The compiler's default optimizations, /O2 on Windows* ( -O2 on Linux* or Mac OS* X), generate optimized code for all IA-32 architecture processors with support for SSE2 instructions and for all Intel®64 architecture processors. The option /arch:IA32 on Windows (-mia32 on Linux or Mac OS X) generates code suitable for IA-32 architecture processors without support for SSE2. In addition, /Qipo (inter-procedural optimization or IPO, -ipo on Linux or Mac OS X), /Qprof_use (profile-guided optimization or PGO, -prof_use on Linux or Mac OS X), and /O3 (high-level loop/memory optimizations, -O3 on Linux or Mac OS X) may further improve performance for many types of applications. These options are available for both Intel® and non-Intel microprocessors but they may perform more optimizations for Intel microprocessors than they perform for non-Intel microprocessors.
Intel® Compiler - How can I generate optimized code to run on any IA-32 or Intel®64 architecture processor?
For more complete information about compiler optimizations, see our Optimization Notice.