Unsuccessful write retry buffer
AssigneeRambus Inc.
InventorsHongzhong Zheng, Brent Haukness
Life of this patent
Estimating the term…
Abstract
A memory module includes at least two memory devices. Each of the memory devices perform verify operations after attempted writes to their respective memory cores. When a write is unsuccessful, each memory device stores information about the unsuccessful write in an internal write retry buffer. The write operations may have only been unsuccessful for one memory device and not any other memory devices on the memory module. When the memory module is instructed, both memory devices on the memory module can retry the unsuccessful memory write operations concurrently. Both devices can retry these write operations concurrently even though the unsuccessful memory write operations were to different addresses.
Claims20
A memory device, comprising:
- a first memory bank to be operated in a manner such that a first portion of attempted writes to the first memory bank are determined by corresponding verify operations to have failed in a manner that causes loss of data;
- a second memory bank to be operated in the manner such that a second portion of attempted writes to the second memory bank are determined by corresponding verify operations to have failed in the manner that causes loss of data;
- a first write retry buffer having a first plurality of entries to each store failed write information associated with the first memory bank, the first plurality of entries to include a first entry storing first failed write information associated with a first failed write operation that is a one of the first portion of attempted writes to the first memory bank that has been determined to have failed;
- a second write retry buffer having a second plurality of entries to each store failed write information associated with the second memory bank, the second plurality of entries to include a second entry storing second failed write information associated with a second failed write operation that is a one of the second portion of attempted writes to the second memory bank that has been determined to have failed;
- a buffer limit indicator to indicate to a memory controller when at least one of the first write retry buffer and the second write retry buffer meets a threshold criteria associated with a number of the plurality of entries that are unused;
- and
- a retry controller that, in response to a first command received by the memory device from the memory controller based on the buffer limit indicator, retries the first failed write operation and the second failed write operation.
A method of operating a memory device, comprising:
- operating a first memory core in a manner that a first portion of attempted writes to the first memory core are determined by corresponding verify operations to have failed in a manner that causes loss of data;
- operating a second memory core in a manner that a second portion of attempted writes to the second memory core are determined by corresponding verify operations to have failed in the manner that causes loss of data;
- storing, in a first write retry buffer, first failed write information that includes a first entry associated with a first failed write operation that is a first one of the first portion of attempted writes to the first memory core that has been determined to have failed;
- storing, in a second write retry buffer, second failed write information that includes a second entry associated with a second failed write operation that is a first one of the second portion of attempted writes to the second memory core that has been determined to have failed;
- indicating, based on a threshold criteria met by a number of entries used in the first write retry buffer and the second write retry buffer, a condition to a memory controller;
- and,
- retrying, in response to a first command that, based on the condition, is received from the memory controller, the first failed write operation and the second failed write operation.
A method, comprising:
- operating a first memory core in a first memory device in a manner that a portion of attempted writes to the first memory core are determined by corresponding verify operations to have failed in a manner that causes loss of data;
- operating a second memory core in the first memory device in a manner that a portion of attempted writes to the second memory core are determined by corresponding verify operations to have failed in the manner that causes loss of data;
- determining, by the first memory device and based on a first verify operation performed by the first memory device, that a first attempted write to the first memory core in the first memory device failed;
- based on determining that the first attempted write failed, storing, by the first memory device and in a first internal write retry buffer in the first memory device, a first entry;
- determining, by the first memory device and based on a second verify operation performed by the first memory device, that a second attempted write to the second memory core in the first memory device failed;
- based on determining that the second attempted write failed, storing, by the second memory device and in a second internal write retry buffer in the first memory device, a second entry;
- outputting, by the first memory device, a first indicator that indicates to a memory controller whether at least one of the first internal write retry buffer and the second internal write retry buffer meet a threshold criteria associated with a first amount of unused entries;
- and
- in response to a command from the memory controller, retrying, by the first memory device and using information from the first entry in the first internal write retry buffer in the first memory device, the first attempted write, and retrying, by the first memory device using information from the second entry in the second internal write retry buffer in the first memory device, the second attempted write.
Dependencies are inferred from the claim text.