diff --git a/projects/rocblas/docs/how-to/Programmers_Guide.rst b/projects/rocblas/docs/how-to/Programmers_Guide.rst index c72825b90721..8b4538033157 100644 --- a/projects/rocblas/docs/how-to/Programmers_Guide.rst +++ b/projects/rocblas/docs/how-to/Programmers_Guide.rst @@ -424,7 +424,7 @@ rocblas_sizeof_datatype function size_t rocblas_sizeof_datatype(rocblas_datatype type) -Returns the size of a rocBLAS runtime data type. +Returns the size of a rocBLAS data type. Answering device memory size queries in functions that do not need memory diff --git a/projects/rocblas/docs/index.rst b/projects/rocblas/docs/index.rst index 03e883fc07d0..6ad97ec1079c 100644 --- a/projects/rocblas/docs/index.rst +++ b/projects/rocblas/docs/index.rst @@ -8,7 +8,7 @@ rocBLAS documentation ******************************************************************** -rocBLAS is the ROCm Basic Linear Algebra Subprograms (BLAS) library. rocBLAS is implemented in the :doc:`HIP programming language ` and optimized for AMD GPUs. +rocBLAS is the ROCm Basic Linear Algebra Subprograms (BLAS) library. rocBLAS is implemented in :doc:`HIP C++ ` and optimized for AMD GPUs. This documentation set contains instructions for installing, understanding, and using the rocBLAS library. To learn more, see :doc:`./what-is-rocblas` diff --git a/projects/rocblas/docs/what-is-rocblas.rst b/projects/rocblas/docs/what-is-rocblas.rst index c4f54ec9e6d4..e6ab44686050 100644 --- a/projects/rocblas/docs/what-is-rocblas.rst +++ b/projects/rocblas/docs/what-is-rocblas.rst @@ -9,7 +9,7 @@ What is rocBLAS? ******************************************************************** rocBLAS is the AMD library for Basic Linear Algebra Subprograms (BLAS) on the :doc:`ROCm platform `. -It is implemented in the :doc:`HIP programming language ` and optimized for AMD GPUs. +It is implemented in the :doc:`HIP C++ ` and optimized for AMD GPUs. For more detailed component information, see :doc:`rocBLAS design notes <../conceptual/rocblas-design-notes>`. rocBLAS aims to provide the following: @@ -17,7 +17,7 @@ rocBLAS aims to provide the following: * Functionality similar to legacy BLAS, adapted to run on AMD GPUs * A high-performance, robust implementation -rocBLAS is written in C++17 and HIP and uses the AMD ROCm runtime. +rocBLAS is written in HIP C++ (using C++17 features) and it uses the HIP runtime. The rocBLAS API is a thin C99 API that uses the hourglass pattern. It contains: