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
cfeea66a3fa8ca3686a7cfa2d0ce8ab0169f168d jdk7-b24
cbc8ad9dd0e085a607427ea35411990982f19a36 jdk7-b25
9410f77cc30c604d1caf7c9fe3a57fa19e1acbe8 jdk7-b26
11b4dc9f2be3523ef989a0db8459eb56b3045c3a jdk7-b27
56652b46f328937f6b9b5130f1e4cd80f48868ef jdk7-b28