Accelerating eight-way parallel Keccak execution
A method comprises fetching, by fetch circuitry, an encoded XOR3PP instruction comprising at least one opcode, a first source identifier to identify a first register, a second source identifier to identify a second register, a third source identifier to identifier a third register, and a fourth source identifier to identify a fourth operand, wherein the first register is to store a first value, the second register is to store a second value, and the third register is to store a third value, decoding, by decode circuitry, the encoded XOR3PP instruction to generate a decoded XOR3PP instruction; and executing, by execution circuitry, the decoded XOR3PP instruction to determine a first rotational value and a second rotational value, perform a rotate operation on at least a portion of the first value based on the first rotational value to generate a rotated third value, perform an XOR operation on at least a portion of the first value, at least a portion of the second value, and the rotated third value to generate an XOR result, perform a rotate operation on the XOR result based on the second rotational value to generate a rotated XOR; and store the rotated XOR result.
1. A hardware processor, comprising:
fetch circuitry to fetch an encoded instruction comprising at least one opcode, a first source identifier to identify a first register to store a first operand value, a second source identifier to identify a second register to store a second operand value, a third source identifier to identify a third register to store a third operand value, and a fourth operand value;
decode circuitry to decode the encoded instruction to generate a decoded instruction; and
execution circuitry to execute the decoded instruction to:
determine a first rotational value and a second rotational value;
perform a first rotate operation using the first rotational value, wherein the first rotate operation is to include a rotation of a 3 operand XOR result generated based on at least a portion of the first operand value, at least a portion of the second operand value, and at least a portion of the third operand value to generate a first rotated result;
perform a second rotate operation on the first rotated result using the second rotational value to generate a second rotated result; and
store the second rotated result.
2. The hardware processor of claim 1 , further comprising:
commit circuitry to commit a result of the executed instruction.
3. The hardware processor of claim 1 , wherein the first register is a 512-bit register that stores a 64-bit operand value.
4. The hardware processor of claim 1 , wherein the first register stores eight 64-bit words that separately represent operand values.
5. The hardware processor of claim 4 , wherein the executed instruction is performed on the operand values separately represented by all eight 64-bit words.
6. The hardware processor of claim 1 , wherein the first rotational value and the second rotational value are based on a single bit of a multi-bit representation of the fourth operand value.
7. The hardware processor of claim 1 , wherein the first, the second and the third registers are separately capable of storing 512-bits of data to store respective first, second and third operand values and the fourth operand value is an 8-bit integer immediate.
8. A method, comprising:
fetching, by fetch circuitry, an encoded instruction comprising at least one opcode, a first source identifier to identify a first register to store a first operand value, a second source identifier to identify a second register to store a second operand value, a third source identifier to identify a third register to store a third operand value, and a fourth operand value;
decoding, by decode circuitry, the encoded instruction to generate a decoded instruction; and
executing, by execution circuitry, the decoded instruction to:
determine a first rotational value and a second rotational value;
perform a first rotate operation using the first rotational value, wherein the first rotate operation is to include a rotation of a 3 operand XOR result generated based on at least a portion of the first operand value, at least a portion of the second operand value, and at least a portion of the third operand value to generate a first rotated result;
perform a second rotate operation on the first rotated result using the second rotational value to generate a second rotated result; and
store the second rotated XOR result.
9. The method of claim 8 , further comprising:
committing a result of the executed instruction.
10. The method of claim 8 , wherein the first register is a 512-bit register that stores a 64-bit operand value.
11. The method of claim 8 , wherein the first register stores eight 64-bit words that separately represent operand values.
12. The method of claim 11 , wherein the executed instruction is performed on the operand values separately represented by all eight 64-bit words.
13. The method of claim 8 , wherein the first rotational value and the second rotational value are based on a single bit of a multi-bit representation of the fourth operand value.
14. The method of claim 8 , wherein the first, the second and the third registers are separately capable of storing 512-bits of data to store respective first, second and third operand values and the fourth operand value is an 8-bit integer immediate.
15. A non-transitory computer readable medium comprising instructions which, when executed by a processor, configure the processor to:
fetch an encoded instruction comprising at least one opcode, a first source identifier to identify a first register to store a first operand value, a second source identifier to identify a second register to store a second operand value, a third source identifier to identify a third register to store a third operand value, and a fourth operand value;
decode the encoded instruction to generate a decoded instruction; and
execute the decoded instruction to:
determine a first rotational value and a second rotational value;
perform a first rotate operation using the first rotational value, wherein the first rotate operation is to include a rotation of a 3 operand XOR result generated based on at least a portion of the first operand value, at least a portion of the second operand value, and at least a portion of the third operand value to generate a first rotated result;
perform a second rotate operation on the first rotated result using the second rotational value to generate a second rotated result; and
store the second rotated result.
16. The computer readable medium of claim 15 , comprising instructions to:
commit a result of the executed instruction.
17. The computer readable medium of claim 15 , wherein the first register is a 512-bit register that stores a 64-bit operand value.
18. The computer readable medium of claim 15 , wherein the first register stores eight 64-bit words that separately represent operand values.
19. The computer readable medium of claim 18 , the executed instruction is performed on the operand values separately represented by all eight 64-bit words.
20. The computer readable medium of claim 15 , wherein the first rotational value and the second rotational value are based on a single bit of a multi-bit representation of the fourth operand value.
21. The computer readable medium of claim 20 , wherein the first, the second and the third registers are separately capable of storing 512-bits of data to store respective first, second and third operand values and the fourth operand value is an 8-bit integer immediate.