* update docker images and CI to ROCm 6.2.0
* fix rocfft brick bounds parameter order
* update formatting for rocm 6.2
---------
Co-authored-by: Robin Voetter <robin@streamhpc.com>
* Update dockerfiles to rocm 6.0
* fix(opengl_interop): include 'hip_gl_interop.h' as 'hipGLGetDevices' and related have been moved to a different header
* fix(bsrmv,spmv): use new rocSPARSE 3.x api with compatibility for rocSPARSE 2.x
* fix: use '__HIP_PLATFORM_NVIDIA__' instead of '__HIP_PLATFORM_NVCC__'
* Define __HIP_PLATFORM_X__ properly
* CMake config command in CI
* Fix docker container to build hipRAND from source
* Upgrade cuda version to 12 in hip-libraries-cuda-ubuntu.Dockerfile
* Remove -Werror --Werror all-warnings from CUDA_FLAGS
* Resolve "Add back warning for CUDA"
* fix(rocsparse_utils): use new status only with rocSPARSE 3.0+
---------
Co-authored-by: Nick Breed <nick@streamhpc.com>
Co-authored-by: Nara Prasetya <nara@streamhpc.com>
Co-authored-by: Mátyás Aradi <matyas@streamhpc.com>
* Re-enable the hello-world example on Windows with CMake
* Update ROCm to 5.7
* Small changes to make the pipeline pass with rocm 5.7 on Nvidia platform
* Re-enable the rocsparse tests for cmake on windows
* Resolve "windows test timeouts"
* Resolve "hip_multi_gpu_data_transfer fails validation."
* Remove unused includes
* Resolve "NVCC VS 2019 Release runtime failure"
* Resolve "NVCC: hip_runtime_compilation failure"
* Resolve "Update install instruction to reflect optional dependencies of certain libraries"
* Resolve "HIP texture management example not working on Windows"
* Print version
* Install ROCm on nvidia via rocm-core and hipcc
* Resolve "Use ceiling_div wherever possible"
* Resolve "Investigate and replace setprecision with a better solution in examples"
* Disable execution of failing examples in VS
* fix(rocsparse_utils): remove rocSPARSE 3.0 status
---------
Co-authored-by: Gergely Mészáros <gergely@streamhpc.com>
Co-authored-by: Mátyás Aradi <matyas@streamhpc.com>
Co-authored-by: Balint Soproni <balint@streamhpc.com>
Co-authored-by: Bence Parajdi <bence@streamhpc.com>
Co-authored-by: Robin Voetter <robin@streamhpc.com>
Co-authored-by: Jaap Blok <jaap@streamhpc.com>
Co-authored-by: Nick Breed <nick@streamhpc.com>
* bump the required cmake version to 3.21.3
* Fix device_globals example name
* Fix hip_streams timeout on AMD windows debug build type
* Update templates
* Update cuda container to ROCm 5.4
* Change std::bind into lambda
* HIP 5.5 fixes
* fix tests not being executed
* Make the reference to the identity and transpose op uniform
* Fix NVCC CI
* Resolve "Increase timeout for CI"
* Update fixed size arrays to C++ standards
* Add missing include in hip_texture_management
* Remove void** cast from hipMalloc
* Fix hip-libraries-cuda-ubuntu Dockerfile
* Make the windows builds less verbose
* Rework Windows CI
* Skip failing rocsparse tests
* Fix cooperative groups example
* ci: Make skipped examples more prominent in windows VS test runner
* Enable rocsparse examples in CI
* Update .gitlab/issue_templates/example.md
Fix small typo
---------
Co-authored-by: Balint Soproni <balint@streamhpc.com>
Co-authored-by: Robin Voetter <robin@streamhpc.com>
Co-authored-by: Nara Prasetya <nara@streamhpc.com>
Co-authored-by: Nol Moonen <nol@streamhpc.com>
Co-authored-by: Mátyás Aradi <matyas@streamhpc.com>
Co-authored-by: Gergely Mészáros <gergely@streamhpc.com>
Co-authored-by: Sam Wu <22262939+samjwu@users.noreply.github.com>
* Applications example Monte Carlo pi
* Added prefix sum application example
* Improve Floyd-Warshall example
- Use Pinned memory for host-device copy
- Use device memory for the computation
- Update application README
- Remove void** cast from hipMalloc
* Added Histogram example
* Add applications/prefix_sum to portable examples
* Implement the convolution example
* Bitonic sort example
---------
Co-authored-by: Nol Moonen <nol@streamhpc.com>
Co-authored-by: Nara Prasetya <nara@streamhpc.com>
Co-authored-by: Balint Soproni <balint@streamhpc.com>
Co-authored-by: Robin Voetter <robin@streamhpc.com>
* hiprtc: vs: copy additional libraries to output directory
Besides hiprtc*.dll, hiprtc-builtins*.dll and amd_comgr*.dll is also required.
In this commit, the PostBuildAction is also replaced by a Content statement. This
provides more robustness in the build system, as using this method there will be
no race conditions if two targets copy the same file to the output directory.
* Resolve "Multi-GPU error occured while hip_opengl_interop sample application is executed"
* vs: copy dynamic libraries for other examples as well
If a library links with a dynamic library, the corresponding dll file
should be placed in the output directory so that the library can be found
even if the ROCm binary path is not added to the system PATH. The solution
used is the same as that used for the HIPRTC example.
* cmake: disable static_host_library on windows
The HIP SDK does at this time not ship with a binary for CMAKE_AR, which is required
to compile this example.
* ci: do not rely on PATH
Removing the ROCm binary path from the global PATH allows us to check
that the examples do not depend on this. The CMakeFile for each example
does currently not copy the libraries to the build directory, and so
for running the tests using CTest the library path is still required.
* Resolve "cmake: windows + nvidia does not pass hip includes to include list"
* Resolve "cmake: make examples use install()"
* Resolve "hipBLAS examples' Makefiles do not work with CUDA"
* cuda dockerfile: replace ARG with GID
* Resolve "Windows CI with Nvidia"
* Fix VS copied dependencies for HIP
---------
Co-authored-by: Nara Prasetya <nara@streamhpc.com>
Co-authored-by: Máté Ferenc Nagy-Egri <mate@streamhpc.com>
* Resolve "hipBLAS examples are not executed in the CI on Nvidia"
* Resolve "Use chevron launch syntax instead of hipLaunchKernelGGL"
* Resolve "VS2017 and VS2022 files for all examples"
* Resolve "NAVI30 Error occured while hip_module_api sample application is executed"
* Resolve "Enable examples disabled VS"
* Resolve ""hiprtc0504.dll" is missing while running "hip_runtime_compilation_vs2019.exe""
* Resolve "opengl/vulkan: GLFW dependency via GLFW_DIR"
* Resolve "Dedicated ROCm-Examples-VS20xx.sln solution files tagged with "HIP_nvcc" should come with only a selection of examples that can be run on non-AMD hardware."
* Resolve "[EXSWSTRHPC-177] Remove hip_cooperative_groups from sln files"
* PropertyGroup just after ProjectConfigurations ItemGroup
* Literal version numbers instead of MSBuildVersion
* Removed new line at the beginning of solution files
* Changed -Wno-#warnings to --disable-warnings for nvcc in simple_distributions_cpp_vs20XX.vcxproj
---------
Co-authored-by: Nara Prasetya <nara@streamhpc.com>
Co-authored-by: Robin Voetter <robin@streamhpc.com>
* Resolve "Set Occupancy and Stream examples as C++ 17 standard in Windows Visual Studio"
* Resolve "Fix Bug in Makefile of rocThrust vectors, remove_points and saxpy"
* Fixing rebasing mistake in HIP-Basic events
* Resolve "Port the command line parsing of rocRAND example and matrix multiplication to the common cmdparser.hpp"
* Resolve "Bandwidth example"
* Add required vulkan packages to rocm dockerfile
* Fix the build type variable name in the cmake build on windows
* increase CI docker build timeout
* Add required vulkan packages to cuda dockerfile
* fix vs build files for device query, matrix multiplication, device sum
* fix target names in vcxproj
* Runtime compilation example
* Remove gfx90c target from saxpy
* Fix README.md titles and typos in examples
* Shared memory example
* update docker files to rocm 5.3
* Resolve "2D Shuffle example"
* Assembly to Executable
* Resolve "LLVM IR to executable example"
* Reordering some statements
* update copyright of llvm ir to executable/assembly to executable
* Resolve "Upstreaming examples fixes"
Co-authored-by: Bibrak Qamar <bibrak@streamhpc.com>
Co-authored-by: Lőrinc Serfőző <lorinc@streamhpc.com>
Co-authored-by: Gergely Mészáros <gergely@streamhpc.com>
Co-authored-by: Beatriz Navidad Vilches <beatriz@streamhpc.com>
Co-authored-by: Saad Rahim <44449863+saadrahim@users.noreply.github.com>
Beatriz Navidad Vilches
Beatriz Navidad Vilches
David Galiffi
Lucas Santiago
Robin Voetter
Karthik Kumar Viswanathan
Lőrinc Serfőző