Chiplet Architecture Data Processing Devices And Methods

A method of operating a data processing device including a substrate and a plurality of chiplets mounted to the substrate and interconnected to one another, the method comprising:

• establishing a chiplet range of global addresses for each chiplet, the chiplet range for each one of the plurality of chiplets being mutually exclusive with the chiplet range for any other one of the plurality of chiplets;
• directing software commands to working elements in each chiplet, at least one of the working elements being responsive to software commands to process data and generate messages, each message including a payload and a global address within a global range of global addresses encompassing the chiplet ranges of all of the chiplets, the software commands used in all of the chiplets being based on a common global address table;
• when the global address of a message from a working element of a chiplet is within the chiplet range for that chiplet, actuating a translation circuit within the chiplet to translate the global address to a local address within that chiplet and routing the message to a working element of that chiplet based on the local address;
• and
• when the global address of the message of a message from a working element of a chiplet is outside the chiplet range for that chiplet, dispatching the message to one or more other ones of the plurality of chiplets.

A non-transitory computer readable medium for storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations for a data processing device that includes a substrate and a plurality of chiplets mounted to the substrate and interconnected to one another, the operations comprising:

• establishing a chiplet range of global addressers for each chiplet, the chiplet range for each one of the plurality of chiplets being mutually exclusive with the chiplet range for any other one of the plurality of chiplets;
• directing software commands to working elements in each chiplet, at least one of the working elements being responsive to software commands to process data and generate messages, each message including a payload and a global address within a global range of global addresses encompassing the chiplet ranges of all of the chiplets, the software commands used in all of the chiplets being based on a common global address table;
• when the global address of a message from a working element of a chiplet is within the chiplet range for that chiplet, actuating a translation circuit within the chiplet to translate the global address to a local address within that chiplet and routing the message to a working element of that chiplet based on the local address;
• and
• when the global address of the message of a message from a working element of a chiplet is outside the chiplet range for that chiplet, dispatching the message to one or more other ones of the plurality of chiplets.