Just-in-time hardware for field programmable gate arrays
A system and method are disclosed for executing a component of a design in a hardware engine. The component is compiled to include an interface that supports an ‘open_loop(n)’ function which, when invoked, requests that the hardware engine run for a specified number of steps before communicating with other hardware or software engines via a runtime system. After the compiled hardware component is transferred to the hardware engine, the hardware engine runs for the specified number of steps unless and until it encounters a system function, such as a ‘display(s)’ function, in the code of the component that requires the runtime system to intervene. The hardware engine pauses awaiting the completion of the system function and continues its execution. The ‘open_loop(n)’ operation of the hardware engine permits components in hardware engines to run at a speed close to the native speed of the target programmable hardware fabric.
1. A method for executing a hardware component of a design that includes a plurality of components executable in a plurality of software or hardware engines, the method comprising:
compiling the hardware component to run in a hardware engine, wherein the compiling includes adding code to support an ‘open_loop’ function, and wherein invoking the ‘open_loop’ function requests that the hardware engine execute a plurality of steps before the hardware component communicates with a component in a different hardware or software engine;
transferring the compiled hardware component to the hardware engine for execution under the control of a scheduler; and
invoking the ‘open_loop’ function of the compiled hardware component at the hardware engine.
2. The method of claim 1 , wherein the hardware engine with the compiled hardware component executes for all of the plurality of steps.
3. The method of claim 1 ,
wherein the compiled hardware component includes a display function for displaying data values used in the compiled hardware component; and
wherein the hardware engine with the compiled hardware component executes for less than all of the plurality of steps in order to execute the display function.
4. The method of claim 1 ,
wherein operation of the compiled hardware component is specified in an intermediate representation of a hardware description language; and
wherein compiling the hardware component includes adding code to the intermediate representation to support an interface that includes the ‘open_loop’ function.
5. The method of claim 4 , wherein the interface is an application binary interface (ABI).
6. The method of claim 4 , wherein the interface includes an end function that stops the compiled hardware component in the hardware engine.
7. The method of claim 1 , wherein the compiled hardware component includes an additional interface for communicating with other components in the plurality of components.
8. The method of claim 7 , wherein the additional interface is an advanced extensible interface (AXI).
9. A system for executing a hardware component of a design that includes a plurality of components, the system comprising:
one or more processors and a memory configured with: a plurality of hardware engines configured to execute one or more of the plurality of components, a plurality of software engines configured to execute one or more of the plurality of components, a scheduler configured to schedule interactions among the plurality of hardware engines and the plurality of software engines, and a runtime system configured to:
compile the hardware component to run in a hardware engine of the plurality of hardware engines, wherein the compiling includes adding code to support an ‘open_loop’ function and wherein invoking the ‘open_loop’ function requests that the hardware engine execute a plurality of steps before the hardware component communicates with a component in a different hardware or software engine;
transfer the compiled hardware component to the hardware engine for execution under the control of the scheduler; and
invoke the ‘open_loop’ function of the compiled hardware component at the hardware engine.
10. The system of claim 9 , wherein the hardware engine with the compiled hardware component executes for all of the plurality of steps.
11. The system of claim 9 ,
wherein the compiled hardware component includes a display function for displaying data values used in the compiled hardware component; and
wherein the hardware engine with the compiled hardware component executes for less than all of the plurality of steps in order for the runtime system to execute the display function.
12. The system of claim 9 ,
wherein operation of the compiled hardware component is specified in an intermediate representation of a hardware description language; and
wherein compiling the hardware component includes adding code to the intermediate representation to support an interface that includes the ‘open_loop’ function.
13. The system of claim 12 , wherein the interface is an application binary interface (ABI).
14. The system of claim 12 , wherein the interface includes an end function that stops the compiled hardware component in the hardware engine.
15. The system of claim 9 , wherein the compiled hardware component includes an additional interface for communicating with other components in the plurality of components.
16. The system of claim 15 , wherein the additional interface is an application extensible interface (AXI).
17. A non-transitory computer readable storage medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform a method for executing a hardware component of a design that includes a plurality of components, the method comprising:
compiling the hardware component to run in a hardware engine, wherein the compiling includes adding code to support an ‘open_loop’ function, and wherein invoking the ‘open_loop’ function requests that the hardware engine execute a plurality of steps before the hardware component communicates with a component in a different hardware or software engine;
transferring the compiled hardware component to the hardware engine for execution under the control of a scheduler; and
invoking the ‘open_loop’ function of the compiled hardware component at the hardware engine.
18. The non-transitory computer-readable storage medium of claim 17 , wherein the hardware engine with the compiled hardware component executes for all of the plurality of steps.
19. The non-transitory computer-readable storage medium of claim 17 ,
wherein the compiled hardware component includes a display function for displaying data values used in the compiled hardware component; and
wherein the hardware engine with the compiled hardware component executes for less than all of the plurality of steps in order to execute the display function.
20. The non-transitory computer-readable storage medium of claim 17 ,
wherein operation of the compiled hardware component is specified in an intermediate representation of a hardware description language; and
wherein compiling the hardware component includes adding code to the intermediate representation to support an interface that includes the ‘open_loop’ function.