IP Library Granted Patent US 12,061,905
Granted Patent B1
US 12,061,905 · App. 18/337,869 · Granted Aug 13, 2024

Transformations in fused multiply-add instructions

Inventors: John Hayes (Mountain View, CA); Volkmar Uhlig (Cupertino, CA)
Assignee: GHOST AUTONOMY INC.
G06F9/3001G06F9/30181
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,061,905
App. No.
18/337,869
Granted
Aug 13, 2024
Kind
B1
Abstract

Transformations in fused multiply-add instructions, including: receiving an extended fused multiply-add (FMA) instruction comprising: a first subset of bits corresponding to each operand of a fused multiply-add (FMA) operation, and a second subset of bits comprising an opcode for the extended FMA instruction, wherein the opcode identifies extended FMA instruction from an instruction set of a plurality of extended FMA instructions each having a different predefined opcode corresponding to a different combination of transformation-operand groupings; and performing, based on the opcode of the extended FMA instruction, the one or more transformations and the FMA operation.

Claims (26)

1. A method comprising:

receiving an extended fused multiply-add (FMA) instruction comprising: a first subset of bits corresponding to each operand of a fused multiply-add (FMA) operation, and a second subset of bits comprising an opcode for the extended FMA instruction, wherein the opcode identifies extended FMA instruction from an instruction set of a plurality of extended FMA instructions each having a different predefined opcode corresponding to a different combination of transformation-operand groupings; and

performing, based on the opcode of the extended FMA instruction, the one or more transformations and the FMA operation.

2. The method of claim 1 , wherein the one or more transformations comprise a bit masking transformation.

3. The method of claim 1 , wherein the one or more transformations comprise a bit shifting operation.

4. The method of claim 1 , wherein the one or more transformations comprise a rounding operation.

5. The method of claim 1 , wherein the one or more transformations comprise a casting operation.

6. The method of claim 1 , wherein the one or more transformations comprise a conditional zeroing operation.

7. The method of claim 1 , wherein the first subset of bits indicates one or more registers storing one or more of the operands.

8. The method of claim 1 , wherein the first subset of bits indicates one or more of the operands.

9. An apparatus configured to perform steps comprising:

receiving an extended fused multiply-add (FMA) instruction comprising: a first subset of bits corresponding to each operand of a fused multiply-add (FMA) operation, and a second subset of bits comprising an opcode for the extended FMA instruction, wherein the opcode identifies extended FMA instruction from an instruction set of a plurality of extended FMA instructions each having a different predefined opcode corresponding to a different combination of transformation-operand groupings; and

performing, based on the opcode of the extended FMA instruction, the one or more transformations and the FMA operation.

10. The apparatus of claim 9 , wherein the one or more transformations comprise a bit masking transformation.

11. The apparatus of claim 9 , wherein the one or more transformations comprise a bit shifting operation.

12. The apparatus of claim 9 , wherein the one or more transformations comprise a rounding operation.

13. The apparatus of claim 9 , wherein the one or more transformations comprise a casting operation.

14. The apparatus of claim 9 , wherein the one or more transformations comprise a conditional zeroing operation.

15. The apparatus of claim 9 , wherein the first subset of bits indicates one or more registers storing one or more of the operands.

16. The apparatus of claim 9 , wherein the first subset of bits indicates one or more of the operands.

17. A computer program product disposed upon a non-transitory computer readable medium, the computer program product comprising computer program instructions that, when executed, cause a computer system to perform steps comprising:

receiving an extended fused multiply-add (FMA) instruction comprising: a first subset of bits corresponding to each operand of a fused multiply-add (FMA) operation, and a second subset of bits comprising an opcode for the extended FMA instruction, wherein the opcode identifies extended FMA instruction from an instruction set of a plurality of extended FMA instructions each having a different predefined opcode corresponding to a different combination of transformation-operand groupings; and

performing, based on the opcode of the extended FMA instruction, the one or more transformations and the FMA operation.

18. The computer program product of claim 17 , wherein the one or more transformations comprise a bit masking transformation.

19. The computer program product of claim 17 , wherein the one or more transformations comprise a bit shifting operation.

20. The computer program product of claim 17 , wherein the one or more transformations comprise a rounding operation.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 18, 2024
From: GHOST AUTONOMY, INC.
To: APPLIED INTUITION, INC.
Reel/Frame 068982/0647 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 20, 2023
From: HAYES, JOHN; UHLIG, VOLKMAR
To: GHOST AUTONOMY INC.
Reel/Frame 063998/0322 →
Continuity (2)
Continuation 17685340 · Mar 2, 2022
Continuation 16942104 · Jul 29, 2020