cuda矩阵转置

一直觉得经典的cuda矩阵转置只能用于矩阵的宽高都能被线程块大小整除的状况，也是很奇怪，不知道怎么造成这个概念的，而后此次又要用到，本想着大干一番，把宽高不能被线程块整除的矩阵转置攻克了，但是没想到一测试，人原本就能够实现，这就尴尬了, 因此在此记录下来，纠正本身的这个错误。

#define BLOCK_DIM 16

__global__ void myTranspose(float *indata,float *odata,int height,int width)
{
	__shared__ float block[BLOCK_DIM][BLOCK_DIM + 1];

	unsigned int xindex = blockIdx.x * BLOCK_DIM + threadIdx.x;
	unsigned int yindex = blockIdx.y * BLOCK_DIM + threadIdx.y;

	/*int height1 = height - height%BLOCK_DIM;
	int width1 = width - width%BLOCK_DIM;
*/
	if(xindex < width && yindex < height)
	{
		unsigned int index_in = yindex * width + xindex;
		block[threadIdx.y][threadIdx.x] = indata[index_in];
	}
	__syncthreads();

	xindex = blockIdx.y * BLOCK_DIM + threadIdx.x;
	yindex = blockIdx.x * BLOCK_DIM + threadIdx.y;

	if(xindex < height && yindex < width)
	{
		unsigned int index_out = yindex * height + xindex;
		odata[index_out] = block[threadIdx.x][threadIdx.y];
	}
}