Review Request: 7141246 build-infra merge: Introduce new JVM_VARIANT* to control which kind of jvm gets built

[Kelly O'Hair]

Looks ok to me.

Note that this will likely collide with the FDS (full debug symbols) makefile changes coming down the pipe.

-kto

On Mar 12, 2012, at 5:18 AM, Erik Joelsson wrote:

> http://cr.openjdk.java.net/~erikj/7141246/webrev.01/ <http://cr.openjdk.java.net/%7Eerikj/7141246/webrev.01/>
> 267 lines changed: 119 ins; 40 del; 108 mod; 4676 unch
> 
> 7141246 build-infra merge: Introduce new JVM_VARIANT* to control which kind of jvm gets built
> 
> The build-infra project is starting to move into jdk8. For the hotspot build to stay compatible with the changes, the variables used to control which kind of jvm gets built needs to be cleaned up. We propose the following scheme:
> 
> JVM_VARIANT={server,client,zero etc}
> JVM_VARIANT_SERVER={true,false}
> JVM_VARIANT_CLIENT={true,false}
> etc.
> 
> With this scheme it becomes possible to choose any type of jvm, while the current scheme can't support only server on 32-bit platform.
> 
> Readability in makefile source is much improved with removal of double negations.
> 
> Backwards compatibility is kept by translating the old variables (ZERO_BUILD, BUILD_ONLY_CLIENT etc) to the new variables in make/defs.gmk.
> 
> 
> Additional note: This change will not make all combinations valid or buildable. However, some previously impossible variations are now possible, most notably skipping client or kernel on 32-bit hosts.
> 
> /Erik