Skip to content

Commit a6387a0

Browse files
committed
add docs
1 parent 7066082 commit a6387a0

File tree

2 files changed

+14
-3
lines changed

2 files changed

+14
-3
lines changed

cuda_core/docs/source/contribute.rst

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
.. SPDX-License-Identifier: Apache-2.0
22
3+
.. contributer_guide
4+
35
Contributing
46
------------
57

cuda_core/docs/source/release/0.3.0-notes.rst

Lines changed: 12 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -7,10 +7,11 @@
77

88
Released on MM DD, 2025
99

10+
1011
Highlights
1112
----------
1213

13-
- Starting this release ``cuda.core`` is licensed under Apache 2.0.
14+
- Starting this release ``cuda.core`` is licensed under Apache 2.0. The biggest implication of this change is that we are open to external contribution now! Please kindly follow the :ref:`Contributor Guide <contributer_guide>` for detailed instructions.
1415

1516

1617
Breaking Changes
@@ -23,17 +24,25 @@ New features
2324
- :class:`Kernel` adds :attr:`Kernel.num_arguments` and :attr:`Kernel.arguments_info` for introspection of kernel arguments. (#612)
2425
- Add pythonic access to kernel occupancy calculation functions via :attr:`Kernel.occupancy`. (#648)
2526
- Support launching cooperative kernels by setting :attr:`LaunchConfig.cooperative_launch` to `True`.
26-
- A name can be assigned to :class:`ObjectCode` instances generated by both :class:`Program` and :class:`Linker` through their respective
27-
options.
27+
- A name can be assigned to :class:`ObjectCode` instances generated by both :class:`Program` and :class:`Linker` through their respective options.
28+
- Expose :class:`Buffer`, :class:`DeviceMemoryResource`, :class:`LegacyPinnedMemoryResource`, and :class:`MemoryResource` to the top namespace.
2829

2930

3031
New examples
3132
------------
3233

34+
- Add a PyTorch-based example.
35+
- Split the :class:`StridedMemoryView` example into two (CPU/GPU).
36+
3337

3438
Fixes and enhancements
3539
----------------------
3640

41+
- ``cuda.core`` now raises more clear and actionable error messages whenever possible.
42+
- :class:`ObjectCode` can be pickled now.
3743
- Look-up of the :attr:`Event.device` and :attr:`Event.context` (the device and CUDA context where an event was created from) is now possible.
44+
- :class:`Event`-based timing is made more robust (also with better error messages).
3845
- The :func:`launch` function's handling of fp16 scalars was incorrect and is fixed.
46+
- :attr:`ProgramOptions.ptxas_options` can now accept more than one argument.
3947
- The :class:`Device` constructor is made faster.
48+
- The CFFI-based example no longer leaves the intermediate files on disk after it finishes.

0 commit comments

Comments
 (0)