Full documentation for HIPIFY is available at hipify.readthedocs.io.
- CUDA 12.2.0 support
- cuDNN 8.9.2 support
- LLVM 16.0.6 support
- Initial rocSPARSE support
- Initial CUDA2ROC documentation generation for rocBLAS, rocSPARSE, and MIOpen:
- in separate files: hipify-clang --md --doc-format=full --doc-roc=separate
- in a single file: hipify-clang --md --doc-format=full --doc-roc=joint
- New options:
- --use-hip-data-types (Use 'hipDataType' instead of 'hipblasDatatype_t' or 'rocblas_datatype')
- --doc-roc=<value> (ROC documentation generation: 'skip' (default), 'separate', or 'joint'; the '--md' or '--csv' option must be specified)
- [#822] Add a new function call transformation type "additional const by value arg"
- [#830] Add a new function call transformation type "move arg from place X to place Y"
- CUDA 12.1.0 support
- cuDNN 8.8.1 support
- LLVM 16.0.0 support
- New options:
- --default-preprocessor (synonymous with '--skip-excluded-preprocessor-conditional-blocks')
- --no-undocumented-features
- --no-warnings-on-undocumented-features
- --versions
- Accessing half2 struct members (undocumented feature)
- INSTALL to 'bin' subfolder
- Partial CUDA 12.0.0 support
- cuDNN 8.7.0 support
- Initial MIOpem support
- cuBLAS 64bit API (CUDA 12.0) initial support
- rocBLAS and MIOpen synthetic tests
- LLVM 15.0.7 support
- Synthetic unit tests for cuBLAS2rocBLAS and cuDNN2MIOpen
- CUDA 11.8.0 support
- cuDNN 8.6.0 support
- Device types support
- LLVM 15.0.4 support
- [Linux] Get rid of any RPATH definitions
- hipRTC support
- Error Handling API support
- hipBLAS synthetic tests
- LLVM 15.0.0 support
- CUDA 11.7.0 support
- cuDNN 8.4.1 support
- CUB initial support
- More synthetic tests
- New options:
- --hip-kernel-execution-syntax
- Patches for LLVM 14.0.x (Windows only)
- Add GNUInstallDirs for CMake on Linux
- LLVM 3.8.0 is out of support
- HIPIFY-specific options support in unit testing
- CUDA 11.6.0 support
- cuDNN 8.3.3 support
- LLVM 14.0.0 support
- CUDA 11.5.1 support
- cuDNN 8.3.2 support
- hipification of cuOccupancyMaxPotentialBlockSize and cuOccupancyMaxPotentialBlockSizeWithFlags
- CUDA 11.4.2 support
- cuDNN 8.3.2 support
- Initial hipRTC support
- GNU C/C++ 11.2 support
- Visual Studio 2022 support
- LLVM 13.0.0 support
- New options:
- --experimental
- --cuda-kernel-execution-syntax
- Packaging for Debian and RPM Linuxes
- Undo args typecasting for 4 Driver API functions cuStreamWaitValue32(64), cuStreamWriteValue32(64) as those args in the corresponding HIP functions become uint finally
- Support for different formats of the generated documentation
- Experimentally supported APIs
- cuDNN 8.2.4 support
- Initial graph API support
- GNU C/C++ 11.1 support
- LLVM 12.0.1 support
- Abandon HIP_DYNAMIC_SHARED
- Synthetic unit tests
- -std=c++14 by default
- CUDA 11.3.0 support
- cuDNN 8.2.0 support
- LLVM 12.0.0 support
- Added the missing type casting of arguments for the following functions: cuStreamWaitValue32(64), cuStreamWriteValue32(64)
- CUDA 11.2.2 support
- cuDNN 8.1.1 support
- Initial Device API support
- LLVM 11.1.0 support
- New options:
- --doc-format= - Documentation format: 'full' (default), 'strict', or 'compact'
- Tests on kernel launch syntax
- CUDA 11.2.0 support
- Stream Ordered Memory API support
- cuDNN 8.1.1 support
- LLVM 11.0.1 support
- Patches for LLVM 10.0.x and 11.0.0 (both Windows and Linux)
- Initial support for APIs versioning
- LLVM 11.0.0 support
- CUDA and HIP APIs and data types versioning are revised
- CUDA and HIP deprecated and removed APIs and data types are revised
- CUDA 11.0.1 support
- Initial CUDA2HIP documentation generation in Markdown and CSV formats
- Initial CUDA and HIP APIs versioning support
- New options:
- --md (Generate Markdown documentation)
- --csv (Generate CSV documentation)
- Improved hipify-perl generation
- cuDNN 8.0.2 support
- compile_commands.json support (-p )
- Improved hipify-perl generation
- CUDA 11.0.0 support
- Linux packaging
- LLVM 10.0.1 support
- The 'deprecated' flag for all corresponding CUDA and HIP APIs
- Started warning on deprecated APIs
- CUDA 10.2.0 support
- cuDNN 7.6.5 support
- LLVM 10.0.0 support
- hipify-clang and clang options separator '--' support