[8u-dev, ppc] RFR for non-clean backport of 8185979: PPC64: Implement SHA2 intrinsic

[Kazunori Ogata]

Hi,

May I get review of non-clean backport of 8185979: PPC64: Implement SHA2 
intrinsic?

The original patch itself can be applied cleanly (besides difference of 
the source directory structure).  However, in jdk8u, it also needs to 
change the arguments of make_runtime_call() based on the 
CCallingCenventionRequiresAsLongs flag.  So I made additional changes to 
src/share/vm/opto/library_call.cpp and runtime.cpp.

To separate the change in the original patch and the additional changes, I 
made webrev incrementally.  webrev.02 below only contains the changes by 
the original patch, and webrev.03 contains all changes including the 
additional ones.  The difference between the two webrevs is the changes in 
library_call.cpp and runtime.cpp.


Original bug report:
https://bugs.openjdk.java.net/browse/JDK-8185979

Webrev based on the original patch:
http://cr.openjdk.java.net/~horii/jdk8u_aes_be/8185979/webrev.02/

Webrev of all changes:
http://cr.openjdk.java.net/~horii/jdk8u_aes_be/8185979/webrev.03/


I verified there is no degradation in jtreg (make test) results in both 
fastdebug and release builds.


Regards,
Ogata