CUDA Tools
[Tools]

Functions
__global__ void	copyDevToDev (FLOAT_ARRAY_TYPE *dA, FLOAT_ARRAY_TYPE *dB, int N)
	Copy an array of vectors on the device.
__global__ void	gpuSimpleComponentSum (FLOAT_ARRAY_TYPE *dVal, FLOAT_ARRAY_TYPE *sum, int N)
	One thread kernel for summation.
__global__ void	gpuSimpleSqSum (FLOAT_ARRAY_TYPE *dVal, float *sqSum, int N)
	One thread kernel for squared summation.

---

Detailed Description

This module contains tools for everyday tasks dealing with CUDA

---

Function Documentation

__global__ void copyDevToDev	(	FLOAT_ARRAY_TYPE *	dA,
		FLOAT_ARRAY_TYPE *	dB,
		int	N
	)

Copy an array of vectors on the device

Parameters:

[in]	dA	Device vector array a
[out]	dB	Device vector array b
[in]	N	Number of vectors

Definition at line 391 of file gpu_tools.cu.

__global__ void gpuSimpleComponentSum	(	FLOAT_ARRAY_TYPE *	dVal,
		FLOAT_ARRAY_TYPE *	sum,
		int	N
	)

One thread kernel for summation

Attention:

This kernel uses only a single thread on the GPU and is not supposed to be used in production code!

Parameters:

[in]	dVal	Input data
[out]	sum	Output of sum
[in]	N	Number of elements to be summed

Definition at line 416 of file gpu_tools.cu.

__global__ void gpuSimpleSqSum	(	FLOAT_ARRAY_TYPE *	dVal,
		float *	sqSum,
		int	N
	)

One thread kernel for squared summation

Attention:

This kernel uses only a single thread on the GPU and is not supposed to be used in production code!

Parameters:

[in]	dVal	Input data
[out]	sqSum	Output of the squared sum
[in]	N	Number of elements to be summed

Definition at line 403 of file gpu_tools.cu.