IP Library › Granted Patent US 11,355,170
Granted Patent B1
US 11,355,170 · App. 17/123,829 · Granted Jun 7, 2022

Reconfigurable processing-in-memory logic

Inventor: Dmitri Yudanov (Rancho Cordova, CA)
Assignee: Micron Technology, Inc.
G11C8/06G11C8/08H03K19/01742
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 11,355,170
App. No.
17/123,829
Granted
Jun 7, 2022
Kind
B1
Abstract

An example system implementing a processing-in-memory pipeline includes: a memory array to store data in a plurality of memory cells electrically coupled to a plurality of wordlines and a plurality of bitlines; a logic array coupled to the memory array, the logic array to implement configurable logic controlling the plurality of memory cells; and a control block coupled to the memory array and the logic array, the control block to control a computational pipeline to perform computations on the data by activating at least one of: one or more bitlines of the plurality of bitlines or one or more wordlines of the plurality of wordlines.

Claims (39)

1. A system, comprising:

a memory array to store data in a plurality of memory cells electrically coupled to a plurality of wordlines and a plurality of bitlines;

a logic array coupled to the memory array, the logic array to implement a configurable logic controlling the plurality of memory cells, wherein the configurable logic comprises a first pull network that is connected via a logical inverter to a second pull network; and

a control block coupled to the memory array and the logic array, the control block to control a computational pipeline to perform computations on the data by activating at least one of: one or more bitlines of the plurality of bitlines or one or more wordlines of the plurality of wordlines.

2. The system of claim 1 , wherein the control block is further to:

pre-charge a selected bitline of the plurality of bitlines, wherein at least one wordline of the plurality of wordlines is open during the pre-charge operation; and

evaluate, using the configurable logic, a row of memory cells addressed by a selected wordline.

3. The system of claim 2 , wherein the control block is further to:

store, in an available memory cell accessible via a wordline of the plurality of wordlines, a state of the selected bitline.

4. The system of claim 1 , wherein the first pull network is provided by one of: a pull-up network (PUN) or a pull-down network (PDN).

5. The system of claim 1 , wherein the configurable logic is controlled by one or more inputs received from the control block.

6. The system of claim 1 , wherein the configurable logic is controlled by one or more inputs received from the memory array.

7. The system of claim 1 , wherein the configurable logic is controlled by one or more inputs received from an external source.

8. The system of claim 1 , wherein the system further comprises a sense amplifier array, and wherein the logic array is integrated with the sense amplifier array.

9. The system of claim 1 , wherein the computational pipeline comprises a systolic array.

10. The system of claim 1 , implemented as a system-on-chip, the system further comprising:

at least one processing core; and

at least one input/output (I/O) interfaces.

11. A system, comprising:

a memory array to store data in a plurality of memory cells electrically coupled to a plurality of wordlines and a plurality of bitlines, the memory array to implement configurable logic controlling the plurality of memory cells, wherein the configurable logic comprises a first pull network that is connected via a logical inverter to a second pull network; and

a control block coupled to the memory array, the control block to control a computational pipeline to perform computations on the data by activating at least one of: one or more bitlines of the plurality of bitlines or one or more wordlines of the plurality of wordlines.

12. The system of claim 11 , wherein the control block is further to:

pre-charge a selected bitline of the plurality of bitlines, wherein at least one wordline of the plurality of wordlines is open during the pre-charge operation; and

evaluate, using the configurable logic, a row of memory cells addressed by a selected wordline.

13. The system of claim 12 , wherein the control block is further to:

store, in an available memory cell accessible via a wordline of the plurality of wordlines, a state of the selected bitline.

14. The system of claim 11 , wherein the control block is further to:

pre-charge a selected bitline of the plurality of bitlines, wherein a first wordline of the plurality of wordlines is open during the pre-charge operation;

pre-charge a second wordline using a voltage level specified by an input supplied by the control block;

evaluate, using the configurable logic, a row of memory cells addressed by a selected wordline.

15. The system of claim 11 , wherein the first pull network is provided by one of: a pull-up network (PUN) or a pull-down network (PDN).

16. A method, comprising:

implementing a configurable logic by storing data in a memory array comprising a plurality of memory cells electrically coupled to a plurality of wordlines and a plurality of bitlines, wherein the configurable logic comprises a first pull network that is connected via a logical inverter to a second pull network;

activating, by a control block coupled to the memory array, a selected bitline of the plurality of bitlines;

pre-charging the selected bitline, wherein at least one wordline of the plurality of wordlines is open during the pre-charge operation; and

evaluating, using the configurable logic, a row of memory cells addressed by a selected wordline.

17. The method of claim 16 , further comprising:

storing, in an available memory cell accessible via a wordline of the plurality of wordlines, a state of the selected bitline.

18. The method of claim 16 , wherein the first pull network is provided by one of: a pull-up network (PUN) or a pull-down network (PDN).

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 10, 2023
From: MICRON TECHNOLOGY, INC.
To: LODESTAR LICENSING GROUP LLC
Reel/Frame 065519/0704 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 16, 2020
From: YUDANOV, DMITRI
To: MICRON TECHNOLOGY, INC.
Reel/Frame 054669/0976 →
Cited By (5)
US 12,197,378 US 12,493,577 US 12,498,931 US 12,511,128 US 12,529,807