add missing data

fix review comments
ROCm · Sep 18, 2024 · 320ca57 · 320ca57
1 parent 298b308
commit 320ca57
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 22 deletions.
diff --git a/docs/reference/cpp_language_extensions.rst b/docs/reference/cpp_language_extensions.rst
@@ -399,6 +399,8 @@ To read a high-resolution timer from the device, HIP provides the following buil
 
   Note that ``clock()`` and ``clock64()`` do not work properly on AMD RDNA3 (GFX11) graphic processors.
 
+.. _atomic functions:
+
 Atomic functions
 ===============================================
 

diff --git a/docs/understand/hardware_capabilities.rst b/docs/understand/hardware_capabilities.rst
@@ -6,7 +6,7 @@
 Hardware features
 *******************************************************************************
 
-This page gives an overview of the different hardware architectures and the features they implement. Hardware features do not imply performance, that depends on the specifications found in the `Accelerator and GPU hardware specifications`_ page.
+This page gives an overview of the different hardware architectures and the features they implement. Hardware features do not imply performance, that depends on the specifications found in the :doc:`rocm:reference/gpu-arch-specs` page.
 
   .. list-table::
       :header-rows: 1
@@ -21,7 +21,7 @@ This page gives an overview of the different hardware architectures and the feat
         - RDNA3
         - CDNA3
       *
-        - Atomic functions on 32-bit integer values in global memory
+        - :ref:`atomic functions` on 32-bit integer values in global memory
         - ✅
         - ✅
         - ✅
@@ -69,7 +69,7 @@ This page gives an overview of the different hardware architectures and the feat
         - ✅
         - ✅
       *
-        - Error correcting code for SRAMs IP core
+        - SRAM error correcting code
         - ✅
         - ✅
         - ✅
@@ -109,7 +109,7 @@ This page gives an overview of the different hardware architectures and the feat
         - ✅
         - ✅
       *
-        - Support for float8 bfloat8
+        - Support for float8 bfloat8 (see: :doc:`rocm:compatibility/precision-support`)
         - ❌
         - ❌
         - ❌
@@ -182,25 +182,25 @@ This page gives an overview of the different hardware architectures and the feat
         - ?
       *
         - Maximum number of resident blocks per compute unit
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
+        - 40
+        - 32
+        - 32
+        - 32
+        - 32
+        - 32
       *
         - Maximum number of resident wavefronts per compute unit
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
+        - 40
+        - 32
+        - 32
+        - 32
+        - 32
+        - 32
       *
         - Maximum number of resident threads per compute unit
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
-        - ?
+        - 1280
+        - 2048
+        - 1024
+        - 2048
+        - 1024
+        - 2048