Sse2 instruction set series II

This section describes the sse2 Integer Operation commands.

5. Data Migration instructions

MovdqaXMM, XMM/m128

MovdqaXMM/m128, XMM
The source memory content value is sent to the destination register. When m128 exists, the memory address must be 16 bytes aligned.

MovdquXMM, XMM/m128

MovdquXMM/m128, XMM
The source memory content value is sent to the destination register, and the memory address does not need to be 16 bytes aligned.

  

Movq2dqXMM, mm
Send the source register content to the destination register with 64-bit low and 64-bit high.

Movdq2qMm, XMM
Send 64-bit low source register content to the destination register.

6. arithmetic operation commands

Key points:

1. The arithmetic operation commands of sse2 are exactly the same as those of MMX commands in the Command mnemonic.

2. The arithmetic operation commands of sse2 operate on the 128-bit XMM registers, while the MMX commands operate on the 64-bit mm registers.

3. In case of memory variables involved in the sse2 command, the variable address must be 16-byte aligned.

4. For instructions, see the MMX Instruction Set section.

 

7. Data packaging and data shuffling commands

 Key points:

 1. The data packaging and rescheduling commands of sse2 are exactly the same as those of MMX data packaging and shuffling commands on the command mnemonic.

2. The arithmetic operation commands of sse2 operate on the 128-bit XMM registers, while the MMX commands operate on the 64-bit mm registers.

 3. In case of memory variables involved in the sse2 command, the variable address must be 16-byte aligned.

4. Generally, if the sse2 command operates 64-bit lower, the MMX command operates 32-bit lower.

 5. For instructions, see the MMX Instruction Set section.