High-speed broadside communications and control system
A real-time computational device includes a programmable real-time processor, a communications input port which is connected to the programmable real-time processor through a first broadside interface, and a communications output port which is connected to the programmable real-time processor through a second broadside interface. Both broadside interfaces enable 1024 bits of data to be transferred across each of the broadside interfaces in a single clock cycle of the programmable real-time processor.
1. A computational device, comprising:
a programmable real-time processor that has an instruction set with a first bit width;
a network communications input port connected to the programmable real-time processor through a first interface that has a second bit width that is greater than the first bit width; and
a network communications output port connected to the programmable real-time processor through a second interface that has the second bit width.
2. The computational device of claim 1 , further comprising:
a non-pipelined real-time processor connected to the programmable real-time processor through a third interface that has the second bit width; and
a direct memory access circuit connected to the programmable real-time processor through fourth interface that has the second bit width.
3. The computational device of claim 2 , wherein the third interface and the fourth interface are different.
4. The computational device of claim 3 , wherein:
the second bit width is at least 1024 bits wide.
5. The computational device of claim 3 , wherein:
the first interface comprises a first 1024-bit data register;
the second interface comprises a second 1024-bit data register;
the third interface comprises a third 1024-bit data register; and
the fourth interface comprises a fourth 1024-bit data register.
6. The computational device of claim 3 , wherein:
the first interface is configured to enable 1024 bits of data to pass between the network communications input port and the programmable real-time processor in a single clock cycle of the programmable real-time processor;
the second interface is configured to enable 1024 bits of data to pass between the network communications output port and the programmable real-time processor in the single clock cycle of the programmable real-time processor;
the third interface is configured to enable 1024 bits of data to pass between the non-pipelined real-time processor and the programmable real-time processor in the single clock cycle of the programmable real-time processor;
the fourth interface is configured to enable 1024 bits of data to pass between the direct memory access circuit and the programmable real-time processor in the single clock cycle of the programmable real-time processor.
7. The computational device of claim 1 , wherein the network communications input port is an Ethernet input port and the network communications output port is an Ethernet output port.
8. The computational device of claim 7 further comprising:
a receive first-in-first-out buffer coupled between the Ethernet input port and the programmable real-time processor such that first interface is directly coupled to the receive first-in-first-out buffer; and
a transmit first-in-first-out buffer coupled between the Ethernet output port and the programmable real-time processor such that second interface is directly coupled to the transmit first-in-first-out buffer.
9. A data transfer method, comprising:
receiving a first data packet at a network communications input port connected to a programmable real-time processor through a first interface, wherein the programmable real-time processor has an instruction set with a first bit width and the first interface has a second bit width that is greater than the first bit width; and
sending the first data packet to a network communications output port connected to the programmable real-time processor through a second interface that has the second bit width.
10. The data transfer method of claim 9 , further comprising:
transferring the first data packet through the first interface, wherein the second bit width is at least thirty-two bytes wide; and
transferring the first data packet through the second interface.
11. The data transfer method of claim 10 , further comprising:
sending a second data packet from the network communications output port concurrently to transferring the first data packet through the second interface.
12. The data transfer method of claim 10 , wherein:
transferring the first data packet through the first interface, comprises transferring the first data packet through the first interface in a single clock cycle of the programmable real-time processor; and
transferring the first data packet through the second interface, comprises transferring the first data packet through the second interface in a single clock cycle of the programmable real-time processor.
13. A system on chip, comprising:
a first processor that has a first bit width;
a network input port connected to the first processor through a first connection that has a second bit width that is greater than the bit width of the first processor; and
a network output port connected to the first processor through a second connection that has the second bit width.
14. The system on chip of claim 13 , further comprising:
a second processor connected to the first processor through a third connection that has the second bit width; and
a direct memory access circuit connected to the first processor through fourth connection that has the second bit width.
15. The system on chip of claim 14 , wherein the third connection and the fourth connection are different.
16. The system on chip of claim 15 , wherein:
the second bit width is at least 1024 bits wide.
17. The system on chip of claim 15 , wherein:
the first connection comprises a first 1024-bit data register;
the second connection comprises a second 1024-bit data register;
the third connection comprises a third 1024-bit data register; and
the fourth connection comprises a fourth 1024-bit data register.
18. The system on chip of claim 15 , wherein:
the first connection is configured to allow 1024 bits of data to pass between the network input port and the first processor in a single clock cycle of the first processor;
the second connection is configured to allow 1024 bits of data to pass between the network output port and the first processor in the single clock cycle of the first processor;
the third connection is configured to allow 1024 bits of data to pass between the second processor and the first processor in the single clock cycle of the first processor;
the fourth connection is configured to allow 1024 bits of data to pass between the direct memory access circuit and the first processor in the single clock cycle of the first processor.
19. The system on chip of claim 15 , wherein a latency from the network input port to the network output port is unvarying.
20. The system on chip of claim 19 , wherein the latency across the network input port to the network output port is less than thirty-two nanoseconds, and wherein the first processor is a first non-pipelined processor and the second processor is a second non-pipelined processor.