Class DriverAPINativeMethods.AsynchronousMemcpy_v2

Any host memory involved must be DMA'able (e.g., allocated with cuMemAllocHost). memcpy's done with these functions execute in parallel with the CPU and, if the hardware is available, may execute in parallel with the GPU. Asynchronous memcpy must be accompanied by appropriate stream synchronization.

Inheritance

System.Object

DriverAPINativeMethods.AsynchronousMemcpy_v2

Inherited Members

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: ManagedCuda

Assembly: ManagedCuda.dll

Syntax

public static class AsynchronousMemcpy_v2

Methods

cuMemcpy2DAsync_v2(ref CUDAMemCpy2D, CUstream)

Perform a 2D memory copy according to the parameters specified in pCopy. See CUDAMemCpy2D. cuMemcpy2DAsync_v2(ref CUDAMemCpy2D, CUstream) returns an error if any pitch is greater than the maximum allowed (memPitch). cuMemAllocPitch_v2(ref CUdeviceptr, ref SizeT, SizeT, SizeT, UInt32) passes back pitches that always work with cuMemcpy2DAsync_v2(ref CUDAMemCpy2D, CUstream). On intra-device memory copies (device <-> device, CUDA array <-> device, CUDA array <-> CUDA array), cuMemcpy2DAsync_v2(ref CUDAMemCpy2D, CUstream) may fail for pitches not computed by cuMemAllocPitch_v2(ref CUdeviceptr, ref SizeT, SizeT, SizeT, UInt32). cuMemcpy2DUnaligned_v2(ref CUDAMemCpy2D) (not async!) does not have this restriction, but may run significantly slower in the cases where cuMemcpy2DAsync_v2(ref CUDAMemCpy2D, CUstream) would have returned an error code.

Declaration

public static CUResult cuMemcpy2DAsync_v2(ref CUDAMemCpy2D pCopy, CUstream hStream)

Parameters

Type	Name	Description
CUDAMemCpy2D	pCopy	Parameters for the memory copy
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpy3DAsync_v2(ref CUDAMemCpy3D, CUstream)

Perform a 3D memory copy according to the parameters specified in pCopy. See CUDAMemCpy3D. cuMemcpy3DAsync_v2(ref CUDAMemCpy3D, CUstream) returns an error if any pitch is greater than the maximum allowed (memPitch).

cuMemcpy3DAsync_v2(ref CUDAMemCpy3D, CUstream) is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument. It only works on page-locked host memory and returns an error if a pointer to pageable memory is passed as input.

The srcLOD and dstLOD members of the CUDAMemCpy3D structure must be set to 0.

Declaration

public static CUResult cuMemcpy3DAsync_v2(ref CUDAMemCpy3D pCopy, CUstream hStream)

Parameters

Type	Name	Description
CUDAMemCpy3D	pCopy	Parameters for the memory copy
CUstream	hStream	Stream indetifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpy3DPeerAsync(ref CUDAMemCpy3DPeer, CUstream)

Perform a 3D memory copy according to the parameters specified in pCopy. See the definition of the CUDAMemCpy3DPeer structure for documentation of its parameters.

Declaration

public static CUResult cuMemcpy3DPeerAsync(ref CUDAMemCpy3DPeer pCopy, CUstream hStream)

Parameters

Type	Name	Description
CUDAMemCpy3DPeer	pCopy	Parameters for the memory copy
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyAsync(CUdeviceptr, CUdeviceptr, SizeT, CUstream)

Copies data between two pointers. dst and src are base pointers of the destination and source, respectively.
ByteCount specifies the number of bytes to copy. Note that this function infers the type of the transfer (host to host, host to device, device to device, or device to host) from the pointer values. This function is only allowed in contexts which support unified addressing. Note that this function is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument

Declaration

public static CUResult cuMemcpyAsync(CUdeviceptr dst, CUdeviceptr src, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
CUdeviceptr	dst	Destination unified virtual address space pointer
CUdeviceptr	src	Source unified virtual address space pointer
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyAtoHAsync_v2(IntPtr, CUarray, SizeT, SizeT, CUstream)

Copies from one 1D CUDA array to host memory. dstHost specifies the base pointer of the destination. srcArray and srcOffset specify the CUDA array handle and starting offset in bytes of the source data. ByteCount specifies the number of bytes to copy.

cuMemcpyAtoHAsync_v2(IntPtr, CUarray, SizeT, SizeT, CUstream) is asynchronous and can optionally be associated to a stream by passing a non-zero stream hStream argument. It only works on page-locked host memory and returns an error if a pointer to pageable memory is passed as input.

Declaration

public static CUResult cuMemcpyAtoHAsync_v2(IntPtr dstHost, CUarray srcArray, SizeT srcOffset, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
System.IntPtr	dstHost	Destination pointer
CUarray	srcArray	Source array
SizeT	srcOffset	Offset in bytes of source array
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyDtoDAsync_v2(CUdeviceptr, CUdeviceptr, SizeT, CUstream)

Copies from device memory to device memory. dstDevice and srcDevice are the base pointers of the destination and source, respectively. ByteCount specifies the number of bytes to copy. Note that this function is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument.

Declaration

public static CUResult cuMemcpyDtoDAsync_v2(CUdeviceptr dstDevice, CUdeviceptr srcDevice, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
CUdeviceptr	dstDevice	Destination device pointer
CUdeviceptr	srcDevice	Source device pointer
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyDtoHAsync_v2(IntPtr, CUdeviceptr, SizeT, CUstream)

Copies from device to host memory. dstHost and srcDevice specify the base pointers of the destination and source, respectively. ByteCount specifies the number of bytes to copy.

cuMemcpyDtoHAsync_v2(IntPtr, CUdeviceptr, SizeT, CUstream) is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument. It only works on page-locked memory and returns an error if a pointer to pageable memory is passed as input.

Declaration

public static CUResult cuMemcpyDtoHAsync_v2(IntPtr dstHost, CUdeviceptr srcDevice, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
System.IntPtr	dstHost	Destination host pointer
CUdeviceptr	srcDevice	Source device pointer
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyHtoAAsync_v2(CUarray, SizeT, IntPtr, SizeT, CUstream)

Copies from host memory to a 1D CUDA array. dstArray and dstOffset specify the CUDA array handle and starting offset in bytes of the destination data. srcHost specifies the base address of the source. ByteCount specifies the number of bytes to copy.

cuMemcpyHtoAAsync_v2(CUarray, SizeT, IntPtr, SizeT, CUstream) is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument. It only works on page-locked memory and returns an error if a pointer to pageable memory is passed as input.

Declaration

public static CUResult cuMemcpyHtoAAsync_v2(CUarray dstArray, SizeT dstOffset, IntPtr srcHost, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
CUarray	dstArray	Destination array
SizeT	dstOffset	Offset in bytes of destination array
System.IntPtr	srcHost	Source host pointer
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyHtoDAsync_v2(CUdeviceptr, IntPtr, SizeT, CUstream)

Copies from host memory to device memory. dstDevice and srcHost are the base addresses of the destination and source, respectively. ByteCount specifies the number of bytes to copy.

cuMemcpyHtoDAsync_v2(CUdeviceptr, IntPtr, SizeT, CUstream) is asynchronous and can optionally be associated to a stream by passing a non-zero hStream argument. It only works on page-locked memory and returns an error if a pointer to pageable memory is passed as input.

Declaration

public static CUResult cuMemcpyHtoDAsync_v2(CUdeviceptr dstDevice, IntPtr srcHost, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
CUdeviceptr	dstDevice	Destination device pointer
System.IntPtr	srcHost	Source host pointer
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.

cuMemcpyPeerAsync(CUdeviceptr, CUcontext, CUdeviceptr, CUcontext, SizeT, CUstream)

Copies from device memory in one context to device memory in another context. dstDevice is the base device pointer of the destination memory and dstContext is the destination context. srcDevice is the base device pointer of the source memory and srcContext is the source pointer.
ByteCount specifies the number of bytes to copy. Note that this function is asynchronous with respect to the host and all work in other streams in other devices.

Declaration

public static CUResult cuMemcpyPeerAsync(CUdeviceptr dstDevice, CUcontext dstContext, CUdeviceptr srcDevice, CUcontext srcContext, SizeT ByteCount, CUstream hStream)

Parameters

Type	Name	Description
CUdeviceptr	dstDevice	Destination device pointer
CUcontext	dstContext	Destination context
CUdeviceptr	srcDevice	Source device pointer
CUcontext	srcContext	Source context
SizeT	ByteCount	Size of memory copy in bytes
CUstream	hStream	Stream identifier

Returns

Type	Description
CUResult	CUDA Error Codes: Success, ErrorDeinitialized, ErrorNotInitialized, ErrorInvalidContext, ErrorInvalidValue. Note that this function may also return error codes from previous, asynchronous launches.