Skip to content

Commit

Permalink
cuda.core v0.1.1 final doc touch (#301)
Browse files Browse the repository at this point in the history
* minor changes to the release note

* add missing entries

* Update 0.1.1-notes.md

* update

* address comments

* mention Python/OS support matrix
  • Loading branch information
leofang authored Dec 20, 2024
1 parent 33b7366 commit 4135a2f
Show file tree
Hide file tree
Showing 3 changed files with 18 additions and 11 deletions.
2 changes: 2 additions & 0 deletions cuda_core/docs/source/install.md
Original file line number Diff line number Diff line change
Expand Up @@ -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:
Expand Down
2 changes: 1 addition & 1 deletion cuda_core/docs/source/release/0.1.0-notes.md
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# `cuda.core` Release notes
# `cuda.core` v0.1.0 Release notes

Released on Nov 8, 2024

Expand Down
25 changes: 15 additions & 10 deletions cuda_core/docs/source/release/0.1.1-notes.md
Original file line number Diff line number Diff line change
@@ -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
Expand All @@ -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.

0 comments on commit 4135a2f

Please sign in to comment.