# US 2009/0222252 A1 · Data driven logic simulation chip and toolchain

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

- Number: US 2009/0222252 A1
- Document ID: US20090222252A1
- Status: Published application
- Kind: A1
- Application number: 12/392,666
- Filed: 2009-02-25
- Published: 2009-09-03
- Earliest claimed: 2008-02-29
- Assignees: ET International, Inc.
- Inventors: Fei Chen; Guang R. Gao
- Web page: https://patentgrep.com/patent/US20090222252A1

## CPC

- G06F 30/33: Design verification, e.g. functional simulation or model checking
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An apparatus and method may be used for compiling a hardware logic design into data-driven logic programs to be executed on a data-driven chip. The apparatus may include storage with a library for defining a net-list synthesized by a synthesis tool. The apparatus may also include a data-driven logic verification chip comprising a plurality of logic processors. The apparatus may further include a code generator for adopting heuristics to convert the net-list into data driven logic programs and for allocating hardware resources to balance computing and storage loads across the plurality of logic processors of the verification chip.

## Claims

### Claim 1

An apparatus, comprising:
storage including a library to define how a net-list is to be synthesized by a synthesis tool;
a data-driven logic verification chip comprising a plurality of logic processors; and
a code generator to adopt one or more heuristics to convert the net-list into data driven logic programs and to allocate hardware resources to balance computing and storage loads across the plurality of logic processors of the verification chip.

### Claim 2 (depends on 1)

The apparatus of claim 1, wherein the net-list comprises a set of primitives selected from a group consisting of primitive gates, latches and memory units; wherein the library is defined such that primitive gates can be implemented by logic instructions and executed by the plurality of logic processors of the verification chip.

### Claim 3 (depends on 1)

The apparatus of claim 1, wherein the library includes a set of primitive logic operations upon which the net-list can be built by the synthesis tool, wherein primitive logic cells are mapped to logic instructions and executed by the plurality of logic processors of the verification chip.

### Claim 4 (depends on 3)

The apparatus of claim 3, wherein a primitive logic cell is configured to accept one or more inputs, wherein each of the inputs and an output of the primitive logic cell is a one bit signal.

### Claim 5 (depends on 3)

The apparatus of claim 3, wherein the net-list comprises a plurality of primitive logic cells, wherein a predefined number of logic cells are grouped into a logic cell bundle and each logic cell bundle is treated as an atomic execution unit.

### Claim 6 (depends on 1)

The apparatus of claim 1, wherein the library further comprises variable array primitives.

### Claim 7 (depends on 1)

The apparatus of claim 1, wherein the verification chip comprises a plurality of logic simulation engines, each of the plurality of logic simulation engines comprising:
a routing unit;
an instruction memory;
a logic processor; and
a data memory.

### Claim 8 (depends on 7)

The apparatus of claim 7, wherein the routing unit is configured to execute a routing algorithm wherein a header of a routed packet includes offsets in an X direction and a Y direction and the packet is determined to reach its destination when both offsets are equal to zero.

### Claim 9 (depends on 7)

The apparatus of claim 7, wherein the logic processor is configured to execute all logic instructions in an active logic cell bundle, examine an output signal of the active logic cell bundle for changes and generate a triggering signal if the output signal is determined to have changed.

### Claim 10 (depends on 9)

The apparatus of claim 9, wherein the triggering signal is transmitted locally in the logic simulation engine and remotely to other logic simulation engines, and
wherein the logic processor is configured to be interrupted when an input signal is changed by a new triggering message and the logic processor is configured to restart execution at a first logic instruction of the logic cell bundle.

### Claim 11 (depends on 8)

The apparatus of claim 8, wherein the logic processor is configured for pipelined execution.

### Claim 12 (depends on 7)

The apparatus of claim 7, wherein the data memory comprises:
a bundle number to indicate a number of logic cell bundles assigned to one logic simulation engine;
a plurality of bundle information records, each of the plurality of bundle information records to store a starting address and a length of logic instructions in one logic cell bundle and to store the current state of the logic cell bundle; and
a plurality of bundle data records, each of the plurality of bundle data records including inputs, outputs, double buffered latches and temporary data,
wherein a triggering message changes the inputs of a bundle data record and causes a state bit of an associated logic cell bundle to be set to active.

### Claim 13 (depends on 12)

The apparatus of claim 12, wherein memory bits in the temporary data section are configured to be reused for combinatorial signals in the logic cell bundle by a heuristic register allocation algorithm when logic instructions are generated by the code generator.

### Claim 14 (depends on 7)

The apparatus of claim 7, wherein the plurality of logic simulation engines are arranged in a mesh network.

### Claim 15

An apparatus, comprising:
a plurality of logic simulation engines;
an IO glue unit and a simulation control unit to load logic instructions into the plurality of logic simulation engines;
an array simulation unit to handle array simulation requests initiated from the plurality of logic simulation engines and to issue input/output instructions to off-chip storage units including simulated arrays; and
an inter-chip communication unit to handle communication requests between logic verification chips.

### Claim 16 (depends on 15)

The apparatus of claim 15, wherein the simulation control unit is configured to handle control functions.

### Claim 17 (depends on 15)

The apparatus of claim 15, wherein there is a direct physical connection with another inter-chip communication unit of an adjacent verification chip.

### Claim 18 (depends on 15)

The apparatus of claim 15, wherein each of the plurality of logic simulation engines comprises:
a routing unit;
an instruction memory;
a logic; and
a data memory.

### Claim 19 (depends on 15)

The apparatus of claim 15, further comprising a memory controller to permit interaction with an external storage.

### Claim 20 (depends on 15)

The apparatus of claim 15, wherein said apparatus is embodied as a single integrated circuit.

### Claim 21

A method, comprising:
defining a set of primitive logic operations upon which a net-list can be built;
synthesizing logic design sources into the net-list;
converting the net-list into data driven logic instructions;
loading the logic instructions into logic simulation engines in a verification chip;
routing data packets to a specific logic simulation engine; and
executing the logic instructions in the logic simulation engine on the routed data packets.

### Claim 22 (depends on 21)

The method of claim 21, wherein the routing comprises:
moving data into and out of the logic simulation engine and relaying traffic passing by the logic simulation engine; and
routing data packets wherein a header of a routed packet includes offsets in an X direction and a Y direction and the packet is determined to reach its destination when both offsets are equal to zero.

### Claim 23 (depends on 21)

The method of claim 21, wherein the defining comprises defining logic instructions associated with logic cell bundles, each logic cell bundle being a group of primitive cells treated as an atomic execution unit.

### Claim 24 (depends on 21)

The method of claim 21, wherein the executing comprises executing locally and remotely generated triggering events.

### Claim 25 (depends on 21)

The method of claim 21, wherein the executing comprises:
executing all logic instructions in an active logic cell bundle;
examining an output signal of the active logic cell bundle for changes; and
generating a triggering signal if the output signal is determined to have changed,
wherein the triggering signal is transmitted locally in the logic simulation engine and remotely to other logic simulation engines, and
wherein a local processor is interrupted when an input signal is changed by a new triggering message and the logic processor restarts execution at a first logic instruction of the logic cell bundle.

### Claim 26 (depends on 21)

The method of claim 21, wherein the executing comprises executing one logic instruction every cycle unless there is an external triggering message coming into the logic simulation engine and the external triggering message is competing for a write port of a data memory.

### Claim 27

A computer program embodied on a computer readable medium, the computer program comprising program code for controlling a processor to execute a method comprising:
defining a set of primitive logic operations upon which a net-list can be built;
synthesizing logic design sources into the net-list;
converting the net-list into data driven logic instructions; and
loading the logic instructions into logic simulation engines in a verification chip for execution of data packets routed to specific ones of the logic simulation engines.
