Справочник Пользователя для AMD 250
Chapter 4
Instruction-Decoding Optimizations
77
Software Optimization Guide for AMD64 Processors
25112
Rev. 3.06
September 2005
4.4
32/64-Bit vs. 16-Bit Forms of the LEA Instruction
Optimization
Use the 32-bit or 64-bit forms of the Load Effective Address (LEA) instruction rather than the 16-bit
form.
form.
Application
This optimization applies to:
•
32-bit software
•
64-bit software
Rationale
The 32-bit and 64-bit LEA instructions are implemented as DirectPath operations with an execution
latency of only two cycles. The 16-bit LEA instruction, however, is a VectorPath instruction, which
lowers the decode bandwidth and has a longer execution latency.
latency of only two cycles. The 16-bit LEA instruction, however, is a VectorPath instruction, which
lowers the decode bandwidth and has a longer execution latency.