Methods, apparatus, instructions and logic to provide permute controls with leading zero count functionality

1. A processor comprising:

decode circuitry to decode a first instruction specifying a vector leading zero count operation; and

an execution circuit, responsive to a decoded first instruction, to:

read values of data fields in a first register;

for each data field of the data fields in the first register, count a number of most significant contiguous bits set to zero, and

store the count as a value in a corresponding data field in a first destination register, wherein the first instruction is a vector leading zero count instruction used at least in part, to generate a set of permute controls in the second destination register, and a completion mask in the third destination register to be used with the set of permute controls.

2. The processor of claim 1 , wherein the first register comprises a first plurality of data fields, wherein each of the first plurality of data fields in the first register is to store a second plurality of bits and the first destination register comprising a third plurality of data fields corresponding to the first plurality of data fields, wherein each of the third plurality of data fields in the first destination register is to store a count of a number of most significant contiguous bits set to zero for a corresponding data field of the first plurality of data fields, the processor further comprising:

a second register comprising a fourth plurality of data fields corresponding to the third plurality of data fields, wherein each of the fourth plurality of data fields is to store a value of one less than a number of bits in the second plurality of bits

a second destination register comprising a fifth plurality of data fields corresponding to the third plurality of data fields;

said decode circuitry to decode a second instruction; and

the execution circuit, responsive to the decoded second instruction, to:

subtract the value of each of the data fields in the first destination register from a corresponding value of each of the data fields in the second register to generate a corresponding difference, and

store each of the corresponding differences as a value in a corresponding data field of the fifth plurality of data fields in the second destination register.

3. The processor of claim 2 , further comprising:

a third register comprising a sixth plurality of data fields corresponding to the fifth plurality of data fields, wherein each of the sixth plurality of data fields is to store a copy of a specific value;

a third destination register comprising a seventh plurality of mask fields, a portion of the seventh plurality of mask fields corresponding to the fifth plurality of data fields;

said decode stage to decode a third instruction specifying a vector comparison operation; and

the one or more execution units, responsive to the decoded second instruction, to:

compare the value of each of the data fields in the second destination register with a corresponding value of each of the data fields in the third register to generate a corresponding mask value, and

store each of the corresponding mask values in a corresponding mask field of the seventh plurality of mask fields in the third destination register.

4. The processor of claim 3 , wherein the specific value is a value of negative one.

5. The processor of claim 4 , wherein the vector comparison operation is a vector equality comparison operation.

6. The processor of claim 4 , wherein the vector comparison operation is a vector greater-than comparison operation.

7. The processor of claim 1 , wherein the value of one less than a number of bits in the second plurality of bits is thirty-one.

8. The processor of claim 1 , wherein the value of one less than a number of bits in the second plurality of bits is sixty-three.

9. A method comprising:

decoding a first instruction;

executing the decoded first instruction to, for each data field a first source register, counting a number of most significant contiguous bits set to zero and storing the count as a value in a corresponding data field in a first destination register, wherein the first instruction is a vector leading zero count instruction used at least in part, to generate a set of permute controls in the second vector destination register, and a completion mask in the mask destination register to be used with the set of permute controls.

10. The method of claim 9 , further comprising:

executing a second instruction to subtract the value of each of the data fields in the first destination register from a corresponding value of each of the data fields in a second source register to generate a corresponding difference, and store each of the corresponding differences as a value in a corresponding data field f a second destination register.

11. The method of claim 10 , further comprising:

executing a third instruction to compare the value of each of the data fields in the second destination register with a corresponding value of each of the data fields in the third source register to generate a corresponding mask value, and store each of the corresponding mask values in a corresponding mask field of a mask destination register.

12. The method of claim 11 , wherein the specific value is a value of negative one.