Avoid instructions that incur expensive length-changing prefix (LCP) stalls

Avoid instructions that incur expensive length-changing prefix (LCP) stalls
on some x86-64 implementations, notably Core2 and Corei7. Specifically, a move of
a 16-bit constant into memory requires a length-changing prefix and can incur significant
penalties. Modified an old patch written by H.J to split such instructions
during peephole2.

2012-04-05  Teresa Johnson  <tejohnson@google.com>
	    H.J. Lu  <hongjiu.lu@intel.com>

	* config/i386/i386.h (ix86_tune_indices): Add
	X86_TUNE_LCP_STALL.
	* config/i386/i386.md (move immediate to memory peephole2):
	Add cases for HImode move when LCP stall avoidance is needed.
	* config/i386/i386.c (initial_ix86_tune_features): Initialize
	X86_TUNE_LCP_STALL entry.

git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@186176 138bc75d-0d04-0410-961f-82ee72b054a4