Management of per-device memory resources. More...
#include <rmm/cuda_device.hpp>#include <rmm/detail/export.hpp>#include <rmm/detail/runtime_shutdown.hpp>#include <rmm/mr/cuda_memory_resource.hpp>#include <rmm/resource_ref.hpp>#include <cuda/memory_resource>#include <map>#include <mutex>#include <utility>
Include dependency graph for per_device_resource.hpp:
This graph shows which files directly or indirectly include this file:
Go to the source code of this file.
Namespaces | |
| rmm | |
| RAPIDS Memory Manager - The top-level namespace for all RMM functionality. | |
| rmm::mr | |
| Memory Resource classes and adaptors. | |
Functions | |
| cuda_memory_resource & | rmm::mr::detail::initial_resource () |
| Returns a reference to the initial resource. More... | |
| std::mutex & | rmm::mr::detail::ref_map_lock () |
| Reference to the lock. More... | |
| auto & | rmm::mr::detail::get_ref_map () |
| Reference to the map from device id -> any_resource. More... | |
| device_async_resource_ref | rmm::mr::get_per_device_resource_ref (cuda_device_id device_id) |
Get the device_async_resource_ref for the specified device. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::set_per_device_resource (cuda_device_id device_id, cuda::mr::any_resource< cuda::mr::device_accessible > new_resource) |
| Set the memory resource for the specified device. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::set_per_device_resource_ref (cuda_device_id device_id, device_async_resource_ref new_resource_ref) |
Set the device_async_resource_ref for the specified device to new_resource_ref More... | |
| device_async_resource_ref | rmm::mr::get_current_device_resource_ref () |
Get the device_async_resource_ref for the current device. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::set_current_device_resource (cuda::mr::any_resource< cuda::mr::device_accessible > new_resource) |
| Set the memory resource for the current device. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::set_current_device_resource_ref (device_async_resource_ref new_resource_ref) |
Set the device_async_resource_ref for the current device. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::reset_per_device_resource (cuda_device_id device_id) |
| Reset the memory resource for the specified device to the initial resource. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::reset_current_device_resource () |
| Reset the memory resource for the current device to the initial resource. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::reset_per_device_resource_ref (cuda_device_id device_id) |
Reset the device_async_resource_ref for the specified device to the initial resource. More... | |
| cuda::mr::any_resource< cuda::mr::device_accessible > | rmm::mr::reset_current_device_resource_ref () |
Reset the device_async_resource_ref for the current device to the initial resource. More... | |
Detailed Description
Management of per-device memory resources.
Provides functions to get/set the default device_async_resource_ref for each CUDA device. The initial resource for each device is a cuda_memory_resource.
- Note
- Memory resources make CUDA API calls without setting the current CUDA device. Therefore a memory resource should only be used with the current CUDA device set to the device that was active when the memory resource was created.
Function Documentation
◆ get_ref_map()
|
inline |
Reference to the map from device id -> any_resource.
- Returns
- Reference to the map from device id -> any_resource
◆ initial_resource()
|
inline |
Returns a reference to the initial resource.
Returns a global instance of a cuda_memory_resource as a function local static.
- Returns
- Reference to the static cuda_memory_resource used as the initial, default resource
◆ ref_map_lock()
|
inline |
Reference to the lock.
- Returns
- Reference to the lock
