TECHNIQUES FOR SHARING PRIORITIES BETWEEN STREAMS OF WORK AND DYNAMIC PARALLELISM

• US 20140344821A1
• Filed: 05/17/2013
• Published: 11/20/2014
• Est. Priority Date: 05/17/2013
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: KINI, Vivek, LAMB, Christopher
• Granted Patent: US 9,575,760 B2
• Time in Patent Office: Days
• US Class Current: 718/103
• CPC Class Codes: G06F 2209/5021 Priority
• CPC Class Codes: G06F 9/38 Concurrent instruction execution, e.g. pipeline or look ahead
• CPC Class Codes: G06F 9/4843 by program, e.g. task dispatcher, supervisor, operating system
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5038 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration scheduling strategies G06F9/4881 and subgroups
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining

Abstract

One embodiment sets forth a method for assigning priorities to kernels launched by a software application and executed within a stream of work on a parallel processing subsystem that supports dynamic parallelism. First, the software application assigns a maximum nesting depth for dynamic parallelism. The software application then assigns a stream priority to a stream. These assignments cause a driver to map the stream priority to a device priority and, subsequently, associate the device priority with the stream. As part of the mapping, the driver ensures that each device priority is at least the maximum nesting depth higher than the device priorities associated with any lower priority streams. Subsequently, the driver launches any kernel included in the stream with the device priority associated with the stream. Advantageously, by strategically assigning the maximum nesting depth and prioritizing streams, an application developer may increase the overall processing efficiency of the software application.

Claims

1. A method for mapping a plurality of stream priorities associated with a software application to a plurality of device priorities supported by a parallel processing subsystem, the method comprising:

identifying a maximum nesting depth;

receiving a first request from the software application that comprises a request to associate a first stream with a first stream priority;

mapping the first stream priority to a first device priority;

receiving a second request from the software application that comprises a request to associate a second stream with a second stream priority, wherein the second stream priority is higher in priority than the first stream priority; and

mapping the second stream priority to a second device priority, wherein the second device priority is at least the maximum nesting depth higher in priority than the first device priority.

2. The method of claim 1, wherein identifying the maximum nesting depth comprises:

receiving a request from the software application specifying the maximum nesting depth;

storing the maximum nesting depth in a memory resource; and

subsequently accessing the memory resource to read the maximum nesting depth.

3. The method of claim 1, wherein the maximum nesting depth comprises a default maximum nesting depth.

4. The method of claim 1, further comprising:

receiving a third request from the software application that comprises a request to launch a first work component within the second stream; and

submitting the first work component to the parallel processing subsystem for execution within the second stream at the second device priority.

5. The method of claim 4, wherein the first work component comprises a first function that is executable via a plurality of parallel threads.

6. The method of claim 5, wherein the first function, when executed within the parallel processing subsystem, is configured to cause the parallel processing subsystem to execute a second function.

7. The method of claim 1, further comprising:

receiving a third request from the software application that comprises a request to associate a third stream with a third stream priority, wherein the third stream priority is higher in priority than the second stream priority;

identifying that the highest device priority supported by the parallel processing subsystem is separated from the second device priority by less than twice the maximum nesting depth; and

mapping the third stream priority to the second device priority.

8. The method of claim 1, wherein the maximum nesting depth is less than a maximum number of execution levels supported by the parallel processing subsystem.

9. The method of claim 1, wherein the maximum nesting depth is equal to a maximum number of execution levels supported by the parallel processing subsystem.

10. A computer-readable storage medium including instructions that, when executed by a processing unit, cause the processing unit to map a plurality of stream priorities associated with a software application to a plurality of device priorities supported by a parallel processing subsystem by performing the steps of:

identifying a maximum nesting depth;

receiving a first request from the software application that comprises a request to associate a first stream with a first stream priority;

mapping the first stream priority to a first device priority;

receiving a second request from the software application that comprises a request to associate a second stream with a second stream priority, wherein the second stream priority is higher in priority than the first stream priority; and

mapping the second stream priority to a second device priority, wherein the second device priority is at least the maximum nesting depth higher in priority than the first device priority.

11. The computer-readable storage medium of claim 10, wherein identifying the maximum nesting depth comprises:

receiving a request from the software application specifying the maximum nesting depth;

storing the maximum nesting depth in a memory resource; and

subsequently accessing the memory resource to read the maximum nesting depth.

12. The computer-readable storage medium of claim 10, wherein the maximum nesting depth comprises a default maximum nesting depth.

13. The computer-readable storage medium of claim 10, further comprising:

receiving a third request from the software application that comprises a request to launch a first work component within the second stream; and

submitting the first work component to the parallel processing subsystem for execution within the second stream at the second device priority.

14. The computer-readable storage medium of claim 13, wherein the first work component comprises a first function that is executable via a plurality of parallel threads.

15. The computer-readable storage medium of claim 14, wherein the first function, when executed within the parallel processing subsystem, is configured to cause the parallel processing subsystem to execute a second function.

16. The computer-readable storage medium of claim 10, further comprising:

receiving a third request from the software application that comprises a request to associate a third stream with a third stream priority, wherein the third stream priority is higher in priority than the second stream priority;

identifying that the highest device priority supported by the parallel processing subsystem is separated from the second device priority by less than twice the maximum nesting depth; and

mapping the third stream priority to the second device priority.

17. The computer-readable storage medium of claim 10, wherein the maximum nesting depth is less than a maximum number of execution levels supported by the parallel processing subsystem.

18. The computer-readable storage medium of claim 10, wherein the maximum nesting depth is equal to a maximum number of execution levels supported by the parallel processing subsystem.

19. A system configured to map a plurality of stream priorities associated with a software application to a plurality of device priorities supported by a parallel processing subsystem, the system comprising:

a driver program configured to;

identify a maximum nesting depth;

receive a first request from the software application that comprises a request to associate a first stream with a first stream priority;

map the first stream priority to a first device priority;

receive a second request from the software application that comprises a request to associate a second stream with a second stream priority, wherein the second stream priority is higher in priority than the first stream priority; and

map the second stream priority to a second device priority, wherein the second device priority is at least the maximum nesting depth higher in priority than the first device priority.

20. The system of claim 19, wherein the driver program is further configured to:

receive a third request from the software application that comprises a request to launch a first work component within the second stream; and

submit the first work component to the parallel processing subsystem for execution within the second stream at the second device priority.