AW: Improving OpenJDK for FreeBSD

[Doerr, Martin]

Hi Harald and Thomas,

> - PPC changes: we already deal with AIX and Linux, and adding BSD is getting a bit #ifdef heavy. I wonder whether we should factor certain things out into (aix|bsd|linux)_ppc.cpp. For example, all of these os/cpu combinations use the same register set, but the ucontext is different for every one of them, so a utility function like `get_gpr_from_context(const ucontext_t* c, int num)` would help. Ultimately, that is up to Martin as ppc lead (I set him on cc).

Right, we should think about some refactoring. In addition, I don’t like having to maintain exact copies of some files for AIX/BSD/Linux.

I think such changes would be appreciated.

One more thing about PPC64: We are planning to remove Power7 support (and older): JDK-8331859<https://bugs.openjdk.org/browse/JDK-8331859>
If anybody is willing to keep old processors supported and is also willing to keep testing them, we may discuss that.

Best regards,
Martin

Von: Thomas Stüfe <thomas.stuefe at gmail.com>
Datum: Freitag, 14. Februar 2025 um 08:32
An: Harald Eilertsen <harald+jdklists at volse.no>
Cc: JDK Dev list <jdk-dev at openjdk.org>, porters-dev <porters-dev at openjdk.org>, David Holmes <david.holmes at oracle.com>, Doerr, Martin <martin.doerr at sap.com>
Betreff: Re: Improving OpenJDK for FreeBSD
Hi Harald,

In addition to what David wrote:

- PPC changes: we already deal with AIX and Linux, and adding BSD is getting a bit #ifdef heavy. I wonder whether we should factor certain things out into (aix|bsd|linux)_ppc.cpp. For example, all of these os/cpu combinations use the same register set, but the ucontext is different for every one of them, so a utility function like `get_gpr_from_context(const ucontext_t* c, int num)` would help. Ultimately, that is up to Martin as ppc lead (I set him on cc).

- the bigger issue to me is the amount of large ifdef sections added to os/bsd and bsd/<arch> : I think its time to separate MacOS from BSD and make both first-class OSes. The amount of duplicate coding would be outweighed by clearer code. Up for discussion, of course; that is just my gut feeling right now.

Cheers, Thomas



On Thu, Feb 13, 2025 at 12:49 PM Harald Eilertsen <harald+jdklists at volse.no<mailto:harald%2Bjdklists at volse.no>> wrote:
On Wed, Feb 05, 2025 at 11:25:51AM +1000, David Holmes wrote:
> Hi Harald,
>
> Can you create a PR against mainline so that we can see the scope of the
> changes and how they have been implemented?

Here you go:
https://github.com/openjdk/jdk/pull/23611

As mentioned in the notes for the PR, this is mainly ment to get your
feedback and as a draft to assess the scope, and to be a base for
discussing how (and if) to proceed with the port.

Looking forward to your thoughts and feedback!

Harald Eilertsen
--
Eilertsens Kodeknekkeri
https://kodeknekkeriet.net<https://kodeknekkeriet.net/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/jdk-dev/attachments/20250214/154cdd1a/attachment-0001.htm>