[petsc-dev] does next model mess up our histories

[Satish Balay]

On Thu, 2 Oct 2014, Barry Smith wrote:

> 
> > I guess another alternative for these veryfew feature branches
> > that need to iterate over nightlytest suites is:
> > 
> > - never merge feature branch to next untile its complete
> 
>    This is silly. I lose all the portability testing that next provides and would never be able to merge into next because I know things will be broken.

No - you would get the protability test as indicated by the second
bullet listed below. This first requirement is to satisfy Jeds
criteria - never rebase after merge to next.

> 
>    Yupp definitely something wrong with our current model. Seems to me we should just restart next from master every few days.

We are long overdue for a next restart. I go ahead and do that - and
leave the re-merge of feature branches to individual owners.

But I still think its ok to undo merges in next if for whatever reason
you need to rebase [but rebase has to be done as required]

I disagree on 'reset next' every few days as a fix to this problem. I suspect
Jed will have a different objection for this reset..

Satish

> > - switch master or next nightlytest to feature branch [for a few days]
> > - fix rebase feature branch as needed.
> > 
> > [would perhaps require test infrastructure improvements - and an
> > apriori knowledge that this feature branch could go through major back
> > and forth changes requiring multiple rebases]
> > 
> > Satish
> 
>