Merge remote-tracking branch 'upstream/main' into main

.github/workflows/ci.yml

@@ -141,7 +141,7 @@ jobs:
     downstream_tests:
         strategy:
             matrix:
-                downstream_project: [meshmode, grudge, pytential, pytato]
+                downstream_project: [arraycontext, meshmode, grudge, pytential, pytato]
             fail-fast: false
         name: Tests for downstream project ${{ matrix.downstream_project }}
         runs-on: ubuntu-latest
@@ -190,4 +190,14 @@ jobs:
 
                 pytest --tb=native -rsxw --durations=10 -m 'not parallel' tests/multigrid/
 
+    validate_cff:
+            name: Validate CITATION.cff
+            runs-on: ubuntu-latest
+            steps:
+            -   uses: actions/checkout@v3
+            -   uses: actions/setup-python@v4
+            -   run: |
+                    pip install cffconvert
+                    cffconvert -i CITATION.cff --validate
+
 # vim: sw=4

.gitlab-ci.yml

@@ -1,9 +1,4 @@
-stages:
-  - test
-  - deploy
-
 Pytest POCL:
-  stage: test
   script:
   - export PYOPENCL_TEST=portable:pthread
   - export EXTRA_INSTALL="pybind11 numpy mako"
@@ -20,7 +15,6 @@ Pytest POCL:
       junit: test/pytest.xml
 
 Pytest Nvidia Titan V:
-  stage: test
   script:
   - export PYOPENCL_TEST=nvi:titan
   - export EXTRA_INSTALL="pybind11 numpy mako"
@@ -38,7 +32,6 @@ Pytest Nvidia Titan V:
       junit: test/pytest.xml
 
 Pytest POCL without arg check:
-  stage: test
   script:
   - export PYOPENCL_TEST=portable:pthread
   - export EXTRA_INSTALL="pybind11 numpy mako"
@@ -56,7 +49,6 @@ Pytest POCL without arg check:
       junit: test/pytest.xml
 
 Pytest Intel:
-  stage: test
   script:
   - export PYOPENCL_TEST=intel
   - export EXTRA_INSTALL="pybind11 numpy mako"
@@ -76,7 +68,6 @@ Pytest Intel:
 
 
 Pytest POCL Twice With Cache:
-  stage: test
   script: |
     export PYOPENCL_TEST=portable:pthread
     export EXTRA_INSTALL="pybind11 numpy mako"
@@ -109,7 +100,6 @@ Pytest POCL Twice With Cache:
 #   - tags
 
 Pytest POCL Examples:
-  stage: test
   script: |
     export PYOPENCL_TEST=portable:pthread
     export EXTRA_INSTALL="pybind11 numpy mako"
@@ -134,7 +124,6 @@ Pytest POCL Examples:
   - tags
 
 Pylint:
-  stage: test
   script:
   # Needed to avoid name shadowing issues when running from source directory.
   - PROJECT_INSTALL_FLAGS="--editable"
@@ -147,7 +136,6 @@ Pylint:
   - tags
 
 Documentation:
-  stage: deploy
   script: |
     EXTRA_INSTALL="pybind11 numpy"
     curl -L -O https://tiker.net/ci-support-v0
@@ -160,7 +148,6 @@ Documentation:
   - python3
 
 Flake8:
-  stage: test
   script:
   - curl -L -O https://gitlab.tiker.net/inducer/ci-support/raw/main/prepare-and-run-flake8.sh
   - . ./prepare-and-run-flake8.sh "$CI_PROJECT_NAME" test examples
@@ -182,26 +169,10 @@ Mypy:
   except:
   - tags
 
-Benchmarks:
-  stage: test
-  script:
-  - CONDA_ENVIRONMENT=.test-conda-env-py3.yml
-  - PROJECT=loopy
-  - PYOPENCL_TEST=portable:pthread
-  - export LOOPY_NO_CACHE=1
-  - export ASV_FACTOR=1.5
-  - curl -L -O https://gitlab.tiker.net/inducer/ci-support/raw/main/build-and-benchmark-py-project.sh
-  - ". ./build-and-benchmark-py-project.sh"
-  tags:
-  - linux
-  - benchmark
-  except:
-  - tags
-
 Downstream:
   parallel:
     matrix:
-    - DOWNSTREAM_PROJECT: [meshmode, grudge, pytential, pytato]
+    - DOWNSTREAM_PROJECT: [arraycontext, meshmode, grudge, pytential, pytato]
   tags:
   - large-node
   - "docker-runner"

CITATION.cff

@@ -0,0 +1,103 @@
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+authors:
+
+# major contributors
+
+- family-names: "Kloeckner"
+  given-names: "Andreas"
+  orcid: "https://orcid.org/0000-0003-1228-519X"
+- family-names: Kulkarni
+  given-names: Kaushik
+  email:  [email protected]
+- family-names: Kempf
+  given-names: Dominic
+  email: [email protected]
+- family-names: Wala
+  given-names: Matt
+  email: [email protected]
+- family-names: Curtis
+  given-names: Nick
+  email: [email protected]
+- family-names: Stevens
+  given-names: James
+  email: [email protected]
+- family-names: Fernando
+  given-names: Isuru
+  email: [email protected]
+
+# smaller fixes
+
+- family-names: Mitchell
+  given-names: Lawrence
+  email: [email protected]
+- family-names: Alvey-Blanco
+  given-names: Addison J.
+  email: [email protected]
+- family-names: Fikl
+  given-names: Alexandru
+  email: [email protected]
+- family-names: Malone
+  given-names: Chris
+  email: [email protected]
+- family-names: Ward
+  given-names: Connor
+  email: [email protected]
+- family-names: Wilcox
+  given-names: Lucas C.
+  email: [email protected]
+- family-names: Koch
+  given-names: Marcel
+  email: [email protected]
+- family-names: Woodman
+  given-names: Marmaduke
+  email: [email protected]
+- family-names: Smith
+  given-names: Matthew
+  email: [email protected]
+- family-names: Diener
+  given-names: Matthias
+  email: [email protected]
+- family-names: Christensen
+  given-names: Nicholas
+  email: [email protected]
+- family-names: Nykto
+  given-names: Nicolas
+  email: [email protected]
+- family-names: Kirby
+  given-names: Robert C.
+  email: [email protected]
+- family-names: Hegmann
+  given-names: Sebastian
+  email: [email protected]
+- family-names: Vorderwuelbecke
+  given-names: Sophia
+  email: [email protected]
+- family-names: Ratnayaka
+  given-names: Thilina
+  email: [email protected]
+- family-names: Gibson
+  given-names: Thomas
+  email: [email protected]
+- family-names: Sun
+  given-names: Tianjiao
+  email: [email protected]
+- family-names: Smith
+  given-names: Timothy A.
+  email: [email protected]
+- family-names: Warburton
+  given-names: Tim
+  email: [email protected]
+- family-names: Wei
+  given-names: Xiaoyu
+  email: [email protected]
+- family-names: Weiner
+  given-names: Zach
+  email: [email protected]
+
+title: "Loopy"
+version: 2024.1
+date-released: 2024-02-16
+url: "https://github.com/inducer/loopy"
+doi: 10.5281/zenodo.10672275
+license: MIT

MANIFEST.in

@@ -19,4 +19,5 @@ include configure.py
 include Makefile.in
 include README.rst
 include LICENSE
+include CITATION.cff
 include requirements*.txt

README.rst

@@ -10,6 +10,9 @@ Loopy: Transformation-Based Generation of High-Performance CPU/GPU Code
 .. image:: https://badge.fury.io/py/loopy.png
     :alt: Python Package Index Release Page
     :target: https://pypi.org/project/loopy/
+.. image:: https://zenodo.org/badge/20281732.svg
+    :alt: Zenodo DOI for latest release
+    :target: https://zenodo.org/doi/10.5281/zenodo.10672274
 
 Loopy lets you easily generate the tedious, complicated code that is necessary
 to get good performance out of GPUs and multi-core CPUs.

doc/ref_kernel.rst

@@ -200,6 +200,8 @@ Tag                             Meaning
 ``"l.N"``                       Local (intra-group) axis N ("local")
 ``"g.N"``                       Group-number axis N ("group")
 ``"unr"``                       Unroll
+``"unr_hint"``                  Unroll using compiler directives
+``"unr_hint.N"``                Unroll at most N times using compiler directives
 ``"ilp"`` | ``"ilp.unr"``       Unroll using instruction-level parallelism
 ``"ilp.seq"``                   Realize parallel iname as innermost loop
 ``"like.INAME"``                Can be used when tagging inames to tag like another
@@ -535,8 +537,6 @@ have the lifetime of a kernel invocation.
 .. autoclass:: AddressSpace
 
 .. autoclass:: TemporaryVariable
-    :members:
-    :undoc-members:
 
 .. _types: