[11u] Proposal: Switch jdk11u development to Git/Skara with 11.0.13 cycle

[Andrew Haley]

On 11/02/2021 18:02, Severin Gehwolf wrote:
> On Thu, 2021-02-11 at 17:20 +0000, Andrew Haley wrote:
>> On 11/02/2021 16:46, Andrew Dinn wrote:
>>> On 11/02/2021 13:46, Andrew Haley wrote:
>>>
>>>> It's also important not to consider 11 in isolation: while we do not
>>>> need to move 8 and 11 simultaneously, I very much do not want to see
>>>> them use different workflows for a long period.
>>> I'm also ok with switching jdk11u.
>>>
>>> I'm not sure jdk8u will be quite so easy to switch as it is made up of
>>> multiple hg repos. We can just clone these hg repos into their own
>>> separate github repos. However, I'm not sure if the Skara tool:
ing will
>>> 'just work' when we try to relate multiple change sets committed to
>>> these independent repos to common JIRAs.
>>
>> We can take input from the Skara folk about that. As I understand
>> it, there are already people who have built a monorepo from JDK 8.
>
> Does Skara help for getting a forest into mono-repo shape? For mainline
> JDK it was JEP 296 which did it[0]. My understanding is that having a
> mono-repo is a requirement for a move to git (using skara tooling). If
> that's true, we'd have to be doing something similar for JDK 8u first.
> Failing that, we'd have to do a different approach for moving JDK 8u.

What problems do you forsee? Nothing would be moved or altered, except
perhaps to delete get_source.sh, but even that isn't essential.

Have a look at https://github.com/AdoptOpenJDK/openjdk-jdk8u.

> Also, do we know how well other efforts of getting the mercurial forest
> into a single git tree fares in terms of preserving history?

IMO that one isn't a show stopper, given that the old Hg stuff still
exists.

> So while moving JDK 11u over seems doable (already a mono-repo; there
> is already a git mirror under the openjdk Github namespace[1]), getting
> JDK 8u done seems significanly different and more risky.
>
> With that in mind, should a potential move of 11u to git be (time)
> dependent on a move of jdk8u to git? I'm not sure...
Certainly not.

I do not propose to make 11 wait for 8; but neither do I want 8 to
languish in a dusty corner.

Mind you, there would be one advantage: the more difficult it is
to contribute to 8u, the less churn there would be. Hmm, (half
joking) maybe we should leave 8 on Hg to stop people from
changing it.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671