Released on TBD.
- The
cuda.bindings.nvvmPython module was added, wrapping the libNVVM C API. - Source build error checking added for missing required headers
- Statically link CUDA Runtime instead of reimplementing it
- Move stream callback wrappers to the Python layer
- Return code construction is made faster
- Fix segfault when converting char* NULL to bytes
- Benchmark suite is updated
- Improvements in the introductory code samples
- Fix performance hint warnings raised by Cython 3
- Improvements in the Overview page