Compressed instruction format
A technique for decoding an instruction in a variable-length instruction set. In one embodiment, an instruction encoding is described, in which legacy, present, and future instruction set extensions are supported, and increased functionality is provided, without expanding the code size and, in some cases, reducing the code size.
1. An apparatus comprising:
a first instruction decode circuitry to indicate to a second instruction decode circuitry a function of each of a plurality of fields within an instruction, wherein the second instruction decode circuitry is to use the function of each of the plurality of fields indicated by the first instruction decode circuitry to translate the instruction from a first format to a second format, wherein the second format is to represent at least one of the plurality of fields with fewer bits than the first format of the instruction, wherein the first format includes at least a prefix field, a REX prefix field, and at least one escape code field; and
execution circuitry to execute the translated instruction.
2. The apparatus of claim 1 , wherein the second format includes a 4-bit field to specify a source operand, a 1 bit field to identify a vector size, and a 1-bit representation of a byte-long prefix of the first format.
3. The apparatus of claim 2 , wherein the second format further includes an escape byte encoded as C5.
4. The apparatus of claim 2 , wherein the identified vector size is one or 128-bit and 256-bit.
5. The apparatus of claim 2 , wherein the second format includes an opcode field, a scale/index/base field, a modR/M field to specify registers and addressing modes, displacement field, and an immediate field.
6. The apparatus of claim 1 , wherein the second format includes 4-bit field to represent a REX byte, 5-bit field to encode an escape field, a 4-bit field to specify a source operand, a 1 bit field to identify a vector size, and a 1-bit representation of a byte-long prefix of the first format.
7. The apparatus of claim 6 , wherein the second format further includes an escape byte encoded as C4.
8. The apparatus of claim 6 , wherein the identified vector size is one or 128-bit and 256-bit.
9. The apparatus of claim 6 , wherein the second format includes an opcode field, a scale/index/base field, a modR/M field to specify registers and addressing modes, displacement field, and an immediate field, wherein the first format includes a prefix field, a REX prefix field, and at least one escape code field.
10. An apparatus comprising:
instruction decode circuitry to decode an instruction encoded in a second format, wherein the second format is a compressed representation of a first format and wherein the first format includes at least a prefix field, a REX prefix field, and at least one escape code field; and
execution circuitry to execute the decoded instruction.
11. The apparatus of claim 10 , wherein the second format includes a 4-bit field to specify a source operand, a 1 bit field to identify a vector size, and a 1-bit representation of a byte-long prefix of the first format.
12. The apparatus of claim 11 , wherein the second format further includes an escape byte encoded as C5.
13. The apparatus of claim 11 , wherein the identified vector size is one or 128-bit and 256-bit.
14. The apparatus of claim 11 , wherein the second format includes an opcode field, a scale/index/base field, a modR/M field to specify registers and addressing modes, displacement field, and an immediate field.
15. The apparatus of claim 10 , wherein the second format includes 4-bit field to represent a REX byte, 5-bit field to encode an escape field, a 4-bit field to specify a source operand, a 1 bit field to identify a vector size, and a 1-bit representation of a byte-long prefix of the first format.
16. The apparatus of claim 15 , wherein the second format further includes an escape byte encoded as C4.
17. The apparatus of claim 15 , wherein the identified vector size is one or 128-bit and 256-bit.
18. The apparatus of claim 15 , wherein the second format includes an opcode field, a scale/index/base field, a modR/M field to specify registers and addressing modes, displacement field, and an immediate field, wherein the first format includes a prefix field, a REX prefix field, and at least one escape code field.