You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 2f5339b
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: README.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,8 +5,8 @@ CUDA Python is the home for accessing NVIDIA’s CUDA platform from Python. It c
5
5
*[cuda.core](https://nvidia.github.io/cuda-python/cuda-core/latest): Pythonic access to CUDA Runtime and other core functionalities
6
6
*[cuda.bindings](https://nvidia.github.io/cuda-python/cuda-bindings/latest): Low-level Python bindings to CUDA C APIs
7
7
*[cuda.pathfinder](https://nvidia.github.io/cuda-python/cuda-pathfinder/latest): Utilities for locating CUDA components installed in the user's Python environment
8
-
*[cuda.coop](https://nvidia.github.io/cccl/python/cooperative): A Python module providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
9
-
*[cuda.compute](https://nvidia.github.io/cccl/python/parallel): A Python module for easy access to CCCL's highly efficient and customizable parallel algorithms, like `sort`, `scan`, `reduce`, `transform`, etc. that are callable on the *host*
8
+
*[cuda.coop](https://nvidia.github.io/cccl/python/coop): A Python module providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
9
+
*[cuda.compute](https://nvidia.github.io/cccl/python/compute): A Python module for easy access to CCCL's highly efficient and customizable parallel algorithms, like `sort`, `scan`, `reduce`, `transform`, etc. that are callable on the *host*
10
10
*[numba.cuda](https://nvidia.github.io/numba-cuda/): A Python DSL that exposes CUDA **SIMT** programming model and compiles a restricted subset of Python code into CUDA kernels and device functions
11
11
*[cuda.tile](https://docs.nvidia.com/cuda/cutile-python/): A new Python DSL that exposes CUDA **Tile** programming model and allows users to write NumPy-like code in CUDA kernels
12
12
*[nvmath-python](https://docs.nvidia.com/cuda/nvmath-python/latest): Pythonic access to NVIDIA CPU & GPU Math Libraries, with [*host*](https://docs.nvidia.com/cuda/nvmath-python/latest/overview.html#host-apis), [*device*](https://docs.nvidia.com/cuda/nvmath-python/latest/overview.html#device-apis), and distributed [*distributed*](https://docs.nvidia.com/cuda/nvmath-python/latest/distributed-apis/index.html) APIs. It also provides low-level Python bindings to host C APIs ([nvmath.bindings](https://docs.nvidia.com/cuda/nvmath-python/latest/bindings/index.html)).
Copy file name to clipboardExpand all lines: cuda_python/DESCRIPTION.rst
+15-11Lines changed: 15 additions & 11 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -9,27 +9,31 @@ CUDA Python is the home for accessing NVIDIA's CUDA platform from Python. It con
9
9
10
10
* `cuda.core <https://nvidia.github.io/cuda-python/cuda-core/latest>`_: Pythonic access to CUDA Runtime and other core functionalities
11
11
* `cuda.bindings <https://nvidia.github.io/cuda-python/cuda-bindings/latest>`_: Low-level Python bindings to CUDA C APIs
12
-
* `cuda.cooperative <https://nvidia.github.io/cccl/python/cooperative>`_: A Python package providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
13
-
* `cuda.parallel <https://nvidia.github.io/cccl/python/parallel>`_: A Python package for easy access to CCCL's highly efficient and customizable parallel algorithms, like ``sort``, ``scan``, ``reduce``, ``transform``, etc, that are callable on the *host*
14
-
* `numba.cuda <https://nvidia.github.io/numba-cuda/>`_: Numba's target for CUDA GPU programming by directly compiling a restricted subset of Python code into CUDA kernels and device functions following the CUDA execution model.
15
-
16
-
For access to NVIDIA CPU & GPU Math Libraries, please refer to `nvmath-python <https://docs.nvidia.com/cuda/nvmath-python/latest>`_.
17
-
18
-
CUDA Python is currently undergoing an overhaul to improve existing and bring up new components. All of the previously available functionalities from the ``cuda-python`` package will continue to be available, please refer to the `cuda.bindings <https://nvidia.github.io/cuda-python/cuda-bindings/latest>`_ documentation for installation guide and further detail.
12
+
* `cuda.pathfinder <https://nvidia.github.io/cuda-python/cuda-pathfinder/latest>`_: Utilities for locating CUDA components installed in the user's Python environment
13
+
* `cuda.coop <https://nvidia.github.io/cccl/python/coop>`_: A Python module providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
14
+
* `cuda.compute <https://nvidia.github.io/cccl/python/compute>`_: A Python module for easy access to CCCL's highly efficient and customizable parallel algorithms, like ``sort``, ``scan``, ``reduce``, ``transform``, etc. that are callable on the *host*
15
+
* `numba.cuda <https://nvidia.github.io/numba-cuda/>`_: A Python DSL that exposes CUDA **SIMT** programming model and compiles a restricted subset of Python code into CUDA kernels and device functions
16
+
* `cuda.tile <https://docs.nvidia.com/cuda/cutile-python/>`_: A new Python DSL that exposes CUDA **Tile** programming model and allows users to write NumPy-like code in CUDA kernels
17
+
* `nvmath-python <https://docs.nvidia.com/cuda/nvmath-python/latest>`_: Pythonic access to NVIDIA CPU & GPU Math Libraries, with `host <https://docs.nvidia.com/cuda/nvmath-python/latest/overview.html#host-apis>`_, `device <https://docs.nvidia.com/cuda/nvmath-python/latest/overview.html#device-apis>`_, and distributed `distributed <https://docs.nvidia.com/cuda/nvmath-python/latest/distributed-apis/index.html>`_ APIs. It also provides low-level Python bindings to host C APIs (`nvmath.bindings <https://docs.nvidia.com/cuda/nvmath-python/latest/bindings/index.html>`_).
18
+
* `nvshmem4py <https://docs.nvidia.com/nvshmem/api/api/language_bindings/python/index.html>`_: Pythonic interface to the NVSHMEM library, enabling Python applications to leverage NVSHMEM's high-performance PGAS (Partitioned Global Address Space) programming model for GPU-accelerated computing
19
+
* `NSIGHT Python <https://docs.nvidia.com/nsight-python/index.html>`_: Python kernel profiling interface that automates performance analysis across multiple kernel configurations using NVIDIA Nsight Tools
20
+
* `CUPTI Python <https://docs.nvidia.com/cupti-python/>`_: Python APIs for creation of profiling tools that target CUDA Python applications via the CUDA Profiling Tools Interface (CUPTI)
21
+
22
+
CUDA Python is currently undergoing an overhaul to improve existing and introduce new components. All of the previously available functionalities from the ``cuda-python`` package will continue to be available, please refer to the `cuda.bindings <https://nvidia.github.io/cuda-python/cuda-bindings/latest>`_ documentation for installation guide and further detail.
19
23
20
24
cuda-python as a metapackage
21
25
============================
22
26
23
-
``cuda-python`` is now a metapackage that contains a collection of subpackages. Each subpackage is versioned independently, allowing installation of each component as needed.
27
+
``cuda-python`` is being restructured to become a metapackage that contains a collection of subpackages. Each subpackage is versioned independently, allowing installation of each component as needed.
24
28
25
29
Subpackage: cuda.core
26
30
---------------------
27
31
28
-
The ``cuda.core`` package offers idiomatic, pythonic access to CUDA Runtime and other functionalities.
32
+
The ``cuda.core`` package offers idiomatic, Pythonic access to CUDA Runtime and other functionalities.
29
33
30
34
The goals are to
31
35
32
-
1. Provide **idiomatic ("pythonic")** access to CUDA Driver, Runtime, and JIT compiler toolchain
36
+
1. Provide **idiomatic ("Pythonic")** access to CUDA Driver, Runtime, and JIT compiler toolchain
33
37
2. Focus on **developer productivity** by ensuring end-to-end CUDA development can be performed quickly and entirely in Python
34
38
3. **Avoid homegrown** Python abstractions for CUDA for new Python GPU libraries starting from scratch
35
39
4. **Ease** developer **burden of maintaining** and catching up with latest CUDA features
@@ -40,7 +44,7 @@ Subpackage: cuda.bindings
40
44
41
45
The ``cuda.bindings`` package is a standard set of low-level interfaces, providing full coverage of and access to the CUDA host APIs from Python.
Copy file name to clipboardExpand all lines: cuda_python/docs/source/index.rst
+27-10Lines changed: 27 additions & 10 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -7,19 +7,32 @@ CUDA Python
7
7
CUDA Python is the home for accessing NVIDIA's CUDA platform from Python. It consists of
8
8
multiple components:
9
9
10
-
- `cuda.core`_: Pythonic access to CUDA runtime and other core functionalities
10
+
- `cuda.core`_: Pythonic access to CUDA Runtime and other core functionalities
11
11
- `cuda.bindings`_: Low-level Python bindings to CUDA C APIs
12
12
- `cuda.pathfinder`_: Utilities for locating CUDA components installed in the user's Python environment
13
-
- `cuda.cccl.cooperative`_: A Python module providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
14
-
- `cuda.cccl.parallel`_: A Python module for easy access to CCCL's highly efficient and customizable parallel algorithms, like ``sort``, ``scan``, ``reduce``, ``transform``, etc, that are callable on the *host*
15
-
- `numba.cuda`_: Numba's target for CUDA GPU programming by directly compiling a restricted subset of Python code into CUDA kernels and device functions following the CUDA execution model.
16
-
* `nvmath-python`_: Pythonic access to NVIDIA CPU & GPU Math Libraries, with both *host* and *device* (through `nvmath.device`_) APIs. It also provides low-level Python bindings to host C APIs (through `nvmath.bindings`_).
17
-
13
+
- `cuda.coop`_: A Python module providing CCCL's reusable block-wide and warp-wide *device* primitives for use within Numba CUDA kernels
14
+
- `cuda.compute`_: A Python module for easy access to CCCL's highly efficient and customizable parallel algorithms, like ``sort``, ``scan``, ``reduce``, ``transform``, etc. that are callable on the *host*
15
+
- `numba.cuda`_: A Python DSL that exposes CUDA **SIMT** programming model and compiles a restricted subset of Python code into CUDA kernels and device functions
16
+
- `cuda.tile`_: A new Python DSL that exposes CUDA **Tile** programming model and allows users to write NumPy-like code in CUDA kernels
17
+
- `nvmath-python`_: Pythonic access to NVIDIA CPU & GPU Math Libraries, with `host`_, `device`_, and distributed `distributed`_ APIs. It also provides low-level Python bindings to host C APIs (`nvmath.bindings`_).
18
+
- `nvshmem4py`_: Pythonic interface to the NVSHMEM library, enabling Python applications to leverage NVSHMEM's high-performance PGAS (Partitioned Global Address Space) programming model for GPU-accelerated computing
19
+
- `NSIGHT Python`_: Python kernel profiling interface that automates performance analysis across multiple kernel configurations using NVIDIA Nsight Tools
20
+
- `CUPTI Python`_: Python APIs for creation of profiling tools that target CUDA Python applications via the CUDA Profiling Tools Interface (CUPTI)
0 commit comments