|
Ginkgo
Generated from pipelines/1554403166 branch based on develop. Ginkgo version 1.9.0
A numerical linear algebra library targeting many-core architectures
|
5 #ifndef GKO_PUBLIC_CORE_BASE_DEVICE_HPP_
6 #define GKO_PUBLIC_CORE_BASE_DEVICE_HPP_
12 #include <type_traits>
14 #include <ginkgo/config.hpp>
41 static std::mutex& get_mutex(
int i);
50 static int& get_num_execs(
int i);
52 static constexpr
int max_devices = 64;
71 static std::mutex& get_mutex(
int i);
80 static int& get_num_execs(
int i);
82 static constexpr
int max_devices = 64;
88 #endif // GKO_PUBLIC_CORE_BASE_DEVICE_HPP_
nvidia_device handles the number of executor on Nvidia devices and have the corresponding recursive_m...
Definition: device.hpp:29
This is the Executor subclass which represents the HIP enhanced device.
Definition: executor.hpp:1777
This is the Executor subclass which represents the CUDA device.
Definition: executor.hpp:1540
The Ginkgo namespace.
Definition: abstract_factory.hpp:20
amd_device handles the number of executor on Amd devices and have the corresponding recursive_mutex.
Definition: device.hpp:60