Windows-x86 (32-bit) build is broken. Time to retire it?
Magnus Ihse Bursie
magnus.ihse.bursie at oracle.com
Thu Dec 7 11:31:35 UTC 2023
On 2023-12-06 14:40, Mark Reinhold wrote:
> Ports come and go over time. They are introduced via JEPs, and so they
> should be deprecated-for-removal via JEPs and then removed via JEPs.
>
> Fortunately, a JEP to remove the Windows/x86 port will be trivial.
Writing the JEP itself is definitely trivial. To actually remove all
code that will not be needed anymore can be more of a challenge.
Deleting some windows_x86 files is easy and all, but there are likely a
lot of interspersed hacks, if-clauses, ifdefs and workarounds that are
no longer needed; theoretically all windows-specific source code files
will need to be checked.
My guess is that if we do remove it, and even if we try to do a thorough
job, we will continue to find win32-ism's here and there for years to
come. But then at least we can just remove them when we find them, if
the platform support has actually been removed.
Just to be clear: I do think it is worth it, even if it is a lot of job.
The win32 api is substandard and quirky, and our code base will be
better if we can get rid of it.
/Magnus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/jdk-dev/attachments/20231207/97eedee4/attachment-0001.htm>
More information about the jdk-dev
mailing list