6708714: Optimize long LShift on 32-bits x86
Summary: For small (1-3 bits) left long shifts in 32-bits VM use sets of add+addc instructions instead of shld+shl on new AMD cpus.
Reviewed-by: never
Contributed-by: shrinivas.joshi@amd.com
1cc8dd79fd1cd13d36b385196271a29632c67c3b jdk7-b24
bf2517e15f0c0f950e5b3143c4ca11e2df73dcc1 jdk7-b25
5ae7db536e3fcf6be78e45b240a9058095e0ed38 jdk7-b26
67052ac87fc927d048e62ec54ff42adb230d3f7c jdk7-b27
18dc4ba4739a537fd146f77da51db16efce28da2 jdk7-b28