Hi Christoph, Thanks for doing this!
On Wed, 2020-03-11 at 08:58 +0000, Langer, Christoph wrote: > Hi, > > for the backports of "JDK-8223678 Add Visual Studio Code workspace > generation support (for native code)" and "JDK-8232748 Build static > versions of certain JDK libraries" it seems useful to first backport > the refactoring of CacheFind. Furthermore it can improve performance > in the Windows Build and will probably also be a nice base for future > backports in the build system area. > > I had to resolve several rejects because there is some code > divergence between jdk11u-dev and the state of jdk13 at the time of > the original change. But in fact, it was quite straightforward and I > replaced all occurences of CacheFind with the new FindFiles function. > Lots of the rejects were also just copyright years. > > In make/hotspot/lib/CompileJvm.gmk, I effectively added the "$(call > FillFindCache, $(JVM_SRC_DIRS))" statement which was only modified in > the original change. I believe we should omit this hunk in the 11u backport, because that logic was added by JDK-8220383[1] which is not in OpenJDK 11u. > Please let me know what you think. > > Bug: https://bugs.openjdk.java.net/browse/JDK-8189861 > Original Change: http://hg.openjdk.java.net/jdk/jdk/rev/e297c7bb6469 > Webrev: http://cr.openjdk.java.net/~clanger/webrevs/8189861.11u/ I've compared it to the original change and it looks good except for the hunk in make/hotspot/lib/CompileJvm.gmk. Andrew's backport then also works for getting the static-libs image built. Thanks, Severin > Thanks & Best regards > Christoph [1] https://bugs.openjdk.java.net/browse/JDK-8220383
