JinlongLi

hip-runtime-nvidia : Depends: cuda (>= 7.5) but it is not installable

4

I need to install hip-runtime-nvidia on NVIDIA T600 GPU ,but it failed $ nvcc -V nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2022 NVIDIA Corporation Built on Tue_May__3_18:49:52_PDT_2022 Cuda...

ld.lld: error: unable to find library -lamdhip64

3

I am compiling code converted from CUDA by Hipify-perl with Rocm-5.0.0， My CMakeLists.txt about HIP lies here: PROJECT(MultiEuler LANGUAGES CXX) set(CMAKE_MODULE_PATH "/opt/rocm/hip/cmake" ${CMAKE_MODULE_PATH}) find_package(HIP) find_package(MPI) option(USE_DOUBLE "if double" ON) option(ENABLE_HIP...

Kernel attributes and __launch__bounds__ Feature

6

**Describe the motivation for the feature request** As shown below: 1. how do i map __launch_bounds__ to AdaptiveCpp interfaces for HIP Code Optimization ? 2. It's enbaled to check registers...

Implementation issue of Reduction upon --acpp-targets='generic'

4

#define sycl_reduction_plus(argus) sycl::reduction(&(argus), sycl::plus()) #define sycl_reduction_max(argus) sycl::reduction(&(argus), sycl::maximum()) #define sycl_reduction_min(argus) sycl::reduction(&(argus), sycl::minimum()) the reduction expression located at: /XFLUIDS/src/solver_GetDt/GlobalDt_block.hpp:33 q.submit([&](sycl::handler &h) { auto reduction_max_x = sycl_reduction_max(uvw_c_max[0]); h.parallel_for(sycl::nd_range(global_ndrange, local_ndrange), reduction_max_x, [=](nd_item index,...

More GPU Performance Optimization

2

# A list of GPUs has been tested by the following Kernel, it's part of our opensource CFD framwork [XFLUIDS](https://github.com/XFluids/XFluids): ## llvm-16.0.6 is implemented on A100, amd Rocm-5.4.1 on RX6800XT,...