Helper class usable as a Thrust CUDA execution policy that uses RMM for temporary memory allocation on the specified stream and which allows the Thrust backend to skip stream synchronizations that are not required for correctness. More...
#include <exec_policy.hpp>
Inheritance diagram for rmm::exec_policy_nosync:
Collaboration diagram for rmm::exec_policy_nosync:
Public Member Functions | |
| exec_policy_nosync (cuda_stream_view stream=cuda_stream_default, device_async_resource_ref mr=mr::get_current_device_resource_ref()) | |
| Construct a new execution policy object. More... | |
Detailed Description
Helper class usable as a Thrust CUDA execution policy that uses RMM for temporary memory allocation on the specified stream and which allows the Thrust backend to skip stream synchronizations that are not required for correctness.
Constructor & Destructor Documentation
◆ exec_policy_nosync()
|
explicit |
Construct a new execution policy object.
- Parameters
-
stream The stream on which to allocate temporary memory mr The resource to use for allocating temporary memory
The documentation for this class was generated from the following file:
