }}
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write
+    steps:
+      - name: Download dist
+        uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1

README.md

@@ -16,7 +16,7 @@ nvSubquadratic consolidates efforts from across NVIDIA Research teams (nvResearc
 
 **Requirements**:
 
-- CUDA-compatible NVIDIA GPU (Ampere or Hopper architecture)
+- CUDA-compatible NVIDIA GPU (Ampere or newer)
 - CUDA Toolkit 12.0 or higher
 - Python 3.11 or higher
 

docs/getting_started.md

@@ -6,7 +6,7 @@ Apptainer, conda, venv) see the project [README](https://github.com/NVIDIA-BioNe
 
 ## Requirements
 
-- CUDA-compatible NVIDIA GPU (Ampere or Hopper architecture)
+- CUDA-compatible NVIDIA GPU (Ampere or newer)
 - CUDA Toolkit 12.0 or higher
 - Python 3.11 or higher
 

docs/index.rst

@@ -30,7 +30,7 @@ To enable the optional fused RMSNorm kernel on Hopper / Blackwell GPUs:
 Requirements
 ------------
 
-- CUDA-compatible NVIDIA GPU (Ampere or Hopper architecture)
+- CUDA-compatible NVIDIA GPU (Ampere or newer)
 - CUDA Toolkit 12.0 or higher
 - Python 3.11 or higher