Method and system for optimizing performance based on cache analysis

• US 8,266,605 B2
• Filed: 02/22/2006
• Issued: 09/11/2012
• Est. Priority Date: 02/22/2006
• Status: Active Grant

• Current Assignee: Wind River Systems Incorporated (Aptiv PLC)
• Original Assignee: Wind River Systems Incorporated (Aptiv PLC)
• Inventors: Wiles, Roger, Koning, Maarten
• Primary Examiner(s): Nguyen, Phillip H
• Application Number:
  US11/359,706
• Publication Number:
  US 20070240117A1
• Time in Patent Office: 2,393 Days
• Field of Search: 717/151, 717/156, 717158-162, 717/141, 717/145, 717152-154, 717/166
• US Class Current: 717/151
• CPC Class Codes: G06F 11/3457 Performance evaluation by simulation
• CPC Class Codes: G06F 2201/885 Monitoring specific for caches
• CPC Class Codes: G06F 8/4442 Reducing the number of cache misses; Data prefetching cache prefetching G06F12/0862

Abstract

Described is a method and system for optimizing a code layout for execution on a processor including internal and/or external cache memory. The method and system includes executing a program having a first layout, generating at least one memory access parameter for the program, the memory access parameter being based on a cache memory of a computing system on which the program is designed to run and constructing a second layout for the program as a function of the at least one memory access parameter.

Claims

1. A system including a memory to store a set of instructions executable by a processor, the set of instructions being operable to:

compile and execute a program having a first layout of source code routines;

generate at least one memory access parameter for the program, the memory access parameter being based on a cache memory of a computing system on which the program is designed to run, the at least one memory access parameter including at least one of execution frequency data and execution criticality data; and

recompile and execute the program having a second layout of source code routines, wherein the at least one memory access parameter is used by a linker to construct the second layout for the program based on at least one of the execution frequency data and the execution criticality data, and wherein a loader rearranges the second layout of the source code routines based on the at least one of the execution frequency data and the execution criticality data to perform loading.

2. The system according to claim 1, wherein the at least one memory access parameter is one of cache access data and execution path data.

3. The system according to claim 1, wherein each of the first and second layouts are one of a code layout and a data layout.

4. The system according to claim 1, wherein the generating the at least one memory access parameter includes:

analyzing accesses to the cache memory.

5. The system according to claim 1, wherein the method is performed on a second computing system.

6. The system according to claim 5, wherein the second computing system is a hardware simulator with instruction execution capabilities.

7. The system according to claim 6, further comprising:

identifying cache lines hit for the preselected path.

8. The system according to claim 1, wherein the generating the at least one memory access parameter includes:

counting a number of times a preselected path in the program is executed.

9. The system according to claim 1, wherein the at least one memory access parameter includes at least one of (i) cache hits, (ii) cache misses, (iii) cache collisions, (iv) contention for a predetermined cache line, (v) a processing time for a predetermined portion of the program (vi) a location in the program of a processor at a predetermined interval and (vii) a manually tagged portion of the program.

10. The system according to claim 1, further comprising:

generating at least one further memory access parameter for the program in the second layout; and

comparing the at least one further memory access parameter to an optimal cache usage profile to determine whether further cache data should be collected.

11. A system, comprising:

a memory to store a program; and

a processor including;

a hardware simulator component to compile and execute the program having a first layout of source code routines, the hardware simulator component generating at least one memory access parameter for the program, the memory access parameter being based on a cache memory of a computing system on which the program is designed to run, the at least one memory access parameter including at least one of execution frequency data and execution criticality data;

a linker component to construct a second layout of source code routines for the program based on at least one of the execution frequency data and the execution criticality data, wherein the hardware simulator component recompiles and executes the program having the second layout of source code; and

a loader rearranges the second layout of the source code routines based on the at least one of the execution frequency data and the execution criticality data to perform loading.

12. The system according to claim 11, wherein the hardware simulator component analyzes accesses to the cache memory as the program is executing.

13. The system according to claim 11, wherein the hardware simulator component counts a number of times a preselected path in the program is executed.

14. The system according to claim 13, wherein the hardware simulator component identifies cache lines hit for the preselected path.

15. The system according to claim 11, wherein the at least one memory access parameter includes at least one of (i) cache hits, (ii) cache misses, (iii) cache collisions, (iv) contention for a predetermined cache line, (v) a processing time for a predetermined portion of the program (vi) a location in the program of a processor at a predetermined interval and (vii) a manually tagged portion of the program.

16. The system according to claim 11, wherein the hardware simulator component generates at least one further memory access parameter for the program in the second layout and compares the at least one further memory access parameter to an optimal cache usage profile.

17. The system according to claim 16, wherein the comparison to the optimal cache usage profile determines whether further cache data should be collected.

18. The system according to claim 11, wherein the computing system on which the program is designed to run includes the cache memory and a main memory, wherein the second code layout has less cache line contentions than the first code layout.

19. The system according to claim 11, wherein the computing system on which the program is designed to run includes the cache memory and a main memory, wherein the second code layout has less cache collisions than the first code layout.