Class to help launch and configure a CUDA kernel.
More...
#include <Kernel.hpp>
◆ Kernel()
Kernel::Kernel |
( |
std::shared_ptr< char[]> |
ptx, |
|
|
std::string |
demangled_name |
|
) |
| |
create a Kernel based on a templated kernel string
- Parameters
-
ptx | |
kernel_name | |
demangled_name | |
◆ benchmark()
benchmark a Kernel
- Parameters
-
kernel_args | |
number | of executions |
device | |
- Returns
- vector of KernelTimes for every execution
◆ configure()
Kernel & Kernel::configure |
( |
dim3 |
grid, |
|
|
dim3 |
block, |
|
|
unsigned int |
shared = 0 |
|
) |
| |
- Parameters
-
grid | vector of grid dimensions |
block | vector of block dimensions |
shared | amount of dynamic shared memory to allocate |
- Returns
- this (for method chaining)
◆ launch()
- Parameters
-
- Returns
- KernelTime
The documentation for this class was generated from the following files: