[jdk11u-dev] RFR: 8248238: Implementation: JEP 388: Windows AArch64 Support [v7]

[Bernhard Urban-Forster]

On Wed, 27 Oct 2021 16:25:12 GMT, Andrew Haley <aph at openjdk.org> wrote:

>> Reka Kovacs has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains one commit:
>> 
>>   Backport 9604ee82690f89320614b37bfef4178abc869777
>
> make/autoconf/basics_windows.m4 line 414:
> 
>> 412:       AC_MSG_ERROR([Could not create $FIXPATH_BIN])
>> 413:     fi
>> 414:     AC_MSG_RESULT([yes])
> 
> I don't really understand why any of these changes are necessary.

This is hack to make cross-compilation on Windows work. On tip it's much cleaner thanks to the WINENV work that Magnus did here: https://github.com/openjdk/jdk/commit/d29c78da19ba78214efe9e7856cde30fdd9ba8ab However backporting this to 11 would be much riskier imho, as it touches a lot of things in the build system. Also it's by far not a clean backport; I attempted a backport myself and gave up after a couple of frustrating days. Therefore I suggest to go with what we have here, even though it has an uglierer usage.

I described how to use it here: https://mail.openjdk.java.net/pipermail/aarch64-port-dev/2021-May/012724.html

> make/autoconf/toolchain.m4 line 926:
> 
>> 924:           # Corresponds to --with-toolchain-path
>> 925:           BASIC_EVAL_BUILD_DEVKIT_VARIABLE([BUILD_DEVKIT_TOOLCHAIN_PATH])
>> 926:           # Corresponds to --with-sysroot
> 
> Or these changes. What have they to do with AArch64?

It's required to make cross-compilation work on Windows.

-------------

PR: https://git.openjdk.java.net/jdk11u-dev/pull/301