Re: [PATCH v7 00/20] submodule: convert the rest of 'update' to C

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Glen Choo <chooglen@xxxxxxxxxx> writes:

> Hm, I haven't looked at where the conflicts are yet, but I'll get to it
> as I'm reviewing the rest of the feedback.
>
> And on that note, what do you think of Ævar's suggestion to split off
> the 'easy to review' and 'obvious' patches into their own preparatory
> series? I wonder if this would make it harder or easier to manage the
> conflicts.

It depends on how small an interaction the "obvious and easy" part
has with topics in flight.  In the best case, if there aren't any
the preparatory series may even graduate before the other topics
that interfere with the main part of this series becomes ready.

In a worse case, what the preparatory work to lay more solid
foundation wants to do may contradict what some of these topics in
flight want to do.  Such semantic conflicts need to be resolved
before the main part (and these interfering topics) can move
forward, and with "split off", the core of the contradicting wish
may become easier to see and what needs to be resolved may become
clearer.

So, I do not think of a way for such a split to make things harder
for later.  Of course, the "easy to review" and "obvious" part has
to be justifiable on its own, i.e. "a larger series wants to build
on this foundation and for it to work this part must be done in this
way", when the other topics wants to do the part in question
differently, becomes a much weaker justification.  But if it is
truly "obvious", it is unlikely that the benefit of the change
becomes harder to justify.





[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux