# US 2006/0156194 A1 · Method for reallocation of a memory of a subsystem, and subsystem

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2006/0156194 A1
- Document ID: US20060156194A1
- Status: Published application
- Kind: A1
- Application number: 11/295,719
- Filed: 2005-12-07
- Published: 2006-07-13
- Earliest claimed: 2004-12-09
- Assignees: Infineon Technologies AG
- Inventors: Walter Mischo
- Web page: https://patentgrep.com/patent/US20060156194A1

## CPC

- G06F 11/0793: {Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40)}
- G06F 11/073: {in a memory management context, e.g. virtual memory or cache management (memory management G06F12/00; testing of static memory units G11C29/00)}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

The method for reallocation of a memory, in particular a command memory, wherein the memory is part of a subsystem and the memory is assigned to a processing unit available on the subsystem, and wherein the occurrence of a memory error in the memory (RAM) is detected, whereupon the content of the memory is regenerated, is characterized in that the content of the memory is regenerated by means of an error routine controlled by the processing unit.

## Claims

### Claim 1

A method for reallocation of a command memory, the memory part of a subsystem, comprising:
assigning the memory to a processing unit available on the subsystem; and
detecting an occurrence of a memory error in the memory, whereupon content of the memory is regenerated by an error routine controlled by the processing unit.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the entire content of the memory is regenerated.

### Claim 3 (depends on 1)

The method according to claim 1, wherein an erroneous part of the content of the memory is regenerated.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the regeneration of the content of the memory is followed by resuming a program at the location where the memory error has occurred.

### Claim 5 (depends on 1)

The method according to claim 1, wherein regeneration of the content of the memory is followed by rebooting a program of the processing unit.

### Claim 6 (depends on 1)

The method according to claim 1, wherein execution of the processor-controlled error routine comprises shutting down the subsystem.

### Claim 7 (depends on 6)

The method according to claim 6, wherein subsequent to the shutting down the subsystem is rebooted.

### Claim 8 (depends on 1)

The method according to claim 1, wherein
the subsystem comprises a further processing unit,
the memory is not assigned to the further processing unit, and
the memory can be accessed by the further processing unit, wherein the content of the memory is regenerated by an error routine controlled by the further processing unit.

### Claim 9

A subsystem, comprising:
a processing unit;
a memory assigned to the processing unit;
an error detection unit; and
a switch connected with the processing unit at an output, and which is adapted to switch data words required for executing an error routine by the processing unit from at least one further memory to the processing unit.

### Claim 10 (depends on 9)

The subsystem according to claim 9, wherein the further memory is integrated on the subsystem.

### Claim 11 (depends on 10)

The subsystem according to claim 10, wherein the further memory comprises a volatile memory device.

### Claim 12 (depends on 9)

The subsystem according to claim 9, further comprising at least one further processing unit, wherein
the memory is not assigned to the further processing unit,
the memory can be accessed by the further processing unit, wherein the content of the memory is regenerated by an error routine controlled by the further processing unit, and
the switch is connected with the further processing unit at an output and is adapted to switch the data words required for executing the error routine by the further processing unit from at least one further memory to the further processing unit.
