Improve docs of std headers (#3416)

NVIDIA · Jan 21, 2025 · 3f8e582 · 3f8e582
1 parent 4812f28
commit 3f8e582
Show file tree

Hide file tree

Showing 3 changed files with 27 additions and 2 deletions.
diff --git a/docs/libcudacxx/standard_api.rst b/docs/libcudacxx/standard_api.rst
@@ -14,6 +14,7 @@ Standard API
    standard_api/ranges_library
    standard_api/synchronization_library
    standard_api/time_library
+   standard_api/type_support
    standard_api/utility_library
 
 Standard Library Backports

diff --git a/docs/libcudacxx/standard_api/ranges_library.rst b/docs/libcudacxx/standard_api/ranges_library.rst
@@ -13,10 +13,10 @@ See the documentation of the standard headers `\<iterator\> <https://en.cpprefer
    * - Header
      - Content
      - Availability
-   * - <cuda/std/iterator>
+   * - `\<cuda/std/iterator\> <https://en.cppreference.com/w/cpp/header/iterator>`_
      - Iterator related concepts and machinery such as ``cuda::std::forward_iterator``
      - CCCL 2.3.0 / CUDA 12.4
-   * - <cuda/std/ranges>
+   * - `\<cuda/std/ranges\> <https://en.cppreference.com/w/cpp/header/ranges>`_
      - Range related concepts and machinery such as ``cuda::std::ranges::forward_range`` and ``cuda::std::ranges::subrange``
      - CCCL 2.4.0 / CUDA 12.5
 

diff --git a/docs/libcudacxx/standard_api/type_support.rst b/docs/libcudacxx/standard_api/type_support.rst
@@ -0,0 +1,24 @@
+.. _libcudacxx-standard-api-type-support:
+
+Type Support Library
+=======================
+
+.. toctree::
+   :hidden:
+   :maxdepth: 1
+
+Any Standard C++ header not listed below is omitted.
+
+.. list-table::
+   :widths: 25 45 30
+   :header-rows: 1
+
+   * - Header
+     - Content
+     - Availability
+   * - `\<cuda/std/climits\> <https://en.cppreference.com/w/cpp/header/climits>`_
+     - Limits of integral types
+     - libcu++ 1.0.0 / CCCL 2.0.0 / CUDA 10.2
+   * - `\<cuda/std/limits\> <https://en.cppreference.com/w/cpp/header/limits>`_
+     - Interface to query properties of all fundamental numeric types
+     - libcu++ 1.0.0 / CCCL 2.0.0 / CUDA 10.2