Method for managing requests for access to random access memory and corresponding system

Patent metadata

Field	Value
Publication number	US-11495275-B2
Application number	US-202117336841-A
Country	US
Kind code	B2
Filing date	Jun 2, 2021
Priority date	Jun 12, 2020
Publication date	Nov 8, 2022
Grant date	Nov 8, 2022

Abstract

A random access memory is connected to a processing unit through a memory interface. Access to the random access is memory is controlled by a process. The memory interface receives a request for access to the memory issued by the processing unit. In response to the request, the memory interface indicates to the processing unit that the memory is not available to receive another access request during a duration of unavailability. This duration can be differentiated depending on whether the received request is a write or read request. The value of the duration of unavailability associated with a write request and the value of the duration of unavailability associated with a read request are individually programmable independently of each other.

First claim

The invention claimed is:

1. A method for managing access to a random access memory connected to a processing unit through a memory interface, comprising:
in response to receiving a request issued by the processing unit at the memory interface for access to the memory, sending an indication by the memory interface to the processing unit that the memory is not available to receive another access request during a duration of unavailability;
wherein said duration of unavailability is differentiated depending on whether the received request is a write or read request; and
wherein a value of the duration of unavailability associated with a write request and a value of the duration of unavailability associated with a read request are individually programmable independently of each other.

2. The method according to claim 1 , further comprising:
timing the processing unit by a clock signal;
wherein the value of the duration of unavailability associated with the write request and the value of the duration of unavailability associated with the read request are each dependent on a value of a power supply voltage for the random access memory and a frequency of the clock signal.

3. The method according to claim 2 , wherein the value of the duration of unavailability associated with the read request is non-zero if the power supply voltage is less than a first threshold and if the frequency is greater than a second threshold.

4. The method according to claim 2 , wherein the value of the duration of unavailability associated with the write request is non-zero if the power supply voltage is less than a third threshold and if the frequency is greater than a fourth threshold.

5. The method according to claim 1 , wherein the value duration of unavailability associated with the write request is less than or equal to the value of duration of unavailability associated with the read request.

6. The method according to claim 1 , further comprising:
timing the processing unit by a clock signal;
wherein the value of the duration of unavailability associated with the write request corresponds to a first number of cycles of the clock signal; and
wherein the value of the duration of unavailability associated with the read request corresponds to a second number of cycles of the clock signal.

7. A system, comprising:
a random access memory;
a processing unit connected to the random access memory through a memory interface;
wherein said memory interface includes:
a reception circuit configured to receive a request for access to the memory emitted by the processing unit; and
a processing circuit connected to the reception circuit and configured to indicate to the processing unit that the random access memory is not available to receive another access request during a duration of unavailability, wherein said duration of unavailability is differentiated depending on whether the received request is a write or read request; and

a programming circuit configured to individually program, independently of each other, a value of the duration of unavailability associated with the write request and a value of the duration of unavailability associated with a read request.

8. The system according to claim 7 , wherein the processing unit is timed by a clock signal, and wherein the value of the duration of unavailability associated with the write request and the value of the duration of unavailability associated with the read request each depend on the power supply voltage value of the random access memory and the frequency of the clock signal.

9. The system according to claim 8 , wherein the value of the duration of unavailability associated with the read request is non-zero if the power supply voltage is less than a first threshold and if the frequency is greater than a second threshold.

10. The system according to claim 8 , wherein the value of the duration of unavailability associated with the write request is non-zero if the power supply voltage is less than a third threshold and if the frequency is greater than a fourth threshold.

11. The system according to claim 7 , wherein the value of the duration of unavailability associated with the write request is less than or equal to the value of the duration of unavailability associated with the read request.

12. The system according to claim 7 , wherein the processing unit is timed by a clock signal, and
wherein the value of the duration of unavailability associated with the write request corresponds to a first number of cycles of the clock signal; and
wherein the value of the duration of unavailability associated with the read request corresponds to a second number of cycles of the clock signal.

13. The system according to claim 12 , wherein the processing circuit comprises:
a first counter timed by the clock signal and having a first end-of-count value corresponding to the value of the duration of unavailability associated with the write request;
a second counter timed by the clock signal and having a second end-of-count value corresponding to the value of the duration of unavailability associated with the read request; and
a state machine configured to: trigger the first counter if the first end-of-count value is non-zero or trigger the second counter if the second end-of-count value is non-zero, output a memory unavailability signal to the processing unit when one of the first and second counters is triggered and output a memory availability signal to the processing unit when said one of the first and second counters that has been triggered has reached an end-of-count value.

14. The system according to claim 13 , wherein the programming circuit comprises:
a first programmable memory circuit connected to the processing circuit and configured to store a first indication representative of a value of the duration of unavailability associated with the write request; and
a second programmable memory circuit connected to the processing circuit and configured to store a second indication representative of a value of the duration of unavailability associated with the read request.

15. The system according to claim 14 , wherein said first indication is the first end-of-count value and said second indication is the second end-of-count value.

16. The system according to claim 7 , forming a microcontroller.

Assignees

• St Microelectronics Rousset

Inventors

• Eva Christophe
• Gril-Maffre Jean-Michel

Classifications

• Y02D10/00

  Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title

• G06F3/0614

  Improving the reliability of storage systems · CPC title

• G06F3/0622

  in relation to access · CPC title

• G11C16/3404

  Convergence or correction of memory cell threshold voltages; Repair or recovery of overerased or overprogrammed cells · CPC title

• G11C7/222 Primary

  Clock generating, synchronizing or distributing circuits within memory device · CPC title