[Swift-devel] duplicated job submission in swift-0.92?

[Michael Wilde]

----- Original Message -----
> We decided the following:
> - I will revert the changes in the 0.92 branch
> - re-commit bug fixes that were committed after the merge
> - merge the 0.92 branch to trunk
> - fix the problems in trunk

Sounds good. But when and how does the fix get to users? 

Either create a 0.92.1 release (sounds hard based on above)
or create a 0.93 release (in which case should we create the 0.93 branch from trunk as soon as this is fixed?)

How long to re-test? (Thats a question for Sarah, Justin, and Ketan)
Could this include the Cray support mods?
 
> One issue that came up was the exact "workflow" in the
> branching/releasing/merging. I think we discussed this previously on
> this mailing list, and it may just be that Sarah has made a wiki page
> about it. But I'll summarize:
> - before a release, trunk is branched into a release branch
> - heavy development may continue in trunk while the branch only sees
> bug
> fixes
> - after the release, the branch is merged back to trunk (so that bug
> fixes in the branch make it back to trunk)
> - rinse and repeat

Sounds good. Add to this a general approach for fix-only point releases to a release if waiting for the next major release will take too long? (Ie 0.92.1 in the current case?)

I suspect we'd benefit from release-often in smaller increments.

Also, the above workflow should consider how harder changes like the fast branch or coaster block allocator would be done. A parallel development branch in which the developer does frequent upgrades to stay close to trunk?

All above sounds very good. Sarah, is this indeed on the ReleasePlans page, and if not, could you put it there, and we can refine that to cover some of the issues above and discuss?

- Mike

> Mihael
> 
> On Thu, 2011-03-31 at 13:35 -0500, Justin M Wozniak wrote:
> > Sure: (630) 252-3351
> >
> > On Thu, 31 Mar 2011, Mihael Hategan wrote:
> >
> > > Can we have a quick phone chat about this?
> > >
> > > On Thu, 2011-03-31 at 12:48 -0500, Justin M Wozniak wrote:
> > >> By doing a bisection I found that the regression seems to occur
> > >> in Swift
> > >> trunk between 3835:3836 (around 01/01/2011). Still looking...
> > >>
> > >> On Wed, 30 Mar 2011, Mihael Hategan wrote:
> > >>
> > >>> On Wed, 2011-03-30 at 22:12 -0500, Michael Wilde wrote:
> > >>>>
> > >>>> ----- Original Message -----
> > >>>>> Why was trunk merged into the stable branch?
> > >>>>
> > >>>> Can you clarify what you mean here?
> > >>>
> > >>> Cog trunk was merged into cog branch 4.1.8 which corresponds to
> > >>> swift
> > >>> 0.92.
> > >>>
> > >>>>
> > >>>> Trunk merged into *what* stable branch, when?
> > >>>>
> > >>>> Do you mean before 0.92 was generated, or after?
> > >>>
> > >>> After. The 0.92 branch contains code from trunk that was added
> > >>> to trunk
> > >>> after 0.92 was merged. In particular 0.92 was, I think, not
> > >>> supposed to
> > >>> contain stuff from the fast branch and it does.
> > >>>
> > >>>>
> > >>>> Is 0.92 not tagged?
> > >>>
> > >>> It's a branch.
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Swift-devel mailing list
> > >>> Swift-devel at ci.uchicago.edu
> > >>> http://mail.ci.uchicago.edu/mailman/listinfo/swift-devel
> > >>>
> > >>
> > >
> > >
> > >
> >
> 
> 
> _______________________________________________
> Swift-devel mailing list
> Swift-devel at ci.uchicago.edu
> http://mail.ci.uchicago.edu/mailman/listinfo/swift-devel

-- 
Michael Wilde
Computation Institute, University of Chicago
Mathematics and Computer Science Division
Argonne National Laboratory