Interrupt-Responsive Non-Volatile Memory Bank Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-volatile memory systems, such as flash memory, face delays and potential loss of interrupts during programming or erase operations, as these operations lock out the entire memory, leading to delayed response times and inefficiencies, especially in real-time embedded applications.
Innovation Solution
A memory system that allows for interrupt-responsive operation using a single non-volatile memory by aborting the current memory routine upon an interrupt, setting a command failed flag, executing an interrupt service routine, and recovering the aborted routine upon completion, thereby enabling immediate interrupt servicing without requiring a second memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a single non-volatile memory is used, then cost and die area are reduced, but interrupt response time deteriorates due to memory lockout during program/erase operations
Solution Approach 1:
The memory space is segmented into multiple banks (first bank and second bank), allowing independent access to different memory regions. When one bank is locked for program/erase operations, the other bank remains accessible for interrupt service routines, thereby maintaining interrupt response time while using a single memory device.
Solution Approach 2:
A memory management unit acts as an intermediary between the processor and the memory banks, directing interrupt service routines to the accessible bank while allowing program/erase operations to proceed in the locked bank. This mediator enables simultaneous operations without conflict.
2Loss of time
If a second non-volatile memory is provided for interrupt service routines, then interrupt response time is improved, but die area and cost increase
Solution Approach 1:
The single non-volatile memory device performs multiple functions through bank switching: it serves both as the main program/erase memory and as the interrupt service routine memory. The memory management unit dynamically assigns banks based on operational needs, making the single memory universally functional for both purposes.
Solution Approach 2:
The functionality of what would traditionally require two separate memory devices is merged into a single memory device with multiple banks. The first bank and second bank work together to provide both main memory and interrupt memory functionality, reducing die area and cost while maintaining performance.
3Loss of time
If SRAM is used for interrupt service routines, then interrupt response time is improved, but power consumption and cell size increase significantly
Solution Approach 1:
The system changes the operational parameters of the non-volatile memory by switching between different banks and operational modes. Instead of using SRAM with its high power consumption, the system uses the same non-volatile memory technology but operates it in different configurations (reading from one bank while allowing program/erase in another), maintaining low power consumption while achieving fast interrupt response.
Data Source
AI summary
An interrupt-responsive non-volatile memory respond to an interrupt by aborting execution by a memory controller of a memory routine in a non-volatile memory, sets, a flag and executes an interrupt service routine; and upon completion of the interrupt service routine, in response to the flag, recovers the execution of the aborted memory routine.