METHODS TO OPTIMIZE A PROGRAM LOOP VIA VECTOR INSTRUCTIONS USING A SHUFFLE TABLE AND A MASK STORE TABLE

• US 20130339682A1
• Filed: 12/15/2011
• Published: 12/19/2013
• Est. Priority Date: 12/15/2011
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Uliel, Tal, Ould-Ahmed-Vall, Elmoustapha, Toll, Bret T.
• Granted Patent: US 9,436,469 B2
• Time in Patent Office: Days
• US Class Current: 712/226
• CPC Class Codes: G06F 9/30 Arrangements for executing machine instructions, e.g. instruction decode for executing microinstructions G06F9/22
• CPC Class Codes: G06F 9/30032 Movement instructions, e.g. MOVE, SHIFT, ROTATE, SHUFFLE
• CPC Class Codes: G06F 9/30036 Instructions to perform operations on packed data, e.g. vector, tile or matrix operations

Abstract

According to one embodiment, a code optimizer is configured to receive first code having a program loop implemented with scalar instructions to store values of a first array to a second array based on values of a third array. The code optimizer is configured to generate second code representing the program loop with vector instructions including a shuffle instruction and a store instruction, the store instruction to shuffle using a shuffle table elements of the first array based on the second array in a vector manner, the store instruction to store using a mask store table the shuffled elements in the third array in a vector manner.

Claims

1. A computer-implemented method, comprising:

receiving first code having a program loop implemented with scalar instructions to store values of a first array to a second array based on values of a third array; and

generating second code representing the program loop with vector instructions, the second code including

a shuffle instruction to shuffle elements of the first array based on the third array using a shuffle table in a vector manner, and

a store instruction to store the shuffled elements of the first array in the second array using a mask store table in a vector manner.

2. The method of claim 1, wherein the second code further comprises instructions to

compare elements of the third array with a predetermined threshold, generating a comparison result, and

generate a mask based on the comparison result, the elements of the first array to be shuffled based on the mask.

3. The method of claim 2, wherein the second code further comprises an instruction to load elements of the shuffle table selected based on the mask, the elements of the first array to be shuffled via the shuffle instruction based on the selected elements of the shuffle table.

4. The method of claim 2, wherein the second code further comprises an instruction to load elements of the mask store table selected based on the mask, the shuffled elements of the first array to be stored via the mask store instruction based on the selected elements of the mask store table.

5. The method of claim 4, wherein the selected elements of the mask store table are determined based on a number of bits of the mask that have a predetermined logical value.

6. The method of claim 2, wherein the second code further comprises an instruction to store the predetermined threshold in a vector register, the elements of the first array to be compared with values stored in the vector register using a vector compare instruction.

7. The method of claim 2, wherein the mask is generated based on sign bits of the comparison result stored in a vector register.

8. A computer-readable storage medium having instructions stored therein, which when executed by a computer, cause the computer to perform a method, the method comprising:

receiving first code having a program loop implemented with scalar instructions to store values of a first array to a second array based on values of a third array; and

generating second code representing the program loop with vector instructions, the second code including

a shuffle instruction to shuffle elements of the first array based on the third array using a shuffle table in a vector manner, and

a store instruction to store the shuffled elements of the first array in the second array using a mask store table in a vector manner.

9. The computer-readable storage medium of claim 8, wherein the second code further comprises instructions to

compare elements of the third array with a predetermined threshold, generating a comparison result, and

generate a mask based on the comparison result, the elements of the first array to be shuffled based on the mask.

10. The computer-readable storage medium of claim 9, wherein the second code further comprises an instruction to load elements of the shuffle table selected based on the mask, the elements of the first array to be shuffled via the shuffle instruction based on the selected elements of the shuffle table.

11. The computer-readable storage medium of claim 9, wherein the second code further comprises an instruction to load elements of the mask store table selected based on the mask, the shuffled elements of the first array to be stored via the mask store instruction based on the selected elements of the mask store table.

12. The computer-readable storage medium of claim 11, wherein the selected elements of the mask store table are determined based on a number of bits of the mask that have a predetermined logical value.

13. The computer-readable storage medium of claim 9, wherein the second code further comprises an instruction to store the predetermined threshold in a vector register, the elements of the first array to be compared with values stored in the vector register using a vector compare instruction.

14. The computer-readable storage medium of claim 9, wherein the mask is generated based on sign bits of the comparison result stored in a vector register.

15. An apparatus, comprising:

a memory to store a shuffle table and a mask store table; and

a code optimizer to receive first code having a program loop implemented with scalar instructions to store values of a first array to a second array based on values of a third array and to generate second code representing the program loop with vector instructions, the second code including

a shuffle instruction to shuffle elements of the first array based on the third array using the shuffle table in a vector manner, and

a store instruction to store the shuffled elements of the first array in the second array using the mask store table in a vector manner.

16. The apparatus of claim 15, wherein the second code further comprises instructions to

compare elements of the third array with a predetermined threshold, generating a comparison result, and

generate a mask based on the comparison result, the elements of the first array to be shuffled based on the mask.

17. The apparatus of claim 16, wherein the second code further comprises an instruction to load elements of the shuffle table selected based on the mask, the elements of the first array to be shuffled via the shuffle instruction based on the selected elements of the shuffle table.

18. The apparatus of claim 16, wherein the second code further comprises an instruction to load elements of the mask store table selected based on the mask, the shuffled elements of the first array to be stored via the mask store instruction based on the selected elements of the mask store table.

19. The apparatus of claim 18, wherein the selected elements of the mask store table are determined based on a number of bits of the mask that have a predetermined logical value.

20. The apparatus of claim 16, wherein the second code further comprises an instruction to store the predetermined threshold in a vector register, the elements of the first array to be compared with values stored in the vector register using a vector compare instruction.