public class TOpenCLDevice
Stores data and information associated with the Open CL computing device.
The object holds all information describing the properties of the device, programs, command queues and cache (buffer objects). To use the device we first define the size of the Cache and load the Programs. Then we can access Kernels which can be enqueued to Command queues.
Properties
| Name | Type | Description |
|---|---|---|
| AddressBits | Int32 | The default compute device address space size specified as an unsigned integer value in bits. Currently supported values are 32 or 64 bits. |
| AMDInfo | TAMDDeviceInfo | Holds AMD specific device information. |
| Available | Boolean | Returns true, if the device is available for use. |
| ByteAddressableStore | Boolean | Returns true for Open CL v1.1. |
| Cache | TOpenCLCache | The Cache object holds a list of Dew.Math.TOpenCLVector and Dew.Math.TOpenCLMatrix objects. |
| CommandQueue | TOpenCLCommandQueueList | The list of available command queues. There is always at least one queue. |
| CompilerAvailable | Boolean | Returns true, if the Open CL driver has a compiler included. |
| D3D10Sharing | Boolean | If true then interoperability between OpenCL and Direct3D 10 is supported. |
| DeviceID | IntPtr | Returns the Device ID assigned to the device by Open CL driver. |
| DeviceIndex | Int32 | Returns the index of the device in the list of devices for the platform. |
| DevicePlatform | TOpenCLPlatform | Returns the platform of the device. |
| DeviceProfile | String | OpenCL profile string. |
| DeviceVersion | String | Returns the device version. |
| DriverVersion | String | Returns the Open CL driver version. |
| ErrorCorrectionSupport | Boolean | Returns true, if global memory uses ECC. |
| ExecutionCapabilityNative | Boolean | Returns true, if the driver can execute native (not Open CL) kernels. |
| Extensions | TStringList | Holds a list of extension names supported by the device. |
| FP16 | Boolean | Returns true if 16bit half precision floating point math is supported on the device. |
| FP64 | Boolean | Returns true if 64bit double precision floating point math is supported on the device. |
| FPDoubleInfo | TFPInfo | Holds information about double precision floating point math. |
| FPHalfInfo | TFPInfo | Holds information about half precision floating point math. |
| FPSingleInfo | TFPInfo | Holds information about single precision floating point math. |
| GlobalMemCachelineSize | Int32 | Returns global memory cache line size in Bytes. |
| GLobalMemCacheSize | Int64 | Returns global memory cache size in KBytes. |
| GlobalMemSize | Int64 | Size of global device memory in KBytes. |
| GLSharing | Boolean | Creating CL context from a GL context or share group is supported, if this property returns true. |
| Image2DMaxHeight | Int32 | Max height of 2D image in pixels. |
| Image2DMaxWidth | Int32 | Max width of 2D image in pixels. |
| Image3DMaxDepth | Int32 | Max depth of 3D image in pixels. |
| Image3DMaxHeight | Int32 | Max height of 3D image in pixels. |
| Image3DMaxWidth | Int32 | Max width of 3D image in pixels. |
| ImageSupport | Boolean | Returns true, if immages are supported by the device. |
| IsAccType | Boolean | Returns true if the device is of Accelerator type. |
| IsCpuType | Boolean | Returns true if the device is of CPU type. |
| IsDefault | Boolean | Returns true if the device is of the default type. |
| IsGpuType | Boolean | Returns true if the device is of GPU type. |
| Kernels | TOpenCLKernelList | The kernel list holds the list of kernels (Open CL functions) that are available. |
| LocalMemSize | Int64 | Returns the size of local memory (if present) in Bytes. |
| LocalMemType | TLocalMemType | Returns lmtLocal, if dedicated local memory storage is available. Returns lmtGlobal oterwhise. |
| MaxClockFrequency | Int32 | Maximum configured clock frequency of the device in MHz. |
| MaxComputeUnits | Int32 | Returns the number of parallel compute units on the OpenCL device. |
| MaxConstantBufferSize | Int64 | Returns the size of constant memory in Bytes. |
| MaxMemAllocSize | Int64 | Max size of memory object allocation in KBytes. |
| MaxParameterSize | Int32 | Returns maximum size of the parameter in bytes. |
| MaxReadImageArgs | Int32 | Max number of simultaneous image objects that can be read by a kernel. |
| MaxSamplers | Int32 | Maximum number of samplers that can be used in a kernel. |
| MaxWorkGroupSize | Int32 | Maximum number of work-items in a work-group executing a kernel on a single compute unit, using the data parallel execution model. |
| MaxWorkItemDimensions | Int32 | Maximum dimensions that specify the global and local work-item IDs used by the data parallel execution model. |
| MaxWriteImageArgs | Int32 | Max number of simultaneous image objects that can be written by a kernel. |
| MemBaseAddrAlign | Int32 | Allignment in bits of memory buffers. |
| MinDataTypeAlignSize | Int32 | Minimum data alignment size. |
| Name | String | Returns the name of the device. |
| NativeCPU | Boolean | If true, application will run optimal CPU code without Open CL when this device will be the target . |
| NVInfo | TNVDeviceInfo | Holds NVidia specific device information. |
| OpenCLVersion | Int32 | Returns Open CL Version. |
| PlatformIndex | Int32 | Returns the index of the platform in the list of platforms. |
| PreferredVectorWidthChar | Int32 | Returns preffered vector width for Open CL char type. |
| PreferredVectorWidthDouble | Int32 | Returns preffered vector width for Open CL double type. |
| PreferredVectorWidthFloat | Int32 | Returns preffered vector width for Open CL float type. |
| PreferredVectorWidthInt | Int32 | Returns preffered vector width for Open CL int type. |
| PreferredVectorWidthLong | Int32 | Returns preffered vector width for Open CL long type. |
| PreferredVectorWidthShort | Int32 | Returns preffered vector width for Open CL short type. |
| ProfilingTimerResolution | Int32 | Describes the resolution of device timer. This is measured in nanoseconds. |
| Programs | TOpenCLProgramList | The program list holds the list of compiled Open CL programs. |
| QueueImmediateExecution | Boolean | If true, the commands queued in the command-queue can be executed within the callers thread when the clEnqueue* is called. |
| QueueOutOfOrderExec | Boolean | If true, the commands queued in the command-queue can be executed in-order or out-oforder. |
| QueueProfiling | Boolean | If true, command queues support profiling. |
| Vendor | String | Returns the name of the Vendor. |
| VendorID | Int64 | Returns the value of the Vendor ID. |
Methods
| Name | Description |
|---|---|
| GetMaxWorkItemSizes | Maximum number of work-items that can be specified in each dimension of the work-group. |