On 2017-10-18 07:39, David Holmes wrote:
On 17/10/2017 11:21 PM, Magnus Ihse Bursie wrote:
On 2017-10-17 14:39, David Holmes wrote:
Hi Erik,
Seems reasonable. Though does highlight we still need to close out
8167078.
100% agree. Do you think we can just delete the files that are not
distributed anyway? The bug is assigned to infrastructure, but it's
rather somewhere in the intersection of hotspot and corelibs. Can you
help out to resolve how to proceed with that?
I think it's mainly infrastructure because once we have only one file
we have to modify the build so that it can be included by both jdk and
hotspot code. But that should be quite easy now we have consolidated
repos and the new file layout.
As long as we only have one file, I agree that it's a build issue. But
if there are differences, it must be resolved by the component teams.
Looks like the jni.h differences have already been reconciled, but
there's some work to do with the platform specific jni_md.h files.
Unfortunately JDK version assumes the OS defines the differences while
hotspot (rightly so) uses the CPU. Not sure what happens for OS like
AIX that don't have a JDK variant ??
Actually, I believe the java.base approach is more sane here -- the
hotspot files mostly contains ifdefs on OS, and the blocks for each OS
is shared across CPUs, more or less.
jvm.h and the platform specific variants are also a challenge.
Yeah, they looked messy indeed. I opened JDK-8189610 for that, and put
it on hotspot. I think the burden is mainly on the hotspot team to
rescue any changes that they need, since the java.base version is what's
publicly exported.
jvmticmlr.h is the same in both locations.
Thanks for noticing this. That allowed me to quickly address the issue! :)
/Magnus
David
/Magnus
Thanks,
David
On 17/10/2017 10:32 PM, Erik Joelsson wrote:
In the consolidation restructuring, I missed merging the contents
of make/hotspot/copy/Copy-java.base.gmk with
make/copy/Copy-java.base.gmk. The only relevant part that needs to
be preserved is the optional copying of libffi.so.? which we need
when building zero internally at Oracle. The rest of the contents
of this file is just copy rules for header files that are currently
disabled. These header files are currently being copied from the
old jdk repository instead of the old hotspot repository. Keeping
these unused rules around serves no point.
This patch moves the libffi copying to the correct place and
discards the rest of the forgotten file.
Webrev: http://cr.openjdk.java.net/~erikj/8189430/webrev.01/
Bug: https://bugs.openjdk.java.net/browse/JDK-8189430
/Erik
