# US 2015/0106657 A1 · Method of providing high integrity processing

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

- Number: US 2015/0106657 A1
- Document ID: US20150106657A1
- Status: Published application
- Kind: A1
- Application number: 14/549,882
- Filed: 2014-11-21
- Published: 2015-04-16
- Earliest claimed: 2011-11-10
- Assignees: GE Aviation Systems LLC
- Inventors: Jonathan Paul Van Stensel
- Web page: https://patentgrep.com/patent/US20150106657A1

## CPC

- G06F 11/273: Tester hardware, i.e. output processing circuits {(G06F11/263 takes precedence)}
- G06F 11/1633: {using mutual exchange of the output between the redundant processing components}
- G06F 11/1675: {Temporal synchronisation or re-synchronisation of redundant processing components}
- G06F 11/2242: {in multi-processor systems, e.g. one processor becoming the primary tester (G06F11/2736 takes precedence)}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method of providing synchronization and integrity checking in a high integrity processing system having at least two redundant processing lanes, with each lane having an application processor, with the application processors running the same application software in a non-lockstep configuration, and outputting transactions requiring access to an addressable space.

## Claims

### Claim 1

A method of providing synchronization and integrity checking in a high integrity processing system having at least two redundant processing lanes, with each lane having an application processor (AP), with the APs running the same application software in a non-lockstep configuration, and outputting transactions requiring access to an addressable space, the method comprising:
outputting a first transaction from a first AP in a first processing lane and running application software in a non-lockstep configuration;
storing the first transaction in a second buffer of a second processing lane and associated with a second AP;
effecting a processing of the first transaction in the first processing lane;
storing the effected processing of the first transaction in the second buffer;
outputting a second transaction from the second AP running the same application software in a non-lockstep configuration;
storing the second transaction in the second buffer;
comparing the first and second transactions to determine if they match; and
returning the effected processing of the first transaction to the second AP from the second buffer when the first and second transactions match.

### Claim 2 (depends on 1)

The method of claim 1, further comprising logging an error in memory when the first and second transactions do not match.

### Claim 3 (depends on 2)

The method of claim 2, further comprising removing at least one of the first transaction, the second transaction, or the effected processing of the first transaction when the first and second transactions do not match.

### Claim 4 (depends on 1)

The method of claim 1 further comprising returning the effected processing of the first transaction to the first AP prior to returning the effected processing of the first transaction to the second AP.

### Claim 5 (depends on 1)

The method of claim 1 further comprising returning the effected processing of the first transaction to the first AP prior to the comparing.

### Claim 6 (depends on 1)

The method of claim 1 wherein outputting a first transaction further comprises outputting a first read transaction of an addressable space.

### Claim 7 (depends on 6)

The method of claim 6 further comprising determining if there is a write transactions pending for an overlapping addressable space of the first transaction prior to the effecting the processing of the first transaction in the first processing lane.

### Claim 8 (depends on 7)

The method of claim 7 wherein when there is a write transaction pending for and overlapping addressable space, the first transaction is stored in a first buffer in the first processing lane and associated with the first AP, behind the write transaction.

### Claim 9 (depends on 1)

The method of claim 1 wherein the comparison comprises popping the first and second transactions off of the second buffer and then comparing the popped off first and second transactions.

### Claim 10 (depends on 1)

The method of claim 1, further comprising flagging an error when the first and second transactions do not match.

### Claim 11 (depends on 1)

The method of claim 1, further comprising effecting a processing of the first transaction in the second processing lane, comparing the results of the processing of the first transaction in the first lane and the processing of the first transaction in the second lane.

### Claim 12 (depends on 11)

The method of claim 11 further comprising the storing the effected processing of the first transaction in the second buffer when the effected processing of the first transaction in the first lane and the effected processing of the first transaction in the second lane match.

### Claim 13 (depends on 11)

The method of claim 11, further comprising logging an error in memory when the effected processing of the first transaction in the first lane and the effected processing of the first transaction in the second lane do not match.

### Claim 14 (depends on 11)

The method of claim 11, further comprising removing at least one of the first transaction, the second transaction, or the effected processing of the first transaction when the first transaction results from the first and second processing lanes do not match.
