alpaka icon indicating copy to clipboard operation
alpaka copied to clipboard

Abstraction Library for Parallel Kernel Acceleration :llama:

Results 247 alpaka issues
Sort by recently updated
recently updated
newest added

We discussed today again in the alpaka meeting that it would be nice to have accelerator tags #1246 so that there is no need to use `#ifdef` to write traits...

Type:Refactoring

In a cmake based project that requires e.g. the CUDA backend, I tried to enable it like this: ```cmake find_package(alpaka 0.9.0 QUIET) if(NOT alpaka_FOUND) message(STATUS "No alpaka install found, disabling...

Type:Question
Type:CMake

Some simple clarifications for the novice alpaka user are added to the cheatsheet and library file. Alpaka vector is a static array, Buffer is a dynamic array. Pitch is the...

Type:Documentation

Workdiv creation has 2 template parameters. `auto const workdiv = alpaka::WorkDivMembers{arraySize / blockSize, (int)blockSize, 1u};` The aim is being able to call it without template parameters, for example `auto const...

The `explicit Vec(F&& generator, std::integer_sequence)` accepted a parameter pack of `TVal` which is defined to be the inner vector value and can possible be non-integral, which makes it invalid in...

Type:Bug

In the meeting on 25 May 2021 we discussed having an alpaka abstraction for the various tensor core APIs found in recent versions of CUDA and ROCm. Opening this issue...

Type:Enhancement
Backend:CUDA
Backend:HIP

An example: A 2D Convolutional filter applied to a matrix. The values of filter-matrix were initially kept in constant memory at the first commit. But due to Gitlab pipeline error...

Type:Example

for alpaka 1.1.0 ``` CMake Warning (dev) at ~/picongpu/thirdParty/cupla/alpaka/cmake/alpakaCommon.cmake:217 (find_package): Policy CMP0144 is not set: find_package uses upper-case _ROOT variables. Run "cmake --help-policy CMP0144" for policy details. Use the cmake_policy...

The `CudaVectorArrayWrapper` class used to provide `operator[]` and some other STL-like features to small CUDA vectors (like `uint4`, `float2`, etc.) It was used in GPU Philox implementations as storage for...

Type:Refactoring
Backend:CUDA
Backend:HIP

A simple PR. A directory called "_benchmarks_" is created and _babelstream_ example is copied into it. There is a new cmake flag _alpaka_BUILD_BENCHMARKS_. If this flag is ON then _alpaka_ACC_CPU_B_SEQ_T_SEQ_ENABLE_...