[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnu-arch-users] Re: Lithe Branching
From: |
Zenaan Harkness |
Subject: |
Re: [Gnu-arch-users] Re: Lithe Branching |
Date: |
Tue, 12 Oct 2004 12:33:42 +1000 |
On Tue, 2004-10-12 at 11:24, Miles Bader wrote:
> address@hidden writes:
> > I can easily 'tla get $stable' to run alternate test star-merges
> > of dev1 and dev2, or both combined.
> >
> > But can I easily create an archive of stable+dev1, to then star-merge
> > in dev2, so I can easily test if it is worth doing just such a star
> > merge at this point, in that sequence, or discover it is perhaps
> > better to firstly commit dev2 first, then star-merge in dev1 (possibly
> > so dev1 can make further changes)?
>
> Why do you need any commits to do this test?
The different sequencing produces different conflicts. I find it useful
to see just the additional diffs/ conflicts for the second patch,
without all the changes from the first-applied patch. That way I can
determine which set of diffs/ conflicts is easier to merge.
I don't know why the alternative sequence of changeset application
produces different results, but they do. I have consistently found
that merging in one direction is easier than merging in the other.
I might be confused about something though... sorry if that's the case.
thanks
zen