Backporting features [was Re: RFR (11u, XXL): Upstream/backport Shenandoah to JDK11u]
Andrew Dinn
adinn at redhat.com
Mon Feb 17 17:51:15 UTC 2020
On 17/02/2020 17:39, Claes Redestad wrote:
>
> On 2020-02-17 17:52, Andrew Dinn wrote:
>> Also, I have to note that if the room for novel security challenges is
>> such a concern then I do wonder how Oracle can square that concern with
>> a blithe view of the divergence their decision to pursue their own jdk8u
>> implies? likewise with their decision to release, over many years, a
>> version of jdk8u that included JFR when other releases did not.
>
> I'm not Oracle, and I'm not here to discuss why Oracle invested in and
> sold proprietary, commercial features on top of the OpenJDK for many
> years, nor why we/they finally deciced to converge and contribute all
> those features to the OpenJDK. I have little sway in influencing any
> such decisions.
Of course, I appreciate that, Claes. The comments above were not
addressed at you but at your line of reasoning. The rationale for them
is that having raised the question as to how seriously the OpenJDK
project should consider divergence from Oracle, in particular wrt the
room for issues like security holes to appear, Oracle's past divergent
behaviour needs to be factored into any answer. That's required to
ensure that actions in this project and their potential outcomes are
considered on their merits /relative/ to those actions taken by Oracle
and results observed as a consequence. If the latter actions have not
been significantly harmful to the project then the former cannot easily
be held up as such. It's a matter of establishing perspective and
balance. Of course, I wasn't intending to suggest that you were in any
way resposible for the decisions which led to those actions.
> I will say though that I am personally very happy with where we are in
> the OpenJDK mainline *now*, and as much as I wish we could converge and
> retrofit all these features from Oracle JDK to OpenJDK 8u, that's an
> absolutely massive effort that I think would induce a lot of risks,
> while detracting from the main mission of moving the OpenJDK forward.
Yes, I 100% agree that the Jigsaw hump has been completely flattened and
much improvement has been made in more recent releases. So, many users
should be considering upgrading and absolutely noe of them should be
considering jdk8u for any new projects. However, there are and always
were going to be legacy jdk8u deployments for years to come.
regards,
Andrew Dinn
-----------
Senior Principal Software Engineer
Red Hat UK Ltd
Registered in England and Wales under Company Registration No. 03798903
Directors: Michael Cunningham, Michael ("Mike") O'Neill
More information about the jdk-updates-dev
mailing list