rocWMMA User Guide# Contents: 1. Getting Started Guide for Linux 1.1. Introduction 1.1.1. Documentation Roadmap 1.2. Prerequisites 1.3. Installing pre-built packages 1.4. Building and Installing rocWMMA 1.4.1. System Requirements 1.4.2. Minimum GPU Requirements 1.4.3. Download rocWMMA 1.4.4. Build only library 1.4.5. Build library + samples 1.4.6. Build library + tests 1.4.7. Build library + Tests + Assembly 2. Introduction 2.1. Synchronous API 2.2. Supported Data Types 2.3. Supported Matrix Layouts 2.3.1. Using rocWMMA API 2.4. rocWMMA Datatypes matrix_a matrix_b accumulator row_major col_major VecT VectorStorage IOConfig IOShape 2.5. rocWMMA Enumeration layout_t layout_t::mem_row_major layout_t::mem_col_major 2.6. rocWMMA API functions fill_fragment() load_matrix_sync() load_matrix_sync() store_matrix_sync() store_matrix_sync() mma_sync() synchronize_workgroup() load_matrix_coop_sync() load_matrix_coop_sync() load_matrix_coop_sync() store_matrix_coop_sync() store_matrix_coop_sync() store_matrix_coop_sync() 3. Programmer’s Guide 3.1. Library Source Code Organization 3.1.1. The library directory 3.1.1.1. library/include/rocwmma/ 3.1.1.2. library/include/internal 3.1.2. The samples directory 3.1.2.1. samples/sgemmv.cpp 3.1.2.2. samples/simple_gemm.cpp 3.1.2.3. samples/simple_dlrm.cpp 3.1.2.4. samples/common.hpp 3.1.3. The test directory 3.1.3.1. test/bin 3.1.3.2. test/dlrm 3.1.3.3. test/gemm 3.1.3.4. test/unit 3.1.4. Infrastructure 4. Contributor’s Guide 4.1. License Agreement 4.2. Pull-request guidelines 4.3. StyleGuide 4.3.1. Interface 4.3.2. Philosophy 4.3.3. Implementation 4.3.4. Format