IP Library Granted Patent US 10,970,043
Granted Patent B2
US 10,970,043 · App. 16/886,613 · Granted Apr 6, 2021

Programmable multiply-add array hardware

Inventors: Liang Han (San Mateo, CA); Xiaowei Jiang (San Mateo, CA)
Assignee: ALIBABA GROUP HOLDING LIMITED
G06F7/5338G06F7/5318G06F7/544G06F7/5443G06F7/607G06F9/30014G06F9/30101G06N3/063
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 10,970,043
App. No.
16/886,613
Granted
Apr 6, 2021
Kind
B2
Abstract

An integrated circuit including a data architecture including N adders and N multipliers configured to receive operands. The data architecture receives instructions for selecting a data flow between the N multipliers and the N adders of the data architecture. The selected data flow includes the options: (1) a first data flow using the N multipliers and the N adders to provide a multiply-accumulate mode and (2) a second data flow to provide a multiply-reduce mode.

Claims (26)

1. A method for operating a data architecture including N adders and N multipliers configured to operate in one of a multiply-reduce mode or a multiply-accumulate mode, the method comprising:

selecting a data flow between the N multipliers and the N adders of the data architecture, wherein the N multipliers and at least some of the N adders of the data architecture are used both in the multiply-reduce mode and the multiply-accumulate mode and wherein the data architecture is configured to change a connection between the N multipliers and the N adders according to the selected data flow, and

operating the data architecture in one of the multiply-reduce mode or the multiply-accumulate mode according to the selected data flow.

2. The method of claim 1 , wherein selecting the data flow includes selecting a first data flow using the N multipliers and N−1 adders of the N adders, wherein one of the N adders is not used.

3. The method of claim 2 , wherein the first data flow comprises the N−1 adders receiving input resulting from the N multipliers.

4. The method of claim 1 , wherein selecting the data flow includes selecting a second data flow using the N multipliers and the N adders.

5. The method of claim 4 , wherein the second data flow comprises each adder of the N adders receiving an input operand from a corresponding multiplier of the N multipliers.

6. The method of claim 1 , wherein the N multipliers includes a first multiplier of which output data is provided to a first adder among the at least some of the N adders in the multiply-reduce mode and to a second adder among the at least some of the N adders in the multiply-accumulate mode.

7. An integrated circuit comprising:

a data architecture including N adders and N multipliers, wherein the data architecture is configured to operate in one of a multiply-reduce mode or a multiply-accumulate mode according to a selected data flow between the N multipliers and the N adders of the data architecture and wherein the data architecture is configured to change a connection between the N multipliers and the N adders according to the selected data flow, the selected data flow comprises:

a first data flow associated with the multiply-accumulate mode; and

a second data flow associated with the multiply-reduce mode,

wherein the N multipliers and the at least some of the N adders are used both in the first data flow and the second data flow.

8. The integrated circuit of claim 7 , wherein the first data flow uses the N multipliers and N−1 adders of the N adders, wherein one of the N adders is not used.

9. The integrated circuit of claim 8 , wherein the first data flow uses the N−1 adders to receive input resulting from the N multipliers.

10. The integrated circuit of claim 7 , wherein the second data flow uses the N multipliers and the N adders.

11. The integrated circuit of claim 10 , wherein the second data flow uses each adder of the N adders to receive an input operand from a corresponding multiplier of the N multipliers.

12. The integrated circuit of claim 7 , wherein the N multipliers include a first multiplier of which output data is provided to a first adder among the at least some of the N adders in the first data flow and to a second adder among the at least some of the N adders in the second data flow.

13. A non-transitory computer-readable storage medium that stores a set of instructions that is executable by at least one processor of a device to cause the device to perform a method for operating a data architecture including N adders and N multipliers configured to operate in one of a multiply-reduce mode or a multiply-accumulate mode, the method comprising:

selecting a data flow between the N multipliers and the N adders of the data architecture, wherein the N multipliers and at least some of the N adders of the data architecture are used both in the multiply-reduce mode and the multiply-accumulate mode and wherein the data architecture is configured to change a connection between the N multipliers and the N adders according to the selected data flow, and

operating the data architecture in one of the multiply-reduce mode or the multiply-accumulate mode according to the selected data flow.

14. The non-transitory computer-readable storage medium of claim 13 , wherein selecting the data flow includes selecting a first data flow using the N multipliers and N−1 adders of the N adders, wherein one of the N adders is not used.

15. The non-transitory computer-readable storage medium of claim 14 , wherein the first data flow comprises the N−1 adders receiving input resulting from the N multipliers.

16. The non-transitory computer-readable storage medium of claim 14 , wherein the N multipliers includes a first multiplier of which output data is provided to a first adder among the at least some of the N adders in the multiply-reduce mode and to a second adder among the at least some of the N adders in the multiply-accumulate mode.

17. The non-transitory computer-readable storage medium of claim 13 , wherein selecting the data flow includes selecting a second data flow using the N multipliers and the N adders.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the second data flow comprises each adder of the N adders receiving an input operand from a corresponding multiplier of the N multipliers.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 5, 2024
From: ALIBABA GROUP HOLDING LIMITED
To: T-HEAD (SHANGHAI) SEMICONDUCTOR CO., LTD.
Reel/Frame 066348/0656 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 28, 2020
From: HAN, LIANG; JIANG, XIAOWEI
To: ALIBABA GROUP HOLDING LIMITED
Reel/Frame 052781/0196 →
Continuity (3)
Continuation 16054783 · Aug 3, 2018
Provisional Application 62610052 · Dec 22, 2017
Related Publication 20200293283A1 · Sep 17, 2020