Re: [Xen-devel] [v10][PATCH 11/16] tools/libxl: detect and avoid conflicts with RDM

[Ian Campbell <ian.campbell@xxxxxxxxxx>]

On Wed, 2015-07-22 at 17:18 +0800, Chen, Tiejun wrote:
> Thanks for your clarification to me.
> 
> > The solution to this is to be systematic in how you handle the 
> > email
> > based review of a series, not to add a further to the reviewer by 
> > using
> > IRC as well.
> > 
> > For example, here is how I handle review of a series I am working 
> > on:
> > 
> > I keep all the replies to a series I'm working on marked unread. 
> > When I
> > come to rework the series I take the first reply to the first patch 
> > and
> > start a draft reply to it quoting the entire review.
> > 
> > I then go through the comments one by one and either:
> > 
> >        * make the _complete_ code change, including adding the 
> > "Changes
> >          in vN" bit to the commit log and delete that comment from 
> > the
> >          reply
> 
> Are you saying this case of resending this particular patch online?

No, I am talking about making the change in my local git tree. Only
once I have addressed _all_ of the review of the whole series would I
resent a complete new version of the series for further review.

> 
> > or
> >        * write a reply in my draft to that particular comment which 
> > does
> >          one or more of:
> > 
> >                * Explain that I don't understand the suggestion,
> >                  probably asking questions to try and clarify what 
> > was
> >                  being asked for.
> 
> Yes, in this case we're arguing, I was really trying to send a sample 
> of 
> this code fragment to ask this before I sent out the complete change.

I was not talking about any specific case here.

> > At the end of this process _every_ bit of review feedback is addressed
> > either by making the requested change or by responding explaining the
> > reason why the change hasn't been made. This is absolutely crucial. You
> > should never silently ignore a piece of review, even if you don't
> 
> I should double check each line but sometimes this doesn't mean I can 
> 
> understand everything correctly to do right thing as you expect. But 
> this is really not what I intend to do.

I have outlined my strategy for dealing with review which helps
avoid/minimise the error rate (i.e. failing to address comments) when
dealing with reviews. I hope it will be helpful for you in forming your
own strategy.

Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel