clEnqueueNDRangeKernel参数说明

原文地址：http://blog.csdn.net/gflytu/article/details/7686130

OpenCL编程中的一个核心函数是clEnqueueNDRangeKernel，对于此函数的理解，有利于对数据在host和device之间的传递进行控制。

clEnqueueNDRangeKernel（cl_command_queue queue，cl_kernel kernel，

cl_uint work_dims，const size_t *global_work_offset,

const size_t *global_work_size, const size_t *local_work_size,

cl_uint num_events, const cl_event *wait_list, cl_event *event)

对于参数queue和kernel与clEnqueueTask中的一样，重点是work_dims和global_work_offset、global_work_size、local_work_size这四个新增参数的意义。

（1）work_dims：the number of dimensions in the data ( if you deal with image object, you should probably set work_dims equal 2 or 3. But for buffer objects, you can set whatever dimensionality you think best. For a buffer object containing a two-dimensional matrix, you might set work-dims equal 2.)

（2）global_work_offset：the global ID offset in each dimension 

（3）global_work_size：the number of work items in each dimension  (the global_work_size argument of clEnqueueNDRangeKernel identifies how many work-items need to be processed for each dimension. )

（4）local_work_size：the number of work_items in a work_group，in each dimension  (local_work_size less than the global_work_size）

可以通过下面的例子来说明各个参数的具体应用：

size_t dim=2；

size_t global_offset[]={3,5};

size_t global_size[]={6,4};

size_t local_size[]={3,2};

clEnqueueNDRangeKernel（queue,kernel,dim,global_offset,global_size,local_size,0,NULL,NULL);

对于上面的参数我们可以通过以下子函数在kernel里获取这些数据：

uint get_work_dim()：returns the number of dimensions in the kernel's index space

size_t get_global_size(uint dim): returns the number of work items for a given dimension

size_t get_global_id(uint dim):returns the element of the work-dim's global ID for a given dimension

size_t get_global_offset(uint dim):returns the initial offset used to compute global IDs

size_t get_num_groups(uint dim): returns the number of work-groups for a given dimension

size_t get_group_id(uint dim):returns the ID of the work-item's work-group for a given dimension

size_t get_local_id(uint dim): returns the ID of the work-item within its work-group for a given dimension

size_t get_local_size(uint dim): return the number of work-items in the work-group for a given dimension

那么我们可以到如下数据：

uint dim=get_work_dim（）；//dim=2

size_t global_id_0=get_global_id(0);//从参数global_offset(3,5)第一个参数3开始，个数为global_size（6,4）的第一参数6

size_t global_id_1=get_global_id(1);//从参数global_offset(3,5）第二个参数5开始，个数为global_size(6,4)的第二个参数4

size_t global_size_0=get_global_size(0);//大小为global_size(6,4)的第一个参数6

size_t global_size_1=get_global_size(1);//大小为global_size(6,4)的第二个参数4

size_t offset_0=get_global_offset(0);//获取global_offset(3,5)的第一个参数3,

size_t offset_1=get_global_offset(1);//获取global_offset(3,5)的第二个参数5

size_t local_id_0=get_local_id(0);//获取local_size(3,2)的第一个参数个数（0,1,2)

size_t local_id_1=get_local_id(1);//获取local_size(3,2)的第二个参数个数（0,1）