karrenberg / wfvLinks
IMPORTANT NOTICE: This implementation is long outdated. Whole-Function Vectorization is an algorithm that transforms a scalar function in such a way that it computes W executions of the original code in parallel using SIMD instructions (W is the target architecture's SIMD width).
☆22Updated 13 years ago
Alternatives and similar repositories for wfv
Users that are interested in wfv are comparing it to the libraries listed below
Sorting:
- RV: A Unified Region Vectorizer for LLVM☆112Updated 4 months ago
- LLVM AMDGPU Assembler Helper Tools☆112Updated 8 years ago
- Sample programs for the LLVM PTX back-end☆39Updated 10 years ago
- A framework that helps implementing swizzle GPU kernels☆42Updated 5 years ago
- ☆153Updated this week
- OpenCL/SPIR-V implementation of HIP☆105Updated 2 years ago
- ☆74Updated 2 years ago
- Enabling on-the-fly manipulations with LLVM IR code of CUDA sources☆112Updated 5 months ago
- ROCm - AMDGPU Compute Application Binary Interface☆41Updated 3 years ago
- ☆58Updated 3 weeks ago
- ☆70Updated 5 years ago
- Instruction THroughput Estimator using MAchine Learning (ITHEMAL)☆150Updated 3 years ago
- CudaPAD is a PTX/SASS viewer for NVIDIA Cuda kernels and provides an on-the-fly view of the assembly.☆123Updated 2 years ago
- GPUVerify: a Verifier for GPU Kernels☆67Updated 3 years ago
- a Halide language To MLIR compiler.☆26Updated 4 years ago
- An LLVM backend generating SPIR-V binary.☆89Updated 2 years ago
- A source-to-source compiler for automatic parallelization of C programs through code annotation.☆62Updated 5 years ago
- SYCL Conformance Tests☆69Updated last week
- A repository to test dialects defined dynamically.☆12Updated 2 years ago
- Benchmarks for auto-vectorization and revectorization, including both hand-vectorized and scalar code☆30Updated 6 years ago
- c++ posit implementation☆44Updated 2 years ago
- Decuda and cudasm, the CUDA binary utilities package. Low-level tools for NVidia G80 GPUs.☆103Updated 15 years ago
- Library to plot integer sets and maps☆53Updated 8 years ago
- A simple profiler to count Nvidia PTX assembly instructions of OpenCL/SYCL/CUDA kernels for roofline model analysis.☆56Updated 6 months ago
- ☆31Updated 2 years ago
- Generic system-wide modern C++ for heterogeneous platforms with SYCL from Khronos Group☆77Updated 4 years ago
- Haystack is an analytical cache model that given a program computes the number of cache misses.☆46Updated 6 years ago
- Polyhedral Parallel Code Generation (source repository: http://repo.or.cz/ppcg.git)☆128Updated 3 years ago
- A fast and highly scalable GPU dynamic memory allocator☆109Updated 10 years ago
- ☆141Updated last week