tgstkCUDAOperations.h File Reference

#include <cuda_runtime.h>

Go to the source code of this file.

Functions
__global__ void	gpuAdd (cudaPitchedPtr devSrc1, cudaPitchedPtr devSrc2, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuConstMultiply (cudaPitchedPtr devSrc, float c, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuCopy (cudaPitchedPtr devSrc, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuDivide (cudaPitchedPtr devSrc1, cudaPitchedPtr devSrc2, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuMultiply (cudaPitchedPtr devSrc1, cudaPitchedPtr devSrc2, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuNorm (cudaPitchedPtr devSrcX, cudaPitchedPtr devSrcY, cudaPitchedPtr devSrcZ, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuSet (cudaPitchedPtr devDest, float value, int w, int h, int d)
__global__ void	gpuSubstract (cudaPitchedPtr devSrc1, cudaPitchedPtr devSrc2, cudaPitchedPtr devDest, int w, int h, int d)
__global__ void	gpuThreshold (cudaPitchedPtr devSrc, cudaPitchedPtr devDest, float threshold, int w, int h, int d)
template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>
__global__ void	gpuConvolutionX (cudaPitchedPtr devSrc, cudaPitchedPtr devDest, float *kernel, int kernelRadius, int w, int h, int d)
template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>
__global__ void	gpuConvolutionY (cudaPitchedPtr devSrc, cudaPitchedPtr devDest, float *kernel, int kernelRadius, int w, int h, int d)
template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>
__global__ void	gpuConvolutionZ (cudaPitchedPtr devSrc, cudaPitchedPtr devDest, float *kernel, int kernelRadius, int w, int h, int d)
template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerDimension>
__global__ void	gpuReduce (cudaPitchedPtr devSrc, float *devDest, int w, int h, int d)

Function Documentation

gpuAdd()

__global__ void gpuAdd	(	cudaPitchedPtr	devSrc1,
		cudaPitchedPtr	devSrc2,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuConstMultiply()

__global__ void gpuConstMultiply	(	cudaPitchedPtr	devSrc,
		float	c,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuConvolutionX()

template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>

__global__ void gpuConvolutionX	(	cudaPitchedPtr	devSrc,
		cudaPitchedPtr	devDest,
		float *	kernel,
		int	kernelRadius,
		int	w,
		int	h,
		int	d
	)

gpuConvolutionY()

template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>

__global__ void gpuConvolutionY	(	cudaPitchedPtr	devSrc,
		cudaPitchedPtr	devDest,
		float *	kernel,
		int	kernelRadius,
		int	w,
		int	h,
		int	d
	)

gpuConvolutionZ()

template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerThread, int kernelBlockRatio>

__global__ void gpuConvolutionZ	(	cudaPitchedPtr	devSrc,
		cudaPitchedPtr	devDest,
		float *	kernel,
		int	kernelRadius,
		int	w,
		int	h,
		int	d
	)

gpuCopy()

__global__ void gpuCopy	(	cudaPitchedPtr	devSrc,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuDivide()

__global__ void gpuDivide	(	cudaPitchedPtr	devSrc1,
		cudaPitchedPtr	devSrc2,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuMultiply()

__global__ void gpuMultiply	(	cudaPitchedPtr	devSrc1,
		cudaPitchedPtr	devSrc2,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuNorm()

__global__ void gpuNorm	(	cudaPitchedPtr	devSrcX,
		cudaPitchedPtr	devSrcY,
		cudaPitchedPtr	devSrcZ,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuReduce()

template<int blockDimX, int blockDimY, int blockDimZ, int pixelsPerDimension>

__global__ void gpuReduce	(	cudaPitchedPtr	devSrc,
		float *	devDest,
		int	w,
		int	h,
		int	d
	)

gpuSet()

__global__ void gpuSet	(	cudaPitchedPtr	devDest,
		float	value,
		int	w,
		int	h,
		int	d
	)

gpuSubstract()

__global__ void gpuSubstract	(	cudaPitchedPtr	devSrc1,
		cudaPitchedPtr	devSrc2,
		cudaPitchedPtr	devDest,
		int	w,
		int	h,
		int	d
	)

gpuThreshold()

__global__ void gpuThreshold	(	cudaPitchedPtr	devSrc,
		cudaPitchedPtr	devDest,
		float	threshold,
		int	w,
		int	h,
		int	d
	)