|
GPUOcelot
|
|
GPUOcelot
|
#include <Device.h>
Public Attributes | |
| ir::Instruction::Architecture | ISA |
| int | addressSpace |
| char | name [256] |
| size_t | totalMemory |
| unsigned int | multiprocessorCount |
| int | memcpyOverlap |
| int | maxThreadsPerBlock |
| int | maxThreadsDim [3] |
| int | maxGridSize [3] |
| int | sharedMemPerBlock |
| int | totalConstantMemory |
| int | SIMDWidth |
| int | memPitch |
| int | regsPerBlock |
| int | clockRate |
| int | textureAlign |
| int | integrated |
| int | concurrentKernels |
| int | major |
| int | minor |
| size_t | stackSize |
| size_t | printfFIFOSize |
| bool | unifiedAddressing |
| int | memoryClockRate |
| int | memoryBusWidth |
| int | l2CacheSize |
| int | maxThreadsPerMultiProcessor |
identifies the device's address space
clock frequency in kHz
Concurrent kernel execution
Is the device integrated or discrete
"native" ISA of the device
Maximum resident threads per multiprocessor
major shader module revision
maximum size of each dimension of a grid
maximum size of each dimension of a block
maximum number of threads per block
true if the device can simultaneously execute a kernel while performing data transfer
Size of L2 cache in bytes
Global memory bus width in bits
maximum pitch allowed by memory copy functions
minor shader model revision
gets the number of multiprocessors/cores on the device
| char executive::Device::PropertiesData::name[256] |
human-readable device name
printfFIFOSize This device shares a unified address with the host
total registers allowed per block
total amount of shared memory available per block in bytes
warp size
alignment requirement for textures
total amount of constant memory on the device
number of bytes of global memory available to the device
Peak memory clock frequency in kilohertz
1.7.3
