SKL Fixes for VectorAPI - 2

[Rukmannagari, Shravya]

Hi All,

I would like to contribute a second set of fixes for VectorAPI SKL issues. The fixes are:

1)      no_mask_reg has been changed from false to true for all byte and short instructions in the assembler

2)      Added support for Byte512 multiplication

3)      Fixed store mask by changing merge masking to false

Could you please review the patch here:
http://cr.openjdk.java.net/~srukmannagar/VectorAPI_SKLFixes/webrev.01/



Thanks,
Shravya.