Modulus-based error-checking technique

1. A method for checking a result of an operation involving a first number and a second number, comprising:

determining a modulus base p of the first number and a modulus base p of the second number;

performing the operation using the modulus base p of the first number and the modulus base p of the second number;

calculating a modulus base p of a result of the operation involving the first number and the second number;

comparing a result of the performing of the operation with a result of the calculation to identify potential errors associated with the operation; and

repeating the determining, performing, calculating, and comparing operations to identify additional potential errors associated with the operation, wherein the determining and calculating operations are repeated using moduli base q;

wherein p and q are Mersenne prime numbers;

wherein the moduli base p and base q are computed using a shared circuit that performs additions using subsets of consecutive bits in the first number and the second number;

wherein the number of bits in a given subset is greater than or equal to the product of a first characteristic number associated with the modulus base p and a second characteristic number associated with the modulus base q; and

wherein a given characteristic number associated with a given modulus base is a number of bits in a binary representation of the given modulus base.

2. The method of claim 1 , wherein the operation includes multiplication or addition; and

wherein the multiplication includes a full product of the first number and the second number or a partial product, in which some of the bits in the first number and the second number are multiplied.

3. The method of claim 1 , wherein the repeating of the determining, performing, calculating, and comparing operations is performed concurrently with a first instance of these operations.

4. The method of claim 1 , wherein the shared circuit performs the additions after logically aligning the subsets of the bits.

5. The method of claim 1 , wherein at least some of the additions are performed using 4-to-2 compressor circuits.

6. The method of claim 1 , wherein p is 3 and the first characteristic number is 2, p is 7 and the first characteristic number is 3, or p is 31 and the first characteristic number is 5.

7. The method of claim 6 , wherein q is 3 and the second characteristic number is 2, q is 7 and the second characteristic number is 3, or q is 31 and the second characteristic number is 5; and

wherein p is different than q.

8. The method of claim 1 , wherein p is different than q.

9. A modulus circuit configured to determine moduli of an input using shared addition circuits that add subsets of consecutive bits in the input, wherein the input can be a first number, a second number, or a result of an operation involving the first number and the second number;

wherein the moduli include a modulus base p and a modulus base q;

wherein p and q are Mersenne prime numbers;

wherein the number of bits in a given subset is greater than or equal to the product of a first characteristic number associated with the modulus base p and a second characteristic number associated with the modulus base q; and

wherein a given characteristic number associated with a given modulus base is a number of bits in a binary representation of the given modulus base.

10. The modulus circuit of claim 9 , wherein the moduli are used to check a result of the operation involving the first number and the second number.

11. The modulus circuit of claim 9 , wherein the operation includes multiplication or addition; and

wherein the multiplication includes a full product of the first number and the second number or a partial product, in which some of the bits in the first number and the second number are multiplied.

12. The modulus circuit of claim 9 , wherein the shared addition circuits perform additions after logically aligning the subsets of the bits.

13. The modulus circuit of claim 9 , wherein at least some of the shared addition circuits include full-adder circuits.

14. The modulus circuit of claim 9 , wherein at least some of the shared addition circuits include 4-to-2 compressor circuits.

15. The modulus circuit of claim 9 , wherein p is 3 and the first characteristic number is 2, p is 7 and the first characteristic number is 3, or p is 31 and the first characteristic number is 5.

16. The modulus circuit of claim 15 , wherein q is 3 and the second characteristic number is 2, q is 7 and the second characteristic number is 3, or q is 31 and the second characteristic number is 5; and

wherein p is different than q.

17. The modulus circuit of claim 9 , wherein p is different than q.

18. The modulus circuit of claim 9 , wherein the modulus circuit is disposed on an integrated circuit.

19. A computer system, comprising an integrated circuit that includes a modulus circuit which is configured to determine moduli of an input using shared addition circuits that add subsets of consecutive bits in the input, wherein the input can be a first number, a second number, or a result of an operation involving the first number and the second number;

wherein the moduli include a modulus base p and a modulus base q;

wherein p and q are Mersenne prime numbers;

wherein the number of bits in a given subset is greater than or equal to the product of a first characteristic number associated with the modulus base p and a second characteristic number associated with the modulus base q; and

wherein a given characteristic number associated with a given modulus base is a number of bits in a binary representation of the given modulus base.

20. The computer system of claim 19 , wherein the moduli are used to check a result of the operation involving the first number and the second number.