amd-rocm

amd-rocm

AMD ROCm — A software stack for general-purpose parallel computing on Instinct/Radeon GPUs. It covers kernel authoring via the HIP programming model and runtime API, libraries such as rocBLAS/MIOpen/RCCL, and tuning for CDNA/RDNA architectures.

Directory Structure

skills/amd-rocm/
  SKILL.md
  references/
    rocm-stack/
      README.md
      what-is-rocm.md
      core-sdk-overview.md
      math-compute-libraries.md
      communication-libraries.md
      runtime-and-compilers.md
      profiling-debugging-tools.md
      control-monitoring-tools.md
      media-storage-libraries.md
      install-linux.md
      install-deep-learning-frameworks.md
      install-docker-spack.md
      install-windows.md
      compatibility-and-release-history.md
      ai-ecosystem.md
      gpu-systems-infra.md
      toolkits.md
    hip/
      README.md
      programming-model.md
      hardware-implementation.md
      compilers-clr.md
      performance-guidelines.md
      cpp-language-extensions.md
      kernel-language-cpp-support.md
      hiprtc.md
      debugging-logging.md
      runtime-api-overview.md
      initialization-version.md
      device-management.md
      error-handling.md
      stream-management.md
      event-management.md
      asynchronous-execution.md
      execution-control-launch.md
      occupancy.md
      module-context-management.md
      memory-management.md
      host-memory.md
      device-memory.md
      unified-memory.md
      virtual-memory.md
      stream-ordered-allocator.md
      texture-surface.md
      graph-management.md
      cooperative-groups.md
      multi-device-peer-to-peer.md
      interop.md
      porting-cuda-to-hip.md
      cuda-to-hip-api-comparison.md
      hipify.md
      math-api.md
      env-variables.md
      deprecated-apis.md
      hardware-features.md
    gpu-arch/
      README.md
      gpu-architecture-overview.md
      gpu-arch-specs.md
      cdna-architecture.md
      rdna-architecture.md
      performance-optimization-guides.md
      mi100-microarchitecture.md
      mi250-microarchitecture.md
      mi300-microarchitecture.md
      precision-support.md
      mi300-mi200-performance-counters.md
      mi350-performance-counters.md
      isa-documentation.md
      machine-readable-isa.md
  samples/
    README.md
    vector-add-kernel.md
    explicit-memory-transfer.md
    shared-memory-tiled-matmul.md
    dynamic-shared-memory.md
    unified-memory.md
    stream-ordered-allocation.md
    stream-overlap.md
    hip-graph-stream-capture.md
    hip-graph-explicit-nodes.md
    cooperative-groups-reduction.md
    warp-shuffle-reduction.md
    multi-gpu-peer-to-peer.md
    occupancy-tuning.md
    hipify-cuda-to-hip.md
    event-timing.md
    device-query.md
    hiprtc-runtime-compilation.md
    rocblas-gemm.md
  scripts/
    README.md
    install.md
    hipcc-build.md
    hipify.md
    gpu-query.md
    profiling.md
    debug.md
    sanitize.md
    frameworks.md
    validation.md

Exploration Procedure

Derive the category from the task, and identify the target page in the category's README.md:

1. Find the category corresponding to the task in the mapping table below
2. of that category references/{category}/README.md（samples/ scripts/ (the README.md directly under it) to identify the target page
3. Read the .md of the relevant page to check the details

For kernel authoring in CUDA C++/Python, the PTX ISA instruction set, and GEMM implementation with CUTLASS/CuTe, refer to the nvidia-cuda skill. This skill covers the HIP programming model, runtime API, ROCm libraries, and CDNA/RDNA architectures. HIP has naming that is nearly 1:1 with the CUDA API (such ashipMalloc ↔ cudaMalloc ), and references/hip/cuda-to-hip-api-comparison.md と references/hip/porting-cuda-to-hip.md summarizes the correspondence with CUDA and migration procedures.

Task → Category Mapping

Task	Category	Reference README
Want to know the overall ROCm stack, the difference between Core SDK and Extras, and the TheRock build system	rocm-stack	references/rocm-stack/README.md
Want to look up numerical computation/communication libraries such as rocBLAS/MIOpen/RCCL, and profiling/monitoring tools (rocprofv3, amd-smi)	rocm-stack	references/rocm-stack/README.md
Want to know installation procedures for Linux/Windows, PyTorch/TensorFlow/JAX setup, and GPU/OS compatibility	rocm-stack	references/rocm-stack/README.md
Want to look up the HIP programming model, kernel authoring, C++ language extensions, and HIPRTC runtime compilation	hip	references/hip/README.md
Want to look up the HIP runtime API (device/stream/event/memory management, HIP Graph, Cooperative Groups)	hip	references/hip/README.md
Want to know about porting from CUDA to HIP, code conversion with hipify, and the CUDA↔HIP API correspondence table	hip	references/hip/README.md
Want to look up CDNA/RDNA architecture design, MI100/MI250/MI300 microarchitecture, and ISA/performance counters	gpu-arch	references/gpu-arch/README.md
Want to know typical usage (vector add, HIP Graph, rocBLAS GEMM, hipify conversion, etc.)	samples	samples/README.md
Want to know about installation, hipcc builds, rocminfo/amd-smi, rocprofv3 profiling, and rocgdb debugging commands	scripts	scripts/README.md