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:
instruction decode logic circuitry to decode an instruction that is to utilize a format including a multi-byte prefix that at least includes four bits to designate a third operand, a single bit to designate a vector size, and two bits to indicate a legacy prefix; and
execution circuitry to execute the decoded instruction according to an opcode of the instruction.
2. The apparatus of claim 1 , wherein a first byte of the multi-byte prefix is 0xC4.
3. The apparatus of claim 1 , wherein a first byte of the multi-byte prefix is 0xC5.
4. The apparatus of claim 1 , wherein the instruction format is to include a scale, index, and base (SIB) field.
5. The apparatus of claim 1 , wherein the instruction format is to include a displacement field.
6. The apparatus of claim 1 , wherein the instruction format is to include an immediate field.
7. The apparatus of claim 1 , wherein the single bit is to indicate the vector size is to designate either 128-bit or a 256-bit vector size.
8. The apparatus of claim 1 , wherein the legacy prefix is one of: none, 0x66, 0xF3, and 0xF2.
9. A method comprising:
decoding an instruction that is to utilize a format including a multi-byte prefix that at least includes four bits to designate a third operand, a single bit to designate a vector size, and two bits to indicate a legacy prefix; and
executing the decoded instruction according to an opcode of the instruction.
10. The method of claim 9 , wherein a first byte of the multi-byte prefix is 0xC4.
11. The method of claim 9 , wherein a first byte of the multi-byte prefix is 0xC5.
12. The method of claim 9 , wherein the instruction format is to include a scale, index, and base (SIB) field.
13. The method of claim 9 , wherein the instruction format is to include a displacement field.
14. The method of claim 9 , wherein the instruction format is to include an immediate field.
15. The method of claim 9 , wherein the single bit to indicate the vector size is to designate either 128-bit or a 256-bit vector size.
16. The method of claim 9 , wherein the legacy prefix is one of: none, 0x66, 0xF3, and 0xF2.
17. The apparatus of claim 1 , further comprising:
memory to store the instruction.
18. The apparatus of claim 1 , wherein the multi-byte prefix is three bytes in size.
19. The apparatus of claim 1 , wherein the multi-byte prefix is two bytes in size.