[En-Nut-Discussion] devnut_m3n: Merging to trunk, the SVN way

Uwe Bonnes bon at elektron.ikp.physik.tu-darmstadt.de
Wed Mar 21 17:45:40 CET 2012


>>>>> "Ole" == Ole Reinhardt <ole.reinhardt at embedded-it.de> writes:

...
    >>  Do I have appropriate rights on sourceforge?

    Ole> I just checked your permissions. AFAIK they should be ok to create
    Ole> a new branch by using (all in one line

Okay!
...

    Ole> Btw: I also planned to give it a try and to create an own trunk
    Ole> integration branch at weekend.

I am on tour this weekend, so no time for a merge that soon.

    >> When checking in revisions from trunk, should I check in the results
    >> after resolving conflicts or should all happen one single step? I
    >> think the first solution is better, but it will create a lot of
    >> clutter in the log.

    Ole> First you should only work on the devnut_uwe_cm3integration branch
    Ole> which is then a fresh copy of trunk. There you will merge file by
    Ole> file from devnut_m3n and resolve conflict by conflict.

So you propose _not_ to go step by step for each branch revision, but do the
merge more on a file by file base! However this will loose all svn history
of the cm3 branch!

    Ole> Do a checkin if everything got cleanly resolved. Always try to
    Ole> correctly decide which changes are of interest. Only merge changes
    Ole> from devnut_m3n to trunk if these are really part of the cortex
    Ole> development or related changes (e.g. phy driver, changes on the rtc
    Ole> api) etc. If there are questions on single files / changes, don't
    Ole> hesitate to ask.

There are some changes Ulrich did outside of the scope of CM3 that I didn't
see redone on the branch side. This is the area of most conflicts. 

    Ole> If this is all ok, we will need to review it. Then we can merge
    Ole> everything back into trunk.

    Ole> But as mentioned before I will also try a merge, latest at the
    Ole> weekend.  We could either compare our results before merging back
    Ole> to trunk or you just wait a few days until I'm ready if you are
    Ole> unsure how to proceed.

We can compare result with the tree I produces about two weeks ago. But
probably not with a tree produced as you outlined above.

Seeing alls these problems, I propose a switch to git as soon as possible
after the merge.

Bye
-- 
Uwe Bonnes                bon at elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------



More information about the En-Nut-Discussion mailing list