Run-Time parallelization of loops in computer programs using bit vectors

• US 8,028,281 B2
• Filed: 01/05/2007
• Issued: 09/27/2011
• Est. Priority Date: 12/15/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bera, Rajendra K.
• Primary Examiner(s): Kang; Insun
• Application Number:
  US11/620,586
• Publication Number:
  US 20070169061A1
• Time in Patent Office: 1,726 Days
• Field of Search: 717/150, 717/160
• US Class Current: 717/160
• CPC Class Codes: G06F 8/452 Loops

Abstract

Parallelization of loops is performed for loops having indirect loop index variables and embedded conditional statements in the loop body. Loops having any finite number of array variables in the loop body, and any finite number of indirect loop index variables can be parallelized. There are two particular limitations of the described techniques: (i) that there are no cross-iteration dependencies in the loop other than through the indirect loop index variables; and (ii) that the loop index variables (either direct or indirect) are not redefined in the loop body.

Claims

1. A method for executing, by a processor of a computer system, a set of program instructions for a loop, wherein the method comprises:

associating a unique proxy value with each indirect loop index variable of the loop, wherein each unique proxy value is a proxy value bit vector, each proxy value bit vector representing a different prime number;

calculating, for each iteration of the loop, an indirectly indexed access pattern based upon the unique values;

determining whether cross-iteration dependencies exist between any two iterations of the loop based upon the indirectly indexed access patterns of the two iterations;

scheduling the program instructions of the loop across iterations into waves based on the cross-iteration dependencies found; and

executing the waves.

2. The method of claim 1, wherein the indirectly indexed access pattern for each iteration is calculated as an indirectly indexed access pattern bit vector by OR'"'"'ing the proxy value bit vectors associated with the respective indirect loop index variables on a decision path of the loop for that iteration.

3. The method of claim 2, wherein the determining of cross-iteration dependencies is calculated as an AND'"'"'ing result of two indirectly indexed access pattern bit vectors for two respective iterations, wherein a null bit vector of the AND'"'"'ing result indicates that no dependencies exist between the two respective iterations.

4. The method of claim 2, wherein an indirect indexed access pattern is calculated for each possible decision path of the loop.

5. The method of claim 1, wherein the indirectly indexed access patterns for the respective iterations of the loop have pattern values, and wherein for each iteration the pattern values of the indirectly indexed access pattern do not exceed three in number regardless of how many statements are in the loop.

6. The method of claim 1, wherein a set of n prime numbers are enumerated in a series such that one of the prime numbers may be referred to as a kth number of the series, and wherein each proxy value bit vector has n bits having respective logic states and represents a prime number by the logic state of the kth one of the n bits.

7. A computer program product for executing, by a processor of a computer system, a set of program instructions for a loop, the computer program product comprising computer software stored on a tangible, computer-readable storage medium for performing:

associating a unique proxy value with each indirect loop index variable of the loop, wherein each unique proxy value is a proxy value bit vector, each proxy value bit vector representing a different prime number;

calculating, for each iteration of the loop, an indirectly indexed access pattern based upon the unique values;

determining whether cross-iteration dependencies exist between any two iterations of the loop based upon the indirectly indexed access patterns of the two iterations;

scheduling the program instructions of the loop across iterations into waves based on the cross-iteration dependencies found; and

executing the waves.

8. The computer program product of claim 7, wherein the indirectly indexed access pattern for each iteration is calculated as an indirectly indexed access pattern bit vector by OR'"'"'ing the proxy value bit vectors associated with the respective indirect loop index variables on a decision path of the loop for that iteration.

9. The computer program product of claim 8, wherein the determining of cross-iteration dependencies is calculated as an AND'"'"'ing result of two indirectly indexed access pattern bit vectors for two respective iterations, wherein a null bit vector of the AND'"'"'ing result indicates that no dependencies exist between the two respective iterations.

10. The computer program product of claim 8, wherein an indirect indexed access pattern is calculated for each possible decision path of the loop.

11. The computer program product of claim 7, the indirectly indexed access patterns for the respective iterations of the loop have pattern values, and wherein for each iteration the pattern values of the indirectly indexed access pattern do not exceed three in number regardless of how many statements are in the loop.

12. The computer program product of claim 7, wherein a set of n prime numbers are enumerated in a series such that one of the prime numbers may be referred to as a kth number of the series, and wherein each proxy value bit vector has n bits having respective logic states and represents a prime number by the logic state of the kth one of the n bits.

13. A computer system having program instructions stored on a computer-readable medium for executing, by a processor of the computer system, a set of the program instructions for a loop, wherein the executing comprises performing:

associating a unique proxy value with each indirect loop index variable of the loop, wherein each unique proxy value is a proxy value bit vector, each proxy value bit vector representing a different prime number;

calculating, for each iteration of the loop, an indirectly indexed access pattern based upon the unique values;

determining whether cross-iteration dependencies exist between any two iterations of the loop based upon the indirectly indexed access patterns of the two iterations;

scheduling the program instructions of the loop across iterations into waves based on the cross-iteration dependencies found; and

executing the waves.

14. The computer program product of claim 13, wherein the indirectly indexed access pattern for each iteration is calculated as an indirectly indexed access pattern bit vector by OR'"'"'ing the proxy value bit vectors associated with the respective indirect loop index variables on a decision path of the loop for that iteration.

15. The computer program product of claim 14, wherein the determining of cross-iteration dependencies is calculated as an AND'"'"'ing result of two indirectly indexed access pattern bit vectors for two respective iterations, wherein a null bit vector of the AND'"'"'ing result indicates that no dependencies exist between the two respective iterations.

16. The computer program product of claim 14, wherein an indirect indexed access pattern is calculated for each possible decision path of the loop.

17. The computer program product of claim 13, wherein the indirectly indexed access patterns for the respective iterations of the loop have pattern values, and wherein for each iteration the pattern values of the indirectly indexed access pattern do not exceed three in number regardless of how many statements are in the loop.

18. The computer program product of claim 13, wherein a set of n prime numbers are enumerated in a series such that one of the prime numbers may be referred to as a kth number of the series, and wherein each proxy value bit vector has n bits having respective logic states and represents a prime number by the logic state of the kth one of the n bits.