Branches webrevs

[Brian Goetz]

Cool!  

The branches that seem to be most approaching something useful are:
 - jep-334 (already ready to PTT)
 - patterns-stage-1
 - datum
 - condy-folding
 - intrinsics-project



> On Oct 31, 2018, at 6:15 PM, Aleksey Shipilev <shade at redhat.com> wrote:
> 
> Hi,
> 
> In Shenandoah project, we eventually figured that having the automatically-generated webrevs against
> upstream is a very convenient tool for debugging (figuring what exactly was changed), review
> (sometimes, accidental changes leak in, and cause head-scratchy bugs), and merges (because you can
> tell if the change is coming from your code).
> 
> This is the example for amber-demo branch (among other patches for other projects):
>  https://builds.shipilev.net/patch-openjdk-amber-amber-demo/
> 
> If you tell me what other branches are interesting to have, I can add them to the schedule as well.
> The rule of thumb is to get the more or less stable branches that are slated for merge to upstream
> at some point.
> 
> (Technically, webrevs are generated against the first "hg in" changeset's parent, which assumes the
> linear history in "default" branch, and that all branches get merged from default regularly. I think
> that is already happening in amber).
> 
> -Aleksey
>