CFV: New JDK Reviewer: Dmitrij Pochepko

Andrew Dinn adinn at
Tue Jul 30 15:04:07 UTC 2019

On 30/07/2019 14:44, Dmitry Samersoff wrote:
> Thank you for sharing your view, we appreciate your candid feedback.
> Needless to say, that we greatly appreciate the value that you
> constantly bring to OpenJDK.
> However, our perception of the situation is different. We believe that
> you might have missed some facts, for example that, Dmitrij did his best
> to address your concerns promptly and the 5 RFRs from Dmitrij are
> waiting for review.

I am aware of these RFRs. They are still on my radar -- although with a
low priority.

I'll briefly explain (even though it is somewhat tangential to the
proper topic of thread) that this low priority is to do with what I
perceive to be the cost-benefits equation for seeing them through into a
release. The equation differs slightly on the benefit side between the
String intrinsics and the power intrinsic -- the latter looks like it
might have more impact in real use cases. In both cases the cost side is
very much related to the problems I identified in my previous note
relating to the quality of Dmitrij's code. I believe these RFRs require
a lot of review time that I have not so far been able to spare. Without
putting in that time I believe committing the patches would pose a
serious maintenance risk.

Anyway, the nett result is that I have not reviewed these RFRs because
my time has been taken up with what I have determined to be much higher
priority tasks. Of course, I will be happy to discuss this further in a
separate dialogue.

However, note that my veto was not primarily about these issues of code
quality but  about Dmitrij's response to my review comments. So, while
you are indeed right that he responded promptly that doesn't really
answer my concerns about the appropriateness of his responses for
someone who might be considered as a potential reviewer.

> I hope that all the issues can be resolved if we just speak to each
> other. We would like to have an opportunity to clarify your remaining
> concerns and work out a solution plan to address it.
> Therefore, as the first step to
> "work together to reach a mutually-agreeable resolution" could we sync
> with you, Andrew Haley and Dmitrij on CW32 by zoom call?

I'm happy to talk offline -- or in another email thread -- about the
outstanding RFRs and why they have not been prioritised by me (nor by
any other competent AArch64 reviewer, of which we have several). I
cannot speak for Andrew Haley but I would very much expect him to be
willing to be involved in that discussion. However, if you want to
discuss Dmitrij's suitability as a reviewer candidate I think you are
going to have to do that here in this thread. The voting rules are quite
explicit about the need to do that. Besides, this is meant to be an open
source project so even if we were to have a discussion in private it
would still need reporting here.

> I’d like to finish by saying that we are very grateful to you for the
> time you have spent looking into that complex code and hope to have a
> constructive dialogue going forward.
Thank you. I too hope we can proceed in a constructive manner.


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, Eric Shander

More information about the jdk-dev mailing list