Separation of control and data plane functions in SoC virtualized I/O device

A method, comprising:

• executing, by a first processor, a first set of commands included in a first execution thread of a plurality of execution threads to generate a first request formatted according to an on-chip communication protocol, wherein the first request is associated with a first plurality of functions performed by a first device, wherein the first plurality of functions includes at least a first function to train a serial data link, and wherein a given one of execution threads includes instructions that are executable independently of instructions included in other ones of the plurality of execution threads;
• executing, by the first processor, a second set of commands included in a second execution thread of the plurality of execution threads to generate a second request formatted according to the on-chip communication protocol, wherein the second request is associated with a second plurality of functions performed by the first device, wherein the second plurality of functions includes a particular function to encapsulate data into packets for transmission via the serial data link, and wherein the second plurality of functions includes at least a second function to modify data received by the first device;
• translating, by a communication circuit, the first and second request from the on-chip communication protocol to a first communication protocol compatible with the first device, wherein the communication circuit is coupled to the first processor on a common integrated circuit chip, and wherein the communication circuit is coupled to the first device via a first link;
• transmitting, by the communication circuit, the first request to the first device via the first link using the first communication protocol;
• and
• transmitting, by the communication circuit, the second request to the first device via the first link using the first communication protocol.

A system, comprising:

• a plurality of processors, wherein a first processor of the plurality of processors is configured to:
• execute, in a first execution thread of a plurality of execution threads, a first instruction included in a first subset of a first plurality of software instructions to generate a first request, wherein the first plurality of software instructions is included in a first guest operating system;
• and
• execute, in a second execution thread of the plurality of execution threads, a second instruction included in a second subset of the first plurality of software instructions to generate a second request;
• wherein a second processor of the plurality of processors is configured to execute a second plurality of software instructions included in a second guest operating system to generate a third request;
• a first device including a plurality of registers, wherein the first device is configured to perform a plurality of command functions and a plurality of virtual functions, wherein the plurality of command functions includes a command to train a serial data link, and wherein the first device includes at least one hardware resource configured to be shared by two or more virtual functions of the plurality of virtual functions, and wherein the plurality of virtual functions include a particular function to encapsulate data into packets for transmission via the serial data link;
• a transmission unit coupled to the first device, wherein the transmission unit is configured to send and receive data packets via the serial data link;
• a second device;
• and
• a communication unit coupled to the plurality of processor on a common integrated circuit chip, wherein the communication unit includes a first port and a second port, wherein the first port is coupled to the first device via a first link and the second port is coupled to the second device via a second link;
• wherein the first port is configured to:
• receive the first request from the first processor, wherein the first request is formatted according to an on-chip communication protocol, and wherein the first request is associated with at least one command function of the plurality of command functions;
• translate the first request from the on-chip communication protocol to a first communication protocol compatible with the first device;
• relay the first request to the first device via the first link using the first communication protocol;
• receive the second request from the first processor, wherein the second request is formatted according to the on-chip communication protocol, and wherein the second request is associated with at least on virtual functions of the plurality of virtual functions;
• and
• translate the second request from the on-chip communication protocol to the first communication protocol.