clEnqueueNativeKernel - Enqueues a command to execute a native
C/C++ function not compiled using the OpenCL compiler.
void (*user_func)(void *),
void *args, size_t cb_args,
const cl_mem *mem_list,
const void **args_mem_loc,
const cl_event *event_wait_list,
A valid command-queue. A native user function can only be
executed on a command-queue created on a device that has
CL_EXEC_NATIVE_KERNEL capability set in
CL_DEVICE_EXECUTION_CAPABILITIES as specified in the table of OpenCL
Device Queries for clGetDeviceInfo(3clc).
A pointer to a host-callable user function.
A pointer to the args list that user_func should
be called with.
The size in bytes of the args list that args
The data pointed to by args and cb_args bytes in
size will be copied and a pointer to this copied region will be passed to
user_func. The copy needs to be done because the memory objects
(cl_mem values) that args may contain need to be modified and
replaced by appropriate pointers to global memory. When
clEnqueueNativeKernel returns, the memory region pointed to by
args can be reused by the application.
The number of buffer objects that are passed in
A list of valid buffer objects, if num_mem_objects
> 0. The buffer object values specified in mem_list are memory
object handles (cl_mem values) returned by clCreateBuffer(3clc)
A pointer to appropriate locations that args
points to where memory object handles (cl_mem values) are stored.
Before the user function is executed, the memory object handles are replaced
by pointers to global memory.
Specify events that need to complete before this
particular command can be executed. If event_wait_list is NULL, then
this particular command does not wait on any event to complete. If
event_wait_list is NULL, num_events_in_wait_list must be 0. If
event_wait_list is not NULL, the list of events pointed to by
event_wait_list must be valid and num_events_in_wait_list must
be greater than 0. The events specified in event_wait_list act as
synchronization points. The context associated with events in
event_wait_list and command_queue must be the same. The memory
associated with event_wait_list can be reused or freed after the
The total number of read-only images specified as arguments to a
kernel cannot exceed CL_DEVICE_MAX_READ_IMAGE_ARGS. Each 2D image
array argument to a kernel declared with the read_only qualifier counts as
The total number of write-only images specified as arguments to a
kernel cannot exceed CL_DEVICE_MAX_WRITE_IMAGE_ARGS. Each 2D image
array argument to a kernel declared with the write-only qualifier counts as
Returns CL_SUCCESS if the user function execution instance was
successfully queued, or one of the errors below:
•CL_INVALID_COMMAND_QUEUE if command_queue
is not a valid command-queue.
•CL_INVALID_CONTEXT if context associated with
command_queue and events in event-wait_list are not the
•CL_INVALID_VALUE if user_func is
•CL_INVALID_VALUE if args is a NULL value
and cb_args > 0, or if args is a NULL value and
num_mem_objects > 0.
•CL_INVALID_VALUE if args is not NULL and
cb_args is 0.
•CL_INVALID_VALUE if num_mem_objects > 0
and mem_list or args_mem_loc are NULL.
•CL_INVALID_VALUE if num_mem_objects = 0
and mem_list or args_mem_loc are not NULL.
•CL_INVALID_OPERATION if the device associated
with command_queue cannot execute the native kernel.
•CL_INVALID_MEM_OBJECT if one or more memory
objects specified in mem_list are not valid or are not buffer
•CL_OUT_OF_RESOURCES if there is a failure to
queue the execution instance of kernel on the command-queue because of
insufficient resources needed to execute the kernel.
•CL_MEM_OBJECT_ALLOCATION_FAILURE if there is a
failure to allocate memory for data store associated with buffer objects
specified as arguments to kernel.
event_wait_list is NULL and num_events_in_wait_list > 0, or
event_wait_list is not NULL and num_events_in_wait_list is 0, or
if event objects in event_wait_list are not valid events.
•CL_OUT_OF_RESOURCES if there is a failure to
allocate resources required by the OpenCL implementation on the device.
•CL_OUT_OF_HOST_MEMORY if there is a failure to
allocate resources required by the OpenCL implementation on the host.
The Khronos Group
Copyright © 2007-2011 The Khronos Group Inc.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and/or associated documentation files (the
"Materials"), to deal in the Materials without restriction,
including without limitation the rights to use, copy, modify, merge,
publish, distribute, sublicense, and/or sell copies of the Materials, and to
permit persons to whom the Materials are furnished to do so, subject to the
condition that this copyright notice and permission notice shall be included
in all copies or substantial portions of the Materials.
- OpenCL Specification
page 174, section 5.8 - Executing Kernels (updated for
1.2 rev 14)