[jdk17u-dev] RFR: 8376188: Proposal to increase jtreg minimum version to 7.5.1

[Adam Farley]

Gotcha, thanks!

Working on that here: https://github.com/openjdk/jdk21u-dev/pull/2605

Warm Regards,

Adam Farley
Software Developer
Project Adoptium

IBM United Kingdom Limited
Registered in England and Wales with number 741598
Registered office: Building C, IBM Hursley Office, Hursley Park Road, Winchester, Hampshire SO21 2JN


________________________________________
From: Andrew Hughes
Sent: Thursday, February 12, 2026 11:02 PM
To: Adam Farley
Cc: jdk-updates-dev at openjdk.org
Subject: [EXTERNAL] Re: [jdk17u-dev] RFR: 8376188: Proposal to increase jtreg minimum version to 7.5.1

On 11:34 Mon 09 Feb 2026, Adam Farley wrote:
> Hi All,
>
> Win8365790Test is not compatible with jtreg versions earlier than 7.5.1+1, as mentioned in this issue.
>
> https://bugs.openjdk.org/browse/JDK-8376188
>
> Option A: We could backport the change to jdk25 that upgrades the minimum jtreg version to 7.5.1
> https://github.com/openjdk/jdk25u-dev/commit/9042aa82a890b0f0348ac8127dbd3268f2620346
> This solves the current problem and provides protection against future backports.
> However, changing the jtreg minimum could have effects on the other tests.
>
> Option B: We can alter the test in JDK 17 and 21 to make it compatible with jtreg 7.3.1
> This solves the current problem, but provides no protection against future backports.
>
> What do you think?
>
> Warm Regards,
>
> Adam Farley
> Software Developer
> Project Adoptium
>

I would go with option A.  I've actually just been suggesting to
Severin that we need to update the jtreg in 8u (which is still stuck
on 4 or 5) for much the same reason.

The other problem with B is it assumes someone first catches that said
test is incompatible with the older jtreg.  There is a minimum jtreg
version, but no maximum jtreg version.  It is therefore very easy for
someone to use a newer jtreg to run the test when backporting and
completely miss that it won't work with the minimum version.

This has now happened twice in 8u (though with jtreg 7 code not
working with jtreg 5).  The last time led to me hastily rewriting a
test in a security backport in the early hours of the morning.  I
would rather avoid that in future and update the minimum jtreg.  It is
also a good opportunity to run other tests and bring updates to them
as well.

The main pain in this is not long-term, but the short term of having
to backport test fixes.  This is why I tend to believe that we'd
actually be better having a shared test suite independent of the
various source trees (as we used to with GNU Classpath & gcj).

Warm regards,
--
Andrew :)
Pronouns: he / him or they / them
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222

Please contact via e-mail, not proprietary chat networks

Unless otherwise stated above:

IBM United Kingdom Limited
Registered in England and Wales with number 741598
Registered office: Building C, IBM Hursley Office, Hursley Park Road, Winchester, Hampshire SO21 2JN