Merge branch 'main' into flip-launch-args

.github/workflows/gh-build-and-test.yml

@@ -300,9 +300,9 @@ jobs:
       - name: Run cuda.core tests
         shell: bash --noprofile --norc -xeuo pipefail {0}
         run: |
-          if [[ $SKIP_CUDA_BINDINGS_TEST == 1 ]]; then
+          if [[ ${{ matrix.python-version }} == "3.13" ]]; then
             # TODO: remove this hack once cuda-python has a cp313 build
-            if [[ ${{ matrix.python-version }} == "3.13" ]]; then
+            if [[ $SKIP_CUDA_BINDINGS_TEST == 1 ]]; then
               echo "Python 3.13 + cuda-python ${{ matrix.cuda-version }} is not supported, skipping the test..."
               exit 0
             fi
@@ -316,9 +316,6 @@ jobs:
           popd
 
           pushd ./cuda_core
-          # TODO: add requirements.txt for test deps?
-          pip install pytest
-          # TODO: add CuPy to test deps (which would require cuRAND)
-          # pip install "cupy-cuda${TEST_CUDA_MAJOR}x"
+          pip install -r "tests/requirements-cu${TEST_CUDA_MAJOR}.txt"
           pytest -rxXs tests/
           popd

.github/workflows/triagelabel.yml

@@ -2,27 +2,21 @@ name: Add Triage Label
 
 on:
   issues:
-    types: [opened]
+    types:
+      - reopened
+      - opened
 
 jobs:
   triage:
     runs-on: ubuntu-latest
+    permissions:
+      issues: write
     steps:
-    - name: Check for existing labels
-      id: check_labels
-      uses: actions/github-script@v6
-      with:
-        script: |
-          const labels = await github.issues.listLabelsOnIssue({
-            owner: context.repo.owner,
-            repo: context.repo.repo,
-            issue_number: context.issue.number
-          });
-          return labels.data.length > 0;
-
-    - name: Add Triage Label
-      if: steps.check_labels.outputs.result == 'false'
-      uses: actions-ecosystem/action-add-labels@v1
-      with:
-        github_token: ${{ secrets.GITHUB_TOKEN }}
-        labels: triage
+      - name: Add or check for existing labels
+        # add the triage label only if no label has been added
+        if: ${{ github.event.issue.labels[0] == null }}
+        run: gh issue edit "$NUMBER" --add-label "triage"
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          GH_REPO: ${{ github.repository }}
+          NUMBER: ${{ github.event.issue.number }}

cuda_core/docs/source/install.md

@@ -12,6 +12,8 @@ dependencies are as follows:
 
 [^1]: Including `cuda-python`.
 
+`cuda.core` supports Python 3.9 - 3.12, on Linux (x86-64, arm64) and Windows (x86-64).
+
 ## Installing from PyPI
 
 `cuda.core` works with `cuda.bindings` (part of `cuda-python`) 11 or 12. For example with CUDA 12:

cuda_core/docs/source/release/0.1.0-notes.md

@@ -1,4 +1,4 @@
-# `cuda.core` Release notes
+# `cuda.core` v0.1.0 Release notes
 
 Released on Nov 8, 2024
 

cuda_core/docs/source/release/0.1.1-notes.md

@@ -1,25 +1,28 @@
-# `cuda.core` Release notes
+# `cuda.core` v0.1.1 Release notes
 
-Released on Dec XX, 2024
+Released on Dec 20, 2024
 
 ## Hightlights
 
 - Add `StridedMemoryView` and `@args_viewable_as_strided_memory` that provide a concrete
   implementation of DLPack & CUDA Array Interface supports.
-- Add `Linker` that can link one or multiple `ObjectCode` instances generated by `Program`s. Under
-  the hood, it uses either the nvJitLink or cuLink APIs depending on the CUDA version detected
-  in the current environment.
-- Add a `cuda.core.experimental.system` module for querying system- or process- wide information.
-- Support TCC devices with a default synchronous memory resource to avoid the use of memory pools
+- Add `Linker` that can link one or multiple `ObjectCode` instances generated by `Program`. Under
+  the hood, it uses either the nvJitLink or driver (`cuLink*`) APIs depending on the CUDA version
+  detected in the current environment.
+- Support `pip install cuda-core`. Please see the Installation Guide for further details.
 
 ## New features
 
+- Add a `cuda.core.experimental.system` module for querying system- or process- wide information.
 - Add `LaunchConfig.cluster` to support thread block clusters on Hopper GPUs.
 
 ## Enchancements
 
-- Ensure "ltoir" is a valid code type to `ObjectCode`.
-- Improve test coverage.
+- The internal handle held by `ObjectCode` is now lazily initialized upon first touch.
+- Support TCC devices with a default synchronous memory resource to avoid the use of memory pools.
+- Ensure `"ltoir"` is a valid code type to `ObjectCode`.
+- Document the `__cuda_stream__` protocol.
+- Improve test coverage & documentation cross-references.
 - Enforce code formatting.
 
 ## Bug fixes
@@ -35,4 +38,6 @@ Released on Dec XX, 2024
   not supported. This will be fixed in a future release.
 - Some `LinkerOptions` are only available when using a modern version of CUDA. When using CUDA <12,
   the backend is the cuLink api which supports only a subset of the options that nvjitlink does.
-  Further, some options aren't available on CUDA versions <12.6
+  Further, some options aren't available on CUDA versions <12.6.
+- To use `cuda.core` with Python 3.13, it currently requires building `cuda-python` from source
+  prior to `pip install`. This extra step will be fixed soon.

cuda_core/examples/saxpy.py

@@ -47,8 +47,9 @@
 # prepare input/output
 size = cp.uint64(64)
 a = dtype(10)
-x = cp.random.random(size, dtype=dtype)
-y = cp.random.random(size, dtype=dtype)
+rng = cp.random.default_rng()
+x = rng.random(size, dtype=dtype)
+y = rng.random(size, dtype=dtype)
 out = cp.empty_like(x)
 dev.sync()  # cupy runs on a different stream from s, so sync before accessing
 
@@ -73,8 +74,8 @@
 # prepare input
 size = cp.uint64(128)
 a = dtype(42)
-x = cp.random.random(size, dtype=dtype)
-y = cp.random.random(size, dtype=dtype)
+x = rng.random(size, dtype=dtype)
+y = rng.random(size, dtype=dtype)
 dev.sync()
 
 # prepare output

cuda_core/examples/strided_memory_view.py

@@ -91,6 +91,7 @@
     gpu_prog = Program(gpu_code, code_type="c++")
     # To know the GPU's compute capability, we need to identify which GPU to use.
     dev = Device(0)
+    dev.set_current()
     arch = "".join(f"{i}" for i in dev.compute_capability)
     mod = gpu_prog.compile(
         target_type="cubin",
@@ -156,7 +157,6 @@ def my_func(arr, work_stream):
 
 # This takes the GPU path
 if cp:
-    dev.set_current()
     s = dev.create_stream()
     # Create input array on GPU
     arr_gpu = cp.ones(1024, dtype=cp.int32)

cuda_core/examples/vector_add.py

@@ -42,8 +42,9 @@
 
 # prepare input/output
 size = 50000
-a = cp.random.random(size, dtype=dtype)
-b = cp.random.random(size, dtype=dtype)
+rng = cp.random.default_rng()
+a = rng.random(size, dtype=dtype)
+b = rng.random(size, dtype=dtype)
 c = cp.empty_like(a)
 
 # cupy runs on a different stream from s, so sync before accessing

cuda_core/tests/requirements-cu11.txt

@@ -0,0 +1,3 @@
+pytest
+# TODO: remove this hack once cupy has a cp313 build
+cupy-cuda11x; python_version < "3.13"

cuda_core/tests/requirements-cu12.txt

@@ -0,0 +1,3 @@
+pytest
+# TODO: remove this hack once cupy has a cp313 build
+cupy-cuda12x; python_version < "3.13"