Marlin renderer patches for jdk8u integration

[Laurent Bourgès]

Hi Andrew,

Thank you for describing the process with more details than the wiki pages.

Would it be possible to have a branch or another repository 8u-marlin to
integrate Marlin 8u patches there ?

21 patches remain to have Marlin 0.9.1.3 like jdk14.
I already prepared, reviewed & tested the complete final patch in my github
repo.

I propose to skip RenderingEngine class changes that make Marlin the
default renderer until everybody agrees to change the default behaviour
(Pisces) in favor to Marlin.

Anyway I will post soon the first patch for 8u review.

Who can sponsor / review this work as I am only a jdk9+ committer ie not 8u
?

Cheers,
Laurent

Le jeu. 16 janv. 2020 à 18:16, Andrew John Hughes <gnu.andrew at redhat.com> a
écrit :

> On 16/01/2020 09:07, Severin Gehwolf wrote:
> > On Thu, 2020-01-16 at 09:55 +0100, Laurent Bourgès wrote:
> >>
> >>  Should I mark it with the jdk8-backport-request label too ?
> >
> > No, please don't. jdk8u-fix-request label should get added for already
> > *R*eviewed bugs (if they need a review, that is). In fact, the "Fix
> > Request" comment should include the review thread and the people who
> > approved it in that case.
> >
> > Thanks,
> > Severin
> >
>
> Yeah, the process is something along the lines of:
>
> 1. Get a copy of the patch to backport from (preferably) 11u.
>
> 2. Shuffle the paths using jdk11/bin/unshuffle_patch.sh -to9 <11patch>
> <9patch> to get a patch that works with split repos. In some cases, this
> may mean multiple patches for different repos, or there may be multiple
> changesets in 11u (if it was done before the split)
>
> 3. Shuffle again using jdk9/common/bin/unshuffle_patch.sh <repo>
> <9patch> <8patch> for each repo patch.
>
> 4. Try and apply the patch to the repo using hg import <patch>. If this
> succeeds, go to #6. Otherwise:
>
> 5. Fix the patch so that it applies. This may require identifying other
> patches which need to backported first, in which case you start this
> process again with that fix.
>
> 6. Build the resulting JDK, altering if necessary. Again, this may end
> up needing dependent backports.
>
> 6. Run any tests that are introduced or touched by the patch to make
> sure they pass. Again, you may find a missing dependency.
>
> 7. You now should have a working patch. If no changes were necessary
> other than path shuffling, proceed to step #10. Otherwise, step #8.
>
> 8. Use the webrev tool to create a webrev [0].  Post a mail to jdk8u-dev
> with the subject "RFR: <bug ID> <bug description>", asking for a review.
> Explain the motivation and what technical changes were necessary to make
> the patch work with 8u. Include a link to the bug and to your webrev
> uploaded on cr.openjdk.java.net. If you can't upload to
> cr.openjdk.java.net, ask someone with OpenJDK authorship status to do so
> on your behalf.
>
> 9. Wait for a successful review from an OpenJDK 8u reviewer. See [1] for
> the list.
>
> 10. Go to the bug in JIRA and add the label 'jdk8u-fix-request'. Write a
> comment preceded with "Fix Request [8u]". Explain the motivation for the
> fix, and either explain that it applies cleanly to 8u with patch
> shuffling, or link to the successful review thread on the mailing list.
> If you don't have bug database access, someone will need to do this on
> your behalf.
>
> 11. Wait for an 8u maintainer to add jdk8u-fix-yes to the bug.
>
> 12. The change can now be pushed to the appropriate jdk8u-dev repository
> [2]. If you don't have committer or above status, someone will need to
> to do so on your behalf.
>
> 13. Pat yourself on the back, having successfully got a patch into
> OpenJDK 8u.
>
> [0] https://openjdk.java.net/guide/webrevHelp.html
> [1] https://openjdk.java.net/census#jdk8u
>
> Hope that helps (and I didn't miss anything),
> --
> Andrew :)
>
> Senior Free Java Software Engineer
> 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
> https://keybase.io/gnu_andrew
>
>