cuda.core v0.2.0

[leofang]

cuda.core v0.2.0 release announcement

Release note

• https://nvidia.github.io/cuda-python/cuda-core/latest/release/0.2.0-notes.html

All functionalities are currently hosted under the cuda.core.experimental namespace. Once the features become stable they will be moved out of experimental.

Documentation

• https://nvidia.github.io/cuda-python/cuda-core/0.2.0/

Sample codes

• https://github.com/NVIDIA/cuda-python/tree/cuda-core-v0.2.0/cuda_core/examples

Breaking Changes

• The stream attribute is removed from LaunchConfig. Instead, the Stream object should now be directly passed to launch as an argument.
• The signature for launch is changed by swapping positional arguments, the new signature is now (stream, config, kernel, *kernel_args)
• Change __cuda_stream__ from attribute to method.
• The Program.compile method no longer accepts the options argument. Instead, you can optionally pass an instance of ProgramOptions to the constructor of Program.
• Device.properties now provides attribute getters instead of a dictionary interface.
• The .handle attribute of various cuda.core objects now returns the underlying Python object instead of a (type-erased) Python integer.

Key Features and Enhancements

• Add ProgramOptions to facilitate the passing of runtime compile options to Program.
• Add pythonic access to Device and Kernel attributes.

For full details please refer to the release note above.

What's Changed

• Add back CuPy as an optional test dependency + Fix an example bug by @leofang in Add back CuPy as an optional test dependency + Fix an example bug #334
• add warning to the nvjitlink ctor when falling back to cuLink by @ksimpson-work in add warning to the nvjitlink ctor when falling back to cuLink #315
• Set up a preliminary doc build/publish pipeline by @leofang in Set up a preliminary doc build/publish pipeline  #325
• Fix doc ci permissions by @leofang in Fix doc ci permissions #338
• Add conda installation instructions by @leofang in Add conda installation instructions #321
• Add a dummy email address to the doc bot by @leofang in Add a dummy email address to the doc bot #343
• multi gpu example by @amwi04 in multi gpu example #304
• Change __cuda_stream__ from attribute to method by @NaderAlAwar in Change __cuda_stream__ from attribute to method #389
• Ensure deprecation warnings from cuda.bindings are swallowed by @ksimpson-work in Ensure deprecation warnings from cuda.bindings are swallowed #404
• Add the options data class to program by @ksimpson-work in Add the options data class to program #237
• Update to use the new NVKS runners by @leofang in Update to use the new NVKS runners #426
• Disable notebook execution for cuda.bindings by @vzhurba01 in Disable notebook execution for cuda.bindings #424
• Stop tracking cached files and Jupyter notebook hashes in doc builds by @vzhurba01 in Stop tracking cached files and Jupyter notebook hashes in doc builds #425
• Documentation remove gpu dependency by @ksimpson-work in Documentation remove gpu dependency #398
• handle CTK version specific options in the linker test by @ksimpson-work in handle CTK version specific options in the linker test #371
• support ptx code type for program by @ksimpson-work in support ptx code type for program #317
• Update release checklist to focus on subpackages by @vzhurba01 in Update release checklist to focus on subpackages #427
• Kernel attributes by @ksimpson-work in Kernel attributes #360
• Use cpu runner to build docs by @leofang in Use cpu runner to build docs #434
• device properties by @ksimpson-work in device properties #409
• Update linker options sequence handling by @ksimpson-work in Update linker options sequence handling #436
• Expose ObjectCode as public API + prune unnecessary input arguments by @ksimpson-work in Expose ObjectCode as public API + prune unnecessary input arguments #435
• Pin Sphinx <8.2.0 to fix doc build by @leofang in Pin Sphinx <8.2.0 to fix doc build #456
• CI: Add Windows GPU runner for tests by @leofang in CI: Add Windows GPU runner for tests #444
• Improve program checks by @ksimpson-work in Improve program checks #394
• Various handle-related changes and improvements by @leofang in Various handle-related changes and improvements #463
• Improve perf of accessing dev.compute_capability by @leofang in Improve perf of accessing dev.compute_capability #459
• Device properties example by @samaid in Device properties example #474
• Add ObjectCode ptx constructor by @brandon-b-miller in Add ObjectCode ptx constructor #470
• Add a Linker example by @vzhurba01 in Add a Linker example #475
• Add error log producing test by @ksimpson-work in Add error log producing test #423
• Apply __new__ approach to disabling __init__ by @rwgk in Apply __new__ approach to disabling __init__ #484
• switch the launch argument order by @ksimpson-work in switch the launch argument order #316
• NEW: Create an Event without recording to Stream by @carterbox in NEW: Create an Event without recording to Stream #487
• Clearer error messages (cuda.core) by @rwgk in Clearer error messages (cuda.core) #458
• Add event timing by @leofang in Add event timing #481
• Fix merge error by @leofang in Fix merge error #495
• add public handle to object code by @ksimpson-work in add public handle to object code #492
• Bump cuda-core ver to v0.2.0 by @leofang in Bump cuda-core ver to v0.2.0 #494
• Increase tolerance in test_timing() to avoid flaky tests. by @rwgk in Increase tolerance in test_timing() to avoid flaky tests. #498
• Fix test_timing flakiness under Windows by @rwgk in Fix test_timing flakiness under Windows #508
• NEW: Add Event to public API by @carterbox in NEW: Add Event to public API #501
• cuda.core: Change selected .decode() calls to .decode("utf-8", errors="backslashreplace") by @rwgk in cuda.core: Change selected .decode() calls to .decode("utf-8", errors="backslashreplace") #510
• clean up device initialization in test by @ksimpson-work in clean up device initialization in test #507
• Add @functools.lru_cache decorator for get_binding_version() by @rwgk in Add @functools.lru_cache decorator for get_binding_version() #512
• Fix dangling pointer problem in _linker.py by @rwgk in Fix dangling pointer problem in _linker.py #516
• cuda.core: release notes update by @rwgk in cuda.core: release notes update #519
• Set release date by @vzhurba01 in Set release date #523

New Contributors

• @amwi04 made their first contribution in multi gpu example #304
• @NaderAlAwar made their first contribution in Change __cuda_stream__ from attribute to method #389
• @samaid made their first contribution in Device properties example #474
• @brandon-b-miller made their first contribution in Add ObjectCode ptx constructor #470
• @carterbox made their first contribution in NEW: Create an Event without recording to Stream #487

Full Changelog: cuda-core-v0.1.1...cuda-core-v0.2.0

---

This discussion was created from the release cuda.core v0.2.0.