Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion cuda_bindings/docs/source/tips_and_tricks.rst
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ Getting the address of underlying C objects from the low-level bindings

All CUDA C types are exposed to Python as Python classes. For example, the :class:`~cuda.bindings.driver.CUstream` type is exposed as a class with methods :meth:`~cuda.bindings.driver.CUstream.getPtr()` and :meth:`~cuda.bindings.driver.CUstream.__int__()` implemented.

There is an important distinction between the ``getPtr()`` method and the behaviour of ``__int__()``. If you need to get the pointer address *of* the underlying ``CUstream`` C object wrapped in the Python class, you can do so by calling ``int(instance_of_CUstream)``, which returns the address as a Python `int`, while calling ``instance_of_CUstream.getPtr()`` returns the pointer *to* the ``CUstream`` C object (that is, ``&CUstream``) as a Python `int`.
There is an important distinction between the ``getPtr()`` method and the behaviour of ``__int__()``. Since a ``CUstream`` is itself just a pointer, calling ``instance_of_CUstream.getPtr()`` returns the pointer *to* the pointer, instead of the value of the ``CUstream`` C object that is the pointer to the underlying stream handle. ``int(instance_of_CUstream)`` returns the value of the ``CUstream`` converted to a Python int and is the actual address of the underlying handle.


Lifetime management of the CUDA objects
Expand Down