amdgpu_targets_gfx900
Vega GPU, codename vega10, including Radeon Vega Frontier Edition, Radeon RX Vega 56/64, Radeon RX Vega 64 Liquid, Radeon Pro Vega 48/56/64/64X, Radeon Pro WX 8200/9100, Radeon Pro V320/V340/SSG, Radeon Instinct MI25
Name | Category | Description |
---|---|---|
hipBLAS | sci-libs | ROCm BLAS marshalling library |
hipCUB | sci-libs | Wrapper of rocPRIM or CUB for GPU parallel primitives |
hipFFT | sci-libs | CU / ROCM agnostic hip FFT implementation |
hipRAND | sci-libs | CU / ROCM agnostic hip RAND implementation |
hipSPARSE | sci-libs | ROCm SPARSE marshalling library |
miopen | sci-libs | AMD's Machine Intelligence Library |
rccl | dev-libs | ROCm Communication Collectives Library (RCCL) |
rocBLAS | sci-libs | AMD's library for BLAS on ROCm |
rocFFT | sci-libs | Next generation FFT implementation for ROCm |
rocPRIM | sci-libs | HIP parallel primitives for developing performant GPU-accelerated code on ROCm |
rocRAND | sci-libs | Generate pseudo-random and quasi-random numbers |
rocSOLVER | sci-libs | Implementation of a subset of LAPACK functionality on the ROCm platform |
rocSPARSE | sci-libs | Basic Linear Algebra Subroutines for sparse computation |
rocThrust | sci-libs | HIP back-end for the parallel algorithm library Thrust |
roctracer | dev-util | Callback/Activity Library for Performance tracing AMD GPU's |
Tensile | dev-util | Stretching GPU performance for GEMMs and tensor contractions |