HIP: C++ Heterogeneous-Compute Interface for Portability
-
Updated
Oct 22, 2021 - C++
#
hip
Here are 56 public repositories matching this topic...
stdgpu: Efficient STL-like Data Structures on the GPU
cpp
gpu
modern-cpp
cpp14
openmp
cuda
stl
data-structures
gpgpu
gpu-acceleration
cpp17
stl-containers
hip
gpu-computing
rocm
cpp20
stl-like
-
Updated
Jul 26, 2021 - C++
Numerical linear algebra software package
-
Updated
Oct 23, 2021 - C++
Haskell Image Processing Library
-
Updated
Nov 21, 2020 - Haskell
Stretching GPU performance for GEMMs and tensor contractions.
python
machine-learning
amd
gpu
assembly
opencl
dnn
matrix-multiplication
neural-networks
gpu-acceleration
blas
hip
gpu-computing
tensors
tensor-contraction
gemm
radeon
auto-tuning
radeon-open-compute
-
Updated
Oct 23, 2021 - Python
An implementation of HIP that works on CPUs, across OSes.
cuda
cpp17
hip
spmd
stl-algorithms
parallel-algorithms
cuda-programming
hip-runtime
hip-kernel-language
hip-portability
-
Updated
Jul 14, 2021 - C++
The Arbor multi-compartment neural network simulation library.
-
Updated
Oct 23, 2021 - C++
benchmark
openmp
cuda
scientific-computing
test-driven-development
hip
gpu-computing
hpc-applications
sycl
-
Updated
Oct 22, 2021 - C++
Radeon Performance Primitives (RPP) library is a comprehensive high performance computer vision library for AMD (CPU and GPU) with HIP and OpenCL back-ends.
computer-vision
hpc
amd
gpu
histogram
contrast
bitwise
hip
rocm
comprehensive
openvx
rpp
mivisionx
radeon-performance-primitives
warp-affine
channel-extract
agumentation
-
Updated
Oct 21, 2021 - C++
Materials to reproduce the results of our paper about 3D muscle modelling.
geometry
scientific-visualization
biomechanics
hip
muscle
opensim
scientific-publications
joint-models
bioengineering
scientific-research
scientific-papers
muscle-model
-
Updated
Oct 5, 2021 - MATLAB
Portably Performant Physical Algebra
cmake
cpp
hpc
vector
cuda
simd
gpgpu
cpp17
hip
avx512
nvidia-cuda
amd-gpu
sandia-national-laboratories
cpp17-library
hpc-tools
snl-science-libs
-
Updated
Oct 19, 2021 - C++
-
Updated
Feb 25, 2021 - TypeScript
go bindings for hip
-
Updated
Oct 29, 2019 - Go
Improve this page
Add a description, image, and links to the hip topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the hip topic, visit your repo's landing page and select "manage topics."
Bug summary
There is evidence that
sub_group::get_group_id()does not return the same value asthreadIdx.x / warpSize(assuming 1D kernel), as expected on CUDA. We should check the implementation of this function. Our implementation of this function performs bit manipulation magic, presumably the optimization went to far...To Reproduce
Compare
sub_group{}.get_group_id()or `sub