2.6.24 - time to switch

[Werner Almesberger]

Andy Green wrote:
> It means if we peel off the onion skins that make the mokopatches in
> order to reapply them in git for history, the inner, earliest patches
> probably won't apply to a modern 2.6.24 tree (they originally worked
> with 2.6.0 or whatever), even though the whole onion would.

No, that doesn't happen. When we change the base tree (or just some
of our patches underneath other patches by us), the patches in SVN+quilt
may need to be updated.

I'm actually not sure how this works in git+stgit, and that's one of
my bigger concerns regarding the workflow:

When you helped me to get your patches from the GIT tree, you explained
that I had to "uncommit" some patches in order to travel backwards. Does
this really mean that past changes disappear for good from the
repository (which would sound like "rewriting history"), or how does
stgit insert those updates on top of existing commits ?

> You don't have to worry about that: you won't have to change until we
> have a chance to evaluate a new system running in parallel with the old
> one and you can satisfy yourself that overall it sucks less.

Good :-)

- Werner