Enqueuing kernels from kernels on GPU/CPU
Graphics processing units (GPUs) and other compute units are allowed to enqueue tasks for themselves by themselves, without needing a host processor to queue the work for the GPU. Built-in functions enable kernels to enqueue kernels for execution on a device. In some embodiments, ndrange kernels execute over an N-dimensional range to provide data-parallel operations. Task kernels provide task-parallel operations. In some embodiments, kernels may be defined using clang block syntax. The order of execution of commands on a compute unit may be constrained or allow execution of commands out-of-order. Compute units may control when kernels enqueued by the compute unit begins execution.
1. A program storage device, on which are stored instructions, comprising instructions that when executed cause one or more compute units to:
enqueue a first kernel by a first compute unit for execution on a second compute unit, wherein the first and second compute units have different capabilities;
determine, based on the execution of the first kernel, that a condition is met; and
in response to the condition being met based on the execution of the first kernel, enqueue a second kernel for execution on the second compute unit.
2. The program storage device of claim 1 , wherein the first compute unit is a central processing unit (CPU) and the second compute unit is a graphic processing unit (GPU).
3. The program storage device of claim 1 , wherein the second kernel is enqueued after execution of the first kernel is complete.
4. The program storage device of claim 1 , wherein the second kernel is enqueued during execution of the first kernel.
5. The program storage device of claim 1 , wherein the first kernel is a data-parallel kernel, and wherein the second kernel is a data-parallel kernel with a different range than the first kernel.
6. The program storage device of claim 1 , wherein the first kernel is a task-parallel kernel, and wherein the second kernel is a task-parallel kernel.
7. The program storage device of claim 1 , wherein the first kernel is a task-parallel kernel, and wherein the second kernel is a data-parallel kernel.
8. The program storage device of claim 1 , wherein the first kernel is a data-parallel kernel, and wherein the second kernel is a task-parallel kernel.
9. The program storage device of claim 1 , wherein the second compute unit enqueues a barrier on a queue of commands to blocks execution of commands enqueued on the queue of commands after the barrier until the barrier completes.
10. The program storage device of claim 1 , wherein the second compute unit enqueues a marker on a queue of commands that does not complete until one or more other commands completes.
11. A computing device, comprising:
one or more compute units; and
a global memory, coupled to the one or more compute units, on which are stored instructions that when executed cause the one or more compute units to:
enqueue a first kernel by a first compute unit for execution on a second compute unit, wherein the first and second compute units have different capabilities;
determine, based on the execution of the first kernel, that a condition is met; and
in response to the condition being met based on the execution of the first kernel, enqueue a second kernel for execution on the second compute unit.
12. The computing device of claim 11 , wherein the first compute unit is a central processing unit (CPU) and the second compute unit is a graphic processing unit (GPU).
13. The computing device of claim 11 , wherein the second kernel is enqueued after execution of the first kernel is complete.
14. The computing device of claim 11 , wherein the second kernel is enqueued during execution of the first kernel.
15. The computing device of claim 11 , wherein the first kernel is a data-parallel kernel, and wherein the second kernel is a data-parallel kernel with a different range than the first kernel.
16. The computing device of claim 11 , wherein the first kernel is a task-parallel kernel, and wherein the second kernel is a task-parallel kernel.
17. The computing device of claim 11 , wherein the first kernel is a task-parallel kernel, and wherein the second kernel is a data-parallel kernel.
18. The computing device of claim 11 , wherein the first kernel is a data-parallel kernel, and wherein the second kernel is a task-parallel kernel.
19. The computing device of claim 11 , wherein the second compute unit enqueues a barrier on a queue of commands to blocks execution of commands enqueued on the queue of commands after the barrier until the barrier completes.
20. The computing device of claim 11 , wherein the second compute unit enqueues a marker on a queue of commands that does not complete until one or more other commands completes.