Processor apparatus and methods optimized for control applications
View Patent ↗Apparatus and methods for real-time control using a data processor. In one aspect, the invention comprises an improved processor having one or more extension instructions (and associated supporting pipeline hardware) which are specially adapted for use in a real-time control algorithm running on the processor. In one exemplary embodiment, the processor is a 32-bit pipelined RISC device having custom multiply (CMUL) and multiply-accumulate (CMAC) instructions added to the extension instruction set to optimize algorithm performance in real-time linear time-invariant (LTI) applications. Specialized extension hardware, and methods for generating a processor design adapted for real-time control applications are also disclosed.
1. An apparatus comprising:
a user-configured and extended digital processor core having a pipeline, comprising: a base instruction set, said set comprising a plurality of instructions adapted to run on said core;
at least one extension instruction adapted to run on said core, said at least one extension instruction being user-designated, and optimized for real-time control applications; and
extension hardware operatively coupled to said pipeline of said core, said extension hardware facilitating the running of said at least one extension instruction by said core;
a storage device in data communication with said processor core; and
at least one algorithm disposed at least partly in said storage device, said at least one algorithm being adapted to provide an output useful for real-time control applications based on at least one input and a plurality of state variables, said algorithm utilizing said at least one extension instruction to optimize the operation thereof;
wherein said base instruction set comprises:
a plurality of first instructions having a first length;
a plurality of second instructions having a second length; and
wherein said processor core further comprises logic adapted to decode and process both said first length and second length instructions from a single program having both first and second length instructions contained therein, wherein said first length is less than said second length; and
said first instructions occur more frequently within said single program than said second instructions.
2. The apparatus of claim 1 , wherein said algorithm is adapted to perform iterative calculations within a loop, said act of performing iterative calculations performed according to the method, comprising:
reading a new input sample;
calculating the output value for the current sampling period;
updating a plurality of internal variables; and
determining a partial product.
3. The apparatus of claim 1 , wherein said at least one extension instruction comprises an extension instruction adapted to perform a signed multiplication of a coefficient by a state variable, and add the result to the value held by an accumulator of an arithmetic logic unit.
4. The apparatus of claim 1 , wherein said extension instruction is optimized for iterative performance of a multiplication and addition operation.
5. The apparatus of claim 1 , wherein said at least one extension instruction is adapted to perform primary functions related to said real-time control applications, and said base instruction set is adapted to perform secondary functions associated with said real-time control applications.
6. The apparatus of claim 5 , wherein said primary functions comprise at least repeated signed multiplication of a coefficient by a state variable, and add the result to the value held by an accumulator of an arithmetic logic unit.
7. The apparatus of claim 6 , wherein said secondary functions relate to a state-space control system model.
8. An apparatus comprising:
a user-configured and extended digital processor core, comprising:
a base instruction set, said set comprising a plurality of instructions adapted to run on said core;
at least one extension instruction adapted to run on said core, said at least one extension instruction being user-designated, and optimized for real-time control applications;
a pipeline apparatus, comprising:
an instruction fetch stage;
an instruction decode stage operatively coupled downstream of said fetch stage;
an execution stage operatively coupled downstream of said decode stage; and
a writeback stage operatively coupled downstream of said execution stage;
wherein said fetch, decode, execute, and writeback stages are adapted to process a plurality of instructions of at least said base instruction set comprising a first plurality of 16-bit instructions and a second plurality of 32-bit instructions, without switching modes; and
extension hardware operatively coupled to said pipeline of said core, said extension hardware facilitating the running of said at least one extension instruction by said core;
a storage device in data communication with said processor core; and
at least one algorithm disposed at least partly in said storage device, said at least one algorithm being adapted to provide an output useful for real-time control applications based on at least one input and a plurality of state variables, said algorithm utilizing said at least one extension instruction to optimize the operation thereof.
9. The apparatus of claim 8 , wherein said algorithm is adapted to perform iterative calculations within a loop, said act of performing iterative calculations performed according to the method, comprising:
reading a new input sample;
calculating the output value for the current sampling period;
updating a plurality of internal variables; and
determining a partial product.
10. The apparatus of claim 8 , wherein said at least one extension instruction comprises an extension instruction adapted to perform a signed multiplication of a coefficient by a state variable, and add the result to the value held by an accumulator of an arithmetic logic unit.
11. The apparatus of claim 8 , wherein said extension instruction is optimized for iterative performance of a multiplication and addition operation.
12. The apparatus of claim 8 , wherein said at least one extension instruction is adapted to perform primary functions related to said real-time control applications, and said base instruction set is adapted to perform secondary functions associated with said real-time control applications.
13. The apparatus of claim 12 , wherein said primary functions comprise at least repeated signed multiplication of a coefficient by a state variable, and add the result to the value held by an accumulator of an arithmetic logic unit.
14. The apparatus of claim 13 , wherein said secondary functions relate to a state-space control system model.
15. An apparatus comprising:
a user-configured and extended digital processor core having a pipeline, comprising:
a base instruction set, said set comprising a plurality of instructions adapted to run on said core;
at least one extension instruction adapted to run on said core, said at least one extension instruction being user-designated, and optimized for real-time control applications; and
extension hardware operatively coupled to said pipeline of said core, said extension hardware facilitating the running of said at least one extension instruction by said core;
a storage device in data communication with said processor core; and
at least one algorithm disposed at least partly in said storage device, said at least one algorithm being adapted to provide an output useful for real-time control applications based on at least one input and a plurality of state variables, said at least one algorithm utilizing said at least one extension instruction to optimize the operation thereof;
wherein said base instruction set comprises:
a plurality of first instructions having a first length;
a plurality of second instructions having a second length; and wherein said processor core further comprises logic adapted to decode and process both said first length and second length instructions from a single program having both first and second length instructions contained therein, wherein said first length is less than said second length; and
said first instructions occur more frequently within said single program than said second instructions.
16. An apparatus comprising:
a user-configured and extended digital processor core, comprising:
a base instruction set, said set comprising a plurality of instructions adapted to run on said core;
at least one extension instruction adapted to run on said core, said at least one extension instruction being user-designated, and optimized for real-time control applications;
a pipeline apparatus, comprising:
an instruction fetch stage;
an instruction decode stage operatively coupled downstream of said fetch stage;
an execution stage operatively coupled downstream of said decode stage; and
a writeback stage operatively coupled downstream of said execution stage;
wherein said fetch, decode, execute, and writeback stages are adapted to process a plurality of instructions of at least said base instruction set comprising a first plurality of 16-bit instructions and a second plurality of 32-bit instructions, without switching modes; and
extension hardware operatively coupled to said pipeline of said core, said extension hardware facilitating the running of said at least one extension instruction by said core;
a storage device in data communication with said processor core; and
at least one algorithm disposed at least partly in said a storage device, said at least one algorithm being adapted to provide an output useful for real-time control applications based on at least one input and a plurality of state variables, said at least one algorithm utilizing said at least one extension instruction to optimize the operation thereof.
17. The apparatus of claim 1 , wherein said at least one algorithm, said at least one extension instruction, and said extension hardware collectively implement a state-space representation of an LTI controller function.
18. The apparatus of claim 1 , further comprising a plurality of formats adapted to represent coefficients and state variables associated with said at least one algorithm.
19. The apparatus of claim 18 , wherein at least one of said plurality of formats comprises a coefficient format, said coefficient format comprising a low-precision floating-point form having a mantissa portion and an exponent portion.
20. The apparatus of claim 19 , wherein the position of a binary point in said mantissa portion is predetermined to allow for fractional values.
21. The apparatus of claim 8 , wherein said at least one algorithm, said at least one extension instruction, and said extension hardware collectively implement a state-space representation of an LTI controller function.
22. The apparatus of claim 8 , further comprising a plurality of formats adapted to represent coefficients and state variables associated with said at least one algorithm.
23. The apparatus of claim 22 , wherein at least one of said plurality of formats comprises a coefficient format, said coefficient format comprising a low-precision floating-point form having a mantissa portion and an exponent portion.
24. The apparatus of claim 23 , wherein the position of a binary point in said mantissa portion is predetermined to allow for fractional values.
25. An apparatus comprising:
a user-configured and extended digital processor core, comprising:
a base instruction set, said set comprising a plurality of instructions adapted to run on said core;
at least one extension instruction adapted to run on said core, said at least one extension instruction being user-designated, and optimized for real-time control applications;
a pipeline apparatus, comprising:
an instruction fetch stage;
an instruction decode stage operatively coupled downstream of said fetch stage;
an execution stage operatively coupled downstream of said decode stage;
a writeback stage operatively coupled downstream of said execution stage;
wherein said fetch, decode, execute, and writeback stages are adapted to process a plurality of instructions of at least said base instruction set comprising a first plurality of instructions comprising a first length and a second plurality of instructions comprising a second length, without switching modes; and
extension hardware operatively coupled to said pipeline of said core, said extension hardware facilitating the running of said at least one extension instruction by said core;
a storage device in data communication with said processor core; and
at least one algorithm disposed at least partly in said storage device, said at least one algorithm being adapted to provide an output useful for real-time control applications based on at least one input and a plurality of state variables, said algorithm utilizing said at least one extension instruction to optimize the operation thereof.
26. The apparatus of claim 25 , wherein said at least one algorithm, said at least one extension instruction, and said extension hardware collectively implement a state-space representation of an LTI controller function.
27. The apparatus of claim 25 , further comprising a plurality of formats adapted to represent coefficients and state variables associated with said at least one algorithm.
28. The apparatus of claim 27 , wherein at least one of said plurality of formats comprises a coefficient format, said coefficient format comprising a low-precision floating-point form having a mantissa portion and an exponent portion.
29. The apparatus of claim 28 , wherein the position of a binary point in said mantissa portion is predetermined to allow for fractional values.