Skip to content
/ HIPIFY Public
forked from ROCm/HIPIFY

HIPIFY: Convert CUDA to Portable C++ Code

1 star 75 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

emankov/HIPIFY

BranchesTags
 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1,104 Commits
bin
bin
 
 
packaging
packaging
 
 
patches
patches
 
 
src
src
 
 
tests
tests
 
 
.gitignore
.gitignore
 
 
CMakeLists.txt
CMakeLists.txt
 
 
README.md
README.md
 
 

Repository files navigation

HIPIFY

Tools to translate CUDA source code into portable HIP C++ automatically

Table of Contents

hipify-clang

hipify-clang is a clang-based tool for translation CUDA sources into HIP sources. It translates CUDA source into an abstract syntax tree, which is being traversed by transformation matchers. After applying all the matchers, the output HIP source is produced.

Advantages:

  1. It is a translator; thus, any even very complicated constructs will be parsed successfully, or an error will be reported.
  2. It supports clang options like -I, -D, --cuda-path, etc.
  3. Seamless support of new CUDA versions as it is clang's responsibility.
  4. Ease in support.

Disadvantages:

  1. The main advantage is also the main disadvantage: the input CUDA code should be correct; incorrect code wouldn't be translated to HIP.
  2. CUDA should be installed and provided in case of multiple installations by --cuda-path option.
  3. All the includes and defines should be provided to transform code successfully.

hipify-clang: dependencies

hipify-clang requires:

  1. LLVM+CLANG of at least version 3.8.0; the latest stable and recommended release: 11.0.0.

  2. CUDA of at least version 7.0, the latest supported version is 11.0.

LLVM release version CUDA latest supported version Windows Linux
3.8.0*, 3.8.1*,
3.9.0*, 3.9.1*		 7.5 + +
4.0.0, 4.0.1,
5.0.0, 5.0.1, 5.0.2		 8.0 + +
6.0.0, 6.0.1 9.0 + +
7.0.0, 7.0.1, 7.1.0 9.2 works only with the patch
due to the clang's bug 38811
patch for 7.0.0**
patch for 7.0.1**
patch for 7.1.0**
 -
not working due to
the clang's bug 36384
8.0.0, 8.0.1 10.0 works only with the patch
due to the clang's bug 38811
patch for 8.0.0**
patch for 8.0.1**
 +
9.0.0, 9.0.1 10.1 + +
10.0.0, 10.0.1 11.0 + +
11.0 Update 1, 11.1 works only with the patch
due to the clang's bug 47332
patch for 10.0.0***
patch for 10.0.1***
11.0.0 11.0 LATEST STABLE CONFIG
11.0 Update 1, 11.1 works only with the patch
due to the clang's bug 47332
patch for 11.0.0***

* LLVM 3.x is not supported anymore but might still work.

** Download the patch and unpack it into your LLVM distributive directory: a few header files will be overwritten; rebuilding of LLVM is not needed.

*** Download the patch and unpack it into your LLVM source directory: the file Cuda.cpp will be overwritten; needs further rebuilding of LLVM.

In most cases, you can get a suitable version of LLVM+CLANG with your package manager.

Failing that or having multiple versions of LLVM, you can download a release archive, build or install it, and set CMAKE_PREFIX_PATH so cmake can find it; for instance: -DCMAKE_PREFIX_PATH=d:\LLVM\11.0.0\dist

hipify-clang: usage

To process a file, hipify-clang needs access to the same headers that would be required to compile it with clang.

For example:

./hipify-clang square.cu --cuda-path=/usr/local/cuda-11.0 -I /usr/local/cuda-11.0/samples/common/inc

hipify-clang arguments are given first, followed by a separator '--', and then the arguments you'd pass to clang if you were compiling the input file. For example:

./hipify-clang cpp17.cu --cuda-path=/usr/local/cuda-11.0 -- -std=c++17

The Clang manual for compiling CUDA may be useful.

For some hipification automation (starting from clang 8.0.0), it is also possible to provide a Compilation Database in JSON format in the compile_commands.json file:

-p <folder containing compile_commands.json> or
-p=<folder containing compile_commands.json>

The compilation database should be provided in the compile_commands.json file or generated by clang based on cmake; options separator '--' must not be used.

For a list of hipify-clang options, run hipify-clang --help.

hipify-clang: building

mkdir build dist
cd build

cmake \
 -DCMAKE_INSTALL_PREFIX=../dist \
 -DCMAKE_BUILD_TYPE=Release \
 ..

make -j install

On Windows, the following option should be specified for cmake at first place: -G "Visual Studio 16 2019 Win64"; the generated hipify-clang.sln should be built by Visual Studio 16 2019 instead of make. Please, see hipify-clang: Windows for the supported tools for building.

Debug build type -DCMAKE_BUILD_TYPE=Debug is also supported and tested; LLVM+CLANG should be built in Debug mode as well. 64-bit build mode (-Thost=x64 on Windows) is also supported; LLVM+CLANG should be built in 64-bit mode as well.

The binary can then be found at ./dist/bin/hipify-clang.

hipify-clang: testing

hipify-clang has unit tests using LLVM lit/FileCheck.

LLVM+CLANG should be built from sources, pre-built binaries are not exhaustive for testing. Before building ensure that the software required for building is of an appropriate version.

LLVM 9.0.1 or older:

  1. download LLVM+CLANG sources;
  2. build LLVM+CLANG:

Linux:

     cmake \
      -DCMAKE_INSTALL_PREFIX=../dist \
      -DLLVM_SOURCE_DIR=../llvm \
      -DLLVM_TARGETS_TO_BUILD="X86;NVPTX" \
      -DCMAKE_BUILD_TYPE=Release \
      ../llvm
     make -j install

Windows:

     cmake \
      -G "Visual Studio 16 2019" \
      -A x64 \
      -DCMAKE_INSTALL_PREFIX=../dist \
      -DLLVM_SOURCE_DIR=../llvm \
      -DLLVM_TARGETS_TO_BUILD="NVPTX" \
      -DCMAKE_BUILD_TYPE=Release \
      -Thost=x64 \
      ../llvm

Run Visual Studio 16 2019, open the generated LLVM.sln, build all, build project INSTALL.

LLVM 10.0.0 or newer:

  1. download LLVM project sources;
  2. build LLVM project:

Linux:

     cmake \
      -DCMAKE_INSTALL_PREFIX=../dist \
      -DLLVM_SOURCE_DIR=../llvm-project \
      -DLLVM_TARGETS_TO_BUILD="X86;NVPTX" \
      -DLLVM_ENABLE_PROJECTS="clang" \
      -DLLVM_TEMPORARILY_ALLOW_OLD_TOOLCHAIN=ON \
      -DCMAKE_BUILD_TYPE=Release \
      ../llvm-project/llvm
     make -j install

Windows:

     cmake \
      -G "Visual Studio 16 2019" \
      -A x64 \
      -DCMAKE_INSTALL_PREFIX=../dist \
      -DLLVM_SOURCE_DIR=../llvm-project \
      -DLLVM_TARGETS_TO_BUILD="NVPTX" \
      -DLLVM_ENABLE_PROJECTS="clang" \
      -DLLVM_TEMPORARILY_ALLOW_OLD_TOOLCHAIN=ON \
      -DCMAKE_BUILD_TYPE=Release \
      -Thost=x64 \
      ../llvm-project/llvm

Run Visual Studio 16 2019, open the generated LLVM.sln, build all, build project INSTALL.

  1. Ensure CUDA of minimum version 7.0 is installed.

    • Having multiple CUDA installations to choose a particular version the DCUDA_TOOLKIT_ROOT_DIR option should be specified:

      • Linux: -DCUDA_TOOLKIT_ROOT_DIR=/usr/include

      • Windows: -DCUDA_TOOLKIT_ROOT_DIR="c:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.0"

        -DCUDA_SDK_ROOT_DIR="c:/ProgramData/NVIDIA Corporation/CUDA Samples/v11.0"

  2. Ensure cuDNN of the version corresponding to CUDA's version is installed.

    • Path to cuDNN should be specified by the CUDA_DNN_ROOT_DIR option:

      • Linux: -DCUDA_DNN_ROOT_DIR=/usr/include

      • Windows: -DCUDA_DNN_ROOT_DIR=d:/CUDNN/cudnn-11.0-windows10-x64-v8.0.4.30

  3. Ensure CUB of the version corresponding to CUDA's version is installed.

    • Path to CUB should be specified by the CUDA_CUB_ROOT_DIR option:

      • Linux: -DCUDA_CUB_ROOT_DIR=/srv/git/CUB

      • Windows: -DCUDA_CUB_ROOT_DIR=d:/GIT/cub

  4. Ensure python of minimum required version 2.7 is installed.

  5. Ensure lit and FileCheck are installed - these are distributed with LLVM.

    • Install lit into python:

      • Linux: python /usr/llvm/11.0.0/llvm-project/llvm/utils/lit/setup.py install

      • Windows: python d:/LLVM/11.0.0/llvm-project/llvm/utils/lit/setup.py install

    • Starting with LLVM 6.0.1 path to llvm-lit python script should be specified by the LLVM_EXTERNAL_LIT option:

      • Linux: -DLLVM_EXTERNAL_LIT=/usr/llvm/11.0.0/build/bin/llvm-lit

      • Windows: -DLLVM_EXTERNAL_LIT=d:/LLVM/11.0.0/build/Release/bin/llvm-lit.py

    • FileCheck:

      • Linux: copy from /usr/llvm/11.0.0/build/bin/ to CMAKE_INSTALL_PREFIX/dist/bin

      • Windows: copy from d:/LLVM/11.0.0/build/Release/bin to CMAKE_INSTALL_PREFIX/dist/bin

      • Or specify the path to FileCheck in CMAKE_INSTALL_PREFIX option

  6. Set HIPIFY_CLANG_TESTS option turned on: -DHIPIFY_CLANG_TESTS=1.

  7. Build and run tests:

hipify-clang: Linux

On Linux the following configurations are tested:

Ubuntu 14: LLVM 4.0.0 - 7.1.0, CUDA 7.0 - 9.0, cuDNN 5.0.5 - 7.6.5.32

Ubuntu 16-18: LLVM 8.0.0 - 11.0.0, CUDA 8.0 - 10.2, cuDNN 5.1.10 - 8.0.4.30

Ubuntu 20: LLVM 9.0.0 - 11.0.0, CUDA 8.0 - 11.1, cuDNN 5.1.10 - 8.0.4.30

Minimum build system requirements for the above configurations:

Python 2.7, cmake 3.5.1, GNU C/C++ 5.4.0.

Here is an example of building hipify-clang with testing support on Ubuntu 20.04.1:

cmake
 -DHIPIFY_CLANG_TESTS=1 \
 -DCMAKE_BUILD_TYPE=Release \
 -DCMAKE_INSTALL_PREFIX=../dist \
 -DCMAKE_PREFIX_PATH=/usr/llvm/11.0.0/dist \
 -DCUDA_TOOLKIT_ROOT_DIR=/usr/include \
 -DCUDA_DNN_ROOT_DIR=/usr/include \
 -DCUDA_CUB_ROOT_DIR=/usr/CUB \
 -DLLVM_EXTERNAL_LIT=/usr/llvm/11.0.0/build/bin/llvm-lit \
 ..

A corresponding successful output:

-- The C compiler identification is GNU 9.3.0
-- The CXX compiler identification is GNU 9.3.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/cc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Found ZLIB: /usr/lib/x86_64-linux-gnu/libz.so (found version "1.2.11")
-- Found LLVM 11.0.0:
--    - CMake module path: /usr/llvm/11.0.0/dist/lib/cmake/llvm
--    - Include path     : /usr/llvm/11.0.0/dist/include
--    - Binary path      : /usr/llvm/11.0.0/dist/bin
-- Linker detection: GNU ld
-- Found PythonInterp: /usr/bin/python3.8 (found suitable version "3.8.5", minimum required is "2.7")
-- Found lit: /usr/local/bin/lit
-- Found FileCheck: /usr/llvm/11.0.0/dist/bin/FileCheck
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Failed
-- Looking for pthread_create in pthreads
-- Looking for pthread_create in pthreads - not found
-- Looking for pthread_create in pthread
-- Looking for pthread_create in pthread - found
-- Found Threads: TRUE
-- Found CUDA: /usr/include (found version "11.0")
-- Configuring done
-- Generating done
-- Build files have been written to: /usr/hipify/build
make test-hipify

A corresponding successful output:

Running HIPify regression tests
========================================
CUDA 11.0 - will be used for testing
LLVM 11.0.0 - will be used for testing
x86_64 - Platform architecture
Linux 5.4.0-51-generic - Platform OS
64 - hipify-clang binary bitness
64 - python 3.8.5 binary bitness
========================================
-- Testing: 66 tests, 12 threads --
PASS: hipify :: unit_tests/casts/reinterpret_cast.cu (1 of 66)
PASS: hipify :: unit_tests/device/atomics.cu (2 of 66)
PASS: hipify :: unit_tests/compilation_database/cd_intro.cu (3 of 66)
PASS: hipify :: unit_tests/device/device_symbols.cu (4 of 66)
PASS: hipify :: unit_tests/device/math_functions.cu (5 of 66)
PASS: hipify :: unit_tests/headers/headers_test_01.cu (6 of 66)
PASS: hipify :: unit_tests/headers/headers_test_02.cu (7 of 66)
PASS: hipify :: unit_tests/headers/headers_test_03.cu (8 of 66)
PASS: hipify :: unit_tests/headers/headers_test_05.cu (9 of 66)
PASS: hipify :: unit_tests/headers/headers_test_06.cu (10 of 66)
PASS: hipify :: unit_tests/headers/headers_test_04.cu (11 of 66)
PASS: hipify :: unit_tests/headers/headers_test_07.cu (12 of 66)
PASS: hipify :: unit_tests/headers/headers_test_10.cu (13 of 66)
PASS: hipify :: unit_tests/headers/headers_test_11.cu (14 of 66)
PASS: hipify :: unit_tests/headers/headers_test_08.cu (15 of 66)
PASS: hipify :: unit_tests/kernel_launch/kernel_launch_01.cu (16 of 66)
PASS: hipify :: unit_tests/headers/headers_test_09.cu (17 of 66)
PASS: hipify :: unit_tests/libraries/CAFFE2/caffe2_02.cu (18 of 66)
PASS: hipify :: unit_tests/libraries/CAFFE2/caffe2_01.cu (19 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/cublas_0_based_indexing.cu (20 of 66)
PASS: hipify :: unit_tests/libraries/CUB/cub_03.cu (21 of 66)
PASS: hipify :: unit_tests/libraries/CUB/cub_01.cu (22 of 66)
PASS: hipify :: unit_tests/libraries/CUB/cub_02.cu (23 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/cublas_sgemm_matrix_multiplication.cu (24 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/rocBLAS/cublas_0_based_indexing_rocblas.cu (25 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/rocBLAS/cublas_1_based_indexing_rocblas.cu (26 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/cublas_1_based_indexing.cu (27 of 66)
PASS: hipify :: unit_tests/libraries/cuComplex/cuComplex_Julia.cu (28 of 66)
PASS: hipify :: unit_tests/libraries/cuDNN/cudnn_softmax.cu (29 of 66)
PASS: hipify :: unit_tests/libraries/cuFFT/simple_cufft.cu (30 of 66)
PASS: hipify :: unit_tests/libraries/cuBLAS/rocBLAS/cublas_sgemm_matrix_multiplication_rocblas.cu (31 of 66)
PASS: hipify :: unit_tests/libraries/cuRAND/poisson_api_example.cu (32 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_03.cu (33 of 66)
PASS: hipify :: unit_tests/libraries/cuRAND/benchmark_curand_generate.cpp (34 of 66)
PASS: hipify :: unit_tests/libraries/cuRAND/benchmark_curand_kernel.cpp (35 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_04.cu (36 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_05.cu (37 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_06.cu (38 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_07.cu (39 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_08.cu (40 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_09.cu (41 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_11.cu (42 of 66)
PASS: hipify :: unit_tests/namespace/ns_kernel_launch.cu (43 of 66)
PASS: hipify :: unit_tests/libraries/cuSPARSE/cuSPARSE_10.cu (44 of 66)
PASS: hipify :: unit_tests/pp/pp_if_else_conditionals.cu (45 of 66)
PASS: hipify :: unit_tests/pp/pp_if_else_conditionals_01.cu (46 of 66)
PASS: hipify :: unit_tests/pp/pp_if_else_conditionals_01_LLVM_10.cu (47 of 66)
PASS: hipify :: unit_tests/pp/pp_if_else_conditionals_LLVM_10.cu (48 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/11_texture_driver/tex2dKernel.cpp (49 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/0_MatrixTranspose/MatrixTranspose.cpp (50 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/11_texture_driver/texture2dDrv.cpp (51 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/13_occupancy/occupancy.cpp (52 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/1_hipEvent/hipEvent.cpp (53 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/2_Profiler/Profiler.cpp (54 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/7_streams/stream.cpp (55 of 66)
PASS: hipify :: unit_tests/samples/MallocManaged.cpp (56 of 66)
PASS: hipify :: unit_tests/samples/2_Cookbook/8_peer2peer/peer2peer.cpp (57 of 66)
PASS: hipify :: unit_tests/samples/allocators.cu (58 of 66)
PASS: hipify :: unit_tests/samples/coalescing.cu (59 of 66)
PASS: hipify :: unit_tests/samples/dynamic_shared_memory.cu (60 of 66)
PASS: hipify :: unit_tests/samples/axpy.cu (61 of 66)
PASS: hipify :: unit_tests/samples/cudaRegister.cu (62 of 66)
PASS: hipify :: unit_tests/samples/intro.cu (63 of 66)
PASS: hipify :: unit_tests/samples/square.cu (64 of 66)
PASS: hipify :: unit_tests/samples/static_shared_memory.cu (65 of 66)
PASS: hipify :: unit_tests/samples/vec_add.cu (66 of 66)
Testing Time: 2.87s
  Expected Passes    : 66
[100%] Built target test-hipify

hipify-clang: Windows

Tested configurations:

LLVM CUDA cuDNN Visual Studio (latest) cmake Python
4.0.0 - 5.0.2 8.0 5.1.10 - 7.1.4.18 2015.14.0, 2017.15.5.2 3.5.1, 3.18.0 3.6.4, 3.8.5
6.0.0 - 6.0.1 9.0 7.0.5.15 - 7.6.5.32 2015.14.0, 2017.15.5.5 3.6.0, 3.18.0 3.7.2, 3.8.5
7.0.0 - 7.1.0 9.2 7.6.5.32 2017.15.9.11 3.13.3, 3.18.0 3.7.3, 3.8.5
8.0.0 - 8.0.1 10.0 7.6.5.32 2017.15.9.15 3.14.2, 3.18.0 3.7.4, 3.8.5
9.0.0 - 9.0.1 10.1 7.6.5.32 2017.15.9.20, 2019.16.4.5 3.16.4, 3.18.0 3.8.0, 3.8.5
10.0.0 - 11.0.0 8.0 - 11.1 7.6.5.32 - 8.0.4.30 2017.15.9.28, 2019.16.7.6 3.18.4 3.9.0
12.0.0git 8.0 - 11.1 7.6.5.32 - 8.0.4.30 2017.15.9.28, 2019.16.7.6 3.18.4 3.9.0

Building with testing support by Visual Studio 16 2019 on Windows 10:

cmake
 -G "Visual Studio 16 2019" \
 -A x64 \
 -DHIPIFY_CLANG_TESTS=1 \
 -DCMAKE_BUILD_TYPE=Release \
 -DCMAKE_INSTALL_PREFIX=../dist \
 -DCMAKE_PREFIX_PATH=d:/LLVM/11.0.0/dist \
 -DCUDA_TOOLKIT_ROOT_DIR="c:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.0" \
 -DCUDA_SDK_ROOT_DIR="c:/ProgramData/NVIDIA Corporation/CUDA Samples/v11.0" \
 -DCUDA_DNN_ROOT_DIR=d:/CUDNN/cudnn-11.0-windows10-x64-v8.0.4.30 \
 -DCUDA_CUB_ROOT_DIR=d:/GIT/cub \
 -DLLVM_EXTERNAL_LIT=d:/LLVM/11.0.0/build/Release/bin/llvm-lit.py \
 -Thost=x64
 ..

A corresponding successful output:

-- Found LLVM 11.0.0:
--    - CMake module path: d:/LLVM/11.0.0/dist/lib/cmake/llvm
--    - Include path     : d:/LLVM/11.0.0/dist/include
--    - Binary path      : d:/LLVM/11.0.0/dist/bin
-- Found PythonInterp: c:/Program Files/Python39/python.exe (found suitable version "3.9.0", minimum required is "3.6")
-- Found lit: c:/Program Files/Python39/Scripts/lit.exe
-- Found FileCheck: d:/LLVM/11.0.0/dist/bin/FileCheck.exe
-- Found CUDA: c:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.0 (found version "11.0")
-- Configuring done
-- Generating done
-- Build files have been written to: d:/hipify/build

Run Visual Studio 16 2019, open the generated hipify-clang.sln, build project test-hipify.

hipify-perl

hipify-perl is autogenerated perl-based script which heavily uses regular expressions.

Advantages:

  1. Ease in use.

  2. It doesn't check the input source CUDA code for correctness.

  3. It doesn't have dependencies on 3rd party tools, including CUDA.

Disadvantages:

  1. Current disability (and difficulty in implementing) of transforming the following constructs:

    • macros expansion;

    • namespaces:

      • redefines of CUDA entities in user namespaces;

      • using directive;

    • templates (some cases);

    • device/host function calls distinguishing;

    • header files correct injection;

    • complicated argument lists parsing.

  2. Difficulties in supporting.

hipify-perl: usage

perl hipify-perl square.cu > square.cu.hip

hipify-perl: building

To generate hipify-perl, run hipify-clang --perl. Output directory for the generated hipify-perl file might be specified by --o-hipify-perl-dir option.

Supported CUDA APIs

To generate the above documentation with the actual information about all supported CUDA APIs in Markdown format, run hipify-clang --md with or without output directory specifying (-o).

Disclaimer

The information contained herein is for informational purposes only, and is subject to change without notice. While every precaution has been taken in the preparation of this document, it may contain technical inaccuracies, omissions and typographical errors, and AMD is under no obligation to update or otherwise correct this information. Advanced Micro Devices, Inc. makes no representations or warranties with respect to the accuracy or completeness of the contents of this document, and assumes no liability of any kind, including the implied warranties of noninfringement, merchantability or fitness for particular purposes, with respect to the operation or use of AMD hardware, software or other products described herein. No license, including implied or arising by estoppel, to any intellectual property rights is granted by this document. Terms and limitations applicable to the purchase or use of AMD's products are as set forth in a signed agreement between the parties or in AMD's Standard Terms and Conditions of Sale.

AMD, the AMD Arrow logo, and combinations thereof are trademarks of Advanced Micro Devices, Inc. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.

Copyright (c) 2014-2020 Advanced Micro Devices, Inc. All rights reserved.

About

HIPIFY: Convert CUDA to Portable C++ Code

Resources

Readme
Activity

Stars

1 star

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages

  • C++ 67.6%
  • Cuda 18.0%
  • Perl 13.8%
  • CMake 0.3%
  • Python 0.2%
  • Shell 0.1%