Merge pull request cupy#5391 from kmaehashi/update-packages

Add `cupy-cuda113` and `cupy-rocm-4-2` to README and docs

README.md

@@ -38,18 +38,19 @@ You can pass `ndarray` to existing CUDA C/C++ programs via [RawKernels](https://
 Wheels (precompiled binary packages) are available for Linux (x86_64) and Windows (amd64).
 Choose the right package for your platform.
 
-| Platform  | Command                        |
-| --------- | ------------------------------ |
-| CUDA 10.0 | `pip install cupy-cuda100`     |
-| CUDA 10.1 | `pip install cupy-cuda101`     |
-| CUDA 10.2 | `pip install cupy-cuda102`     |
-| CUDA 11.0 | `pip install cupy-cuda110`     |
-| CUDA 11.1 | `pip install cupy-cuda111`     |
-| CUDA 11.2 | `pip install cupy-cuda112`     |
-| CUDA 11.3 | `pip install cupy-cuda112` (*) |
-| ROCm 4.0  | `pip install cupy-rocm-4-0` (experimental; see [docs](https://docs.cupy.dev/en/latest/install.html#using-cupy-on-amd-gpu-experimental) for details) |
-
-*(\*) CUDA 11.3 support is currently under verification. Meanwhile you can use wheels for CUDA 11.2 which is compatible with CUDA 11.3.*
+| Platform      | Command                       |
+| ------------- | ----------------------------- |
+| CUDA 10.0     | `pip install cupy-cuda100`    |
+| CUDA 10.1     | `pip install cupy-cuda101`    |
+| CUDA 10.2     | `pip install cupy-cuda102`    |
+| CUDA 11.0     | `pip install cupy-cuda110`    |
+| CUDA 11.1     | `pip install cupy-cuda111`    |
+| CUDA 11.2     | `pip install cupy-cuda112`    |
+| CUDA 11.3     | `pip install cupy-cuda113`    |
+| ROCm 4.0 (*)  | `pip install cupy-rocm-4-0`   |
+| ROCm 4.2 (*)  | `pip install cupy-rocm-4-2`   |
+
+*(\*) ROCm support is an experimental feature. Refer to the [docs](https://docs.cupy.dev/en/latest/install.html#using-cupy-on-amd-gpu-experimental) for details.
 
 See the [Installation Guide](https://docs.cupy.dev/en/stable/install.html) if you are using Conda/Anaconda or building from source.
 

docs/source/install.rst

@@ -6,7 +6,7 @@ Requirements
 
 * `NVIDIA CUDA GPU <https://developer.nvidia.com/cuda-gpus>`_ with the Compute Capability 3.0 or larger.
 
-* `CUDA Toolkit <https://developer.nvidia.com/cuda-toolkit>`_: v10.0 / v10.1 / v10.2 / v11.0 / v11.1 / v11.2
+* `CUDA Toolkit <https://developer.nvidia.com/cuda-toolkit>`_: v10.0 / v10.1 / v10.2 / v11.0 / v11.1 / v11.2 / v11.3
 
     * If you have multiple versions of CUDA Toolkit installed, CuPy will automatically choose one of the CUDA installations.
       See :ref:`install_cuda` for details.
@@ -90,6 +90,8 @@ Package names are different depending on your CUDA Toolkit version.
      - ``$ pip install cupy-cuda111``
    * - v11.2
      - ``$ pip install cupy-cuda112``
+   * - v11.3
+     - ``$ pip install cupy-cuda113``
 
 .. note::