IP Library › Granted Patent US 10,445,177
Granted Patent B2
US 10,445,177 · App. 16/026,954 · Granted Oct 15, 2019

Controller-based memory scrub for DRAMs with internal error-correcting code (ECC) bits contemporaneously during auto refresh or by using masked write commands

Inventors: David Reed (Santa Clara, CA); Alok Gupta (Santa Clara, CA)
Assignee: Nvidia Corporation
G06F11/1068G06F3/064G06F3/0619G06F3/0659G06F3/0688G06F11/106G11C7/02G11C11/4082G11C11/4087G11C11/4093G11C11/40615G11C29/04G11C2029/0409G11C2029/0411G11C2029/4402G11C2211/4062
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 10,445,177
App. No.
16/026,954
Granted
Oct 15, 2019
Kind
B2
Abstract

A method for updating a DRAM memory array is disclosed. The method comprises: a) transitioning the DRAM memory array from an idle state to a refresh state in accordance with a command from a memory controller; b) initiating a refresh on the DRAM memory array using DRAM internal control circuitry by activating a row of data into an associated sense amplifier buffer; and c) during the refresh, performing an ERR Correction Code (ECC) scrub operation of selected bits in the activated row of the DRAM memory array.

Claims (15)

1. A method of correcting at least one error in a memory core of a Dynamic Random Access Memory (DRAM) by a memory controller, comprising:

initiating, by said memory controller, a request to said DRAM to perform a read-modify-write operation on a set of data bits in said memory core,

wherein said request designates that no bits in said set of data bits are to be modified for said read-modify-write operation; and

wherein said request causes said DRAM to perform steps comprising:

reading said set of data bits and a corresponding set of error correction code (ECC) bits from said memory core;

correcting at least one error in said set of data bits based on said set of ECC bits and without otherwise modifying said set of data bits for said read-modify-write operation to generate a corrected set of data bits; and

storing said corrected set of data bits to said memory core.

2. The method of claim 1 , further comprising wherein said ECC bits is re-computed for said corrected set of data bits.

3. The method of claim 1 , wherein said memory controller employs a refresh column counter to select one or more columns in requesting said read-modify-write operation.

4. The method of claim 3 , further comprising:

incrementing said refresh column counter inside said memory controller,

wherein said incremented refresh column counter indicates a subsequent read-modify-write operation to be requested to said DRAM,

wherein said subsequent read-modify-write operation designates that no bits in a set of data bits corresponding to said incremented refresh column counter are to be modified.

5. The method of claim 1 , wherein said DRAM is a low power double data rate (LPDDR) DRAM.

6. The method of claim 1 , wherein said refresh column counter is incremented by an increment of more than one word.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 9, 2018
From: REED, DAVID; GUPTA, ALOK
To: NVIDIA CORPORATION
Reel/Frame 046294/0863 →
Continuity (2)
Division 14963035 · Dec 8, 2015
Related Publication 20180314591A1 · Nov 1, 2018
Cited By (1)
US 12,360,847