# US 2023/0267000 A1 · Processing apparatus and system for executing data processing on a plurality of pieces of channel information

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

- Number: US 2023/0267000 A1
- Document ID: US20230267000A1
- Status: Published application
- Kind: A1
- Application number: 18/089,146
- Filed: 2022-12-27
- Published: 2023-08-24
- Earliest claimed: 2022-02-23
- Assignees: Beijing ESWIN Computing Technology Co., Ltd.
- Applicants: Beijing ESWIN Computing Technology Co., Ltd.
- Inventors: Shilin LUO
- Web page: https://patentgrep.com/patent/US20230267000A1

## CPC

- G06F 9/383: {Operand prefetching (cache prefetching G06F12/0862)}
- G06F 9/4881: {Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues}
- G06F 9/3016: {Decoding the operand specifier, e.g. specifier format}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

A processing apparatus and a system for executing data processing on a plurality of pieces of channel information are disclosed. The processing apparatus includes: a channel information acquiring circuit, configured to acquire target channel information among the plurality of pieces of channel information; a storing circuit, including a plurality of storage regions corresponding to the plurality of pieces of channel information; a reading circuit, configured to sequentially pre-store the data information from the plurality of storage regions of the storing circuit to the pre-storing circuit; and a pre-storing circuit, configured to pre-store the data information from the plurality of storage regions of the storing circuit to wait for use in the data processing, and output pre-stored target data information corresponding to the target channel information, according to the acquired target channel information, for use in the data processing.

## Claims

### Claim 1

A processing apparatus for executing data processing on a plurality of pieces of channel information, comprising a channel information acquiring circuit, a storing circuit, a reading circuit, and a pre-storing circuit, wherein
the channel information acquiring circuit is configured to acquire target channel information among the plurality of pieces of channel information;
the storing circuit comprises a plurality of storage regions corresponding to the plurality of pieces of channel information, and each of the plurality of storage regions is configured to store data information for any one piece of channel information among the plurality of pieces of channel information;
the reading circuit is configured to sequentially pre-store the data information from the plurality of storage regions of the storing circuit to the pre-storing circuit; and
the pre-storing circuit is configured to pre-store the data information from the plurality of storage regions of the storing circuit to wait for use in the data processing, and output pre-stored target data information corresponding to the target channel information, according to the acquired target channel information, for use in the data processing.

### Claim 2 (depends on 1)

The processing apparatus according to claim 1, wherein the pre-storing circuit comprises:
a plurality of storing sub-circuits, wherein each of the plurality of storing sub-circuits is configured to store the data information corresponding to any one piece of the plurality of pieces of channel information or at least one piece of selected channel information, to wait for use in the data processing; and
a multiplexer, configured to select a storing sub-circuit corresponding to the target channel information from the plurality of storing sub-circuits, according to the target channel information, to wait for use in the data processing.

### Claim 3 (depends on 2)

The processing apparatus according to claim 2, wherein the reading circuit comprises:
a first Round Robin scheduling circuit, configured to perform Round Robin and determine one piece of channel information among the plurality of pieces of channel information to execute the pre-storing; and
a data reading control circuit, configured to read the data information in the storage region corresponding to the one piece of the channel information into the storing sub-circuit, according to the one piece of channel information, to wait for use in the data processing.

### Claim 4 (depends on 3)

The processing apparatus according to claim 3, wherein the first Round Robin scheduling circuit is further configured to,
in response to completion of a pre-storing operation on a current channel information and the data information in the storing circuit being non-empty, execute a next time of Round Robin to determine a next piece of channel information to execute the pre-storing operation.

### Claim 5 (depends on 3)

The processing apparatus according to claim 3, wherein the channel information acquiring circuit comprises:
a second Round Robin scheduling circuit, configured to perform Round Robin and determine the target channel information among the plurality of pieces of channel information to execute the data processing.

### Claim 6 (depends on 5)

The processing apparatus according to claim 5, wherein the second Round Robin scheduling circuit is further configured to,
in response to completion of the data processing on a current channel information, execute a next time of Round Robin to determine a next piece of channel information to execute the data processing.

### Claim 7 (depends on 2)

The processing apparatus according to claim 2, wherein the storing sub-circuit comprises a first-in-first-out queue based on combined logic, and
the first-in-first-out queue is configured to enable a request signal for the first-in-first-out queue and data information associated with the request signal to occur in a same clock cycle.

### Claim 8 (depends on 1)

The processing apparatus according to claim 1, wherein the processing apparatus is a direct memory access apparatus, and the data processing comprises a data moving operation.

### Claim 9 (depends on 8)

The processing apparatus according to claim 8, wherein the data moving operation comprises moving of the data information or moving of target data indicated by the data information, the target data is data required by a neural-network processor for executing a neural network operation, and the neural-network processor comprises at least one processing unit array and a plurality of global buffer clusters; and
the data information indicates address information of the target data, the address information is obtained based on a mapping relationship between the target data and at least one processing unit in the processing unit array, and each of the global buffer clusters is configured to receive the moved data information or the moved target data on one piece of channel information among the plurality of pieces of channel information.

### Claim 10 (depends on 8)

The processing apparatus according to claim 8, further comprising:
an interface control circuit, configured to acquire the target data information from the pre-storing circuit, generate a data moving request according to the target data information, and execute the data moving operation according to the data moving request.

### Claim 11 (depends on 1)

A computer system, comprising:
the processing apparatus according to claim 1; and
a neural-network processor, wherein the neural-network processor is electrically coupled to the processing apparatus, and the neural-network processor comprises at least one processing unit array and a plurality of global buffer clusters; and
the data information indicates address information of target data, the address information is obtained based on a mapping relationship between the target data and at least one processing unit in the processing unit array, and each of the global buffer clusters is configured to receive the moved data information or the moved target data indicated by the data information on one piece of channel information among the plurality of pieces of channel information.

### Claim 12 (depends on 11)

The computer system according to claim 11, further comprising:
an interface decoder, configured to receive an initial instruction and decode the initial instruction to obtain the data information, and supply the data information decoded to the processing apparatus.

### Claim 13 (depends on 12)

The computer system according to claim 12, further comprising:
an auxiliary processor, configured to supply the initial instruction to the interface decoder.

### Claim 14 (depends on 13)

The computer system according to claim 13, wherein the auxiliary processor is coupled to the interface decoder through a second data transmission protocol,
the neural-network processor is coupled to the processing apparatus through a first data transmission protocol, and
the first data transmission protocol is different from the second data transmission protocol.

### Claim 15

A scheduling method for executing data processing on a plurality of pieces of channel information, comprising:
pre-storing data information for any one piece of channel information among the plurality of pieces of channel information from a plurality of storage regions corresponding to the plurality of piece of channel information, to wait for use in the data processing;
acquiring target channel information among the plurality of pieces of channel information; and
outputting the target data information pre-stored corresponding to the target channel information, according to the target channel information acquired, to wait for use in the data processing.

### Claim 16 (depends on 15)

The scheduling method according to claim 15, wherein pre-storing data information for any one piece of channel information among the plurality of pieces of channel information from the plurality of storage regions corresponding to the plurality of piece of channel information, to wait for use in the data processing, comprises:
performing Round Robin and determining one piece of channel information among the plurality of pieces of channel information to execute the pre-storing; and
reading data information in a storage region corresponding to the one piece of the channel information into a storing sub-circuit, according to the one piece of channel information, to wait for use in the data processing.

### Claim 17 (depends on 15)

The scheduling method according to claim 15, wherein acquiring the target channel information among the plurality of pieces of channel information comprises:
performing Round Robin and determining the target channel information among the plurality of pieces of channel information to execute the data processing; and
outputting the target data information pre-stored corresponding to the target channel information, according to the target channel information acquired, to wait for use in the data processing, comprises:
selecting a storing sub-circuit corresponding to the target channel information, according to the target channel information, to wait for use in the data processing.

### Claim 18 (depends on 15)

The scheduling method according to claim 15, wherein the scheduling method is used in a direct memory access apparatus, and the data processing comprises a data moving operation.

### Claim 19 (depends on 18)

The scheduling method according to claim 18, wherein the data moving operation comprises moving of the data information or moving of target data indicated by the data information, the target data is data required by a neural-network processor for executing a neural network operation, and the neural-network processor comprises at least one processing unit array and a plurality of global buffer clusters; and
the data information indicates address information of the target data, the address information is obtained based on a mapping relationship between the target data and at least one processing unit in the processing unit array, and each of the global buffer clusters is configured to receive the moved data information or the moved target data on one piece of channel information among the plurality of pieces of channel information.

### Claim 20 (depends on 16)

The scheduling method according to claim 16, wherein performing Round Robin and determining one piece of channel information among the plurality of pieces of channel information to execute the pre-storing comprises:
executing a next time of Round Robin to determine a next piece of channel information to execute a pre-storing operation, in response to completion of the pre-storage operation on a current channel information and data information in a storing circuit being non-empty.
