Re: [Qemu-devel] Large patch set advice

[Daniel P . Berrangé]

On Wed, Apr 25, 2018 at 01:57:01PM -0600, Warner Losh wrote:
> Every Open Source project has its own quirks. Nobody wants the 520 commits
> on the branch that I started with (too many merged rather than rebases in
> the last 5 years). The 320 real commits are a mess. I’ve been preening them
> to get rid of the silly stuff like (back this out, put it back, all the
> ‘spelling/typo/white space’ fixes). At the end of the rebase, I still
> wasn’t to ‘the same’ so I had a bunch of ‘true up’ commits to make it the
> same...

> My first question seems almost rhetorical: that’s not an acceptable state
> of affairs, and curation is needed to upstream. Right?

Yeah, no one is going to want to review 300 patches !

> (4) Deal with the cases where multiple people have worked on a patch by
> putting SoBs for all of them (or reworking them if I can’t get a hold
> of the original authors) on the commits that are merged. I didn’t see
> a specific policy for this, but this seems sane. The alternative seems
> worse (having elements that don’t compile)

The signed off by line is attesting that you agree to this:

   http://developercertificate.org/

IIUC nothing there requires you to add a SoB line for each author,
but you have to be confident that you can agree to terms a) & b),
given the the work you are building on top of.


> (5) Send them in small batches. I’d go insane trying to manage
> 200 concurrent code reviews, and I can’t image that I’m unique
> in this. I also image that fixes in the early part of the series
> may have ripples to the later parts if my past experience with
> these things has any relevance.

Can't  disagree with that !

> Thanks for any help you can give me.

IIUC, the current bsd-user code in GIT is completely broken & unusable.
After all 500 patches have been applied, I'm wondering how much of
the original code is actually left ? If it is a very small percentage,
an alternative strategy might be to send a patch deleting everything
in tree, and then treat your code as if you were adding bsd-user as a
brand new feature. This could potentially let you merge & restructure
the patches into a smaller more easily reviewable set of work. 


Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|