Re: [Xen-devel] [v7 PATCH 09/10] xl: enable for specifying soft-affinity in the config file

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Tue, 2014-06-10 at 17:36 +0200, Dario Faggioli wrote:
> On mar, 2014-06-10 at 15:38 +0100, Ian Campbell wrote:
> > On Tue, 2014-06-10 at 02:45 +0200, Dario Faggioli wrote:
> > > in a similar way to how hard-affinity is specified (i.e.,
> > > exactly how plain vcpu-affinity was being specified before
> > > this change).
> > 
> > It seems that the bulk of this is just code motion, is that right?
> > 
> I'd call it more refactoring than motion, as what I'm doing is actually
> adding a config switch "cpus_soft=", but I'm generalizing the code so
> that it can be used to deal with both the new and the already existing
> one ("cpus=")... That's why I'm not advertising it as code motion. The
> refactoring was requested during v2 and v3 reviews.

The affect is that a bunch of code has moved (or refactored) and then
had new functionality added at the same time, which makes review harder.
Anyway, it's done now but next time please try and separate out into
movement+related fixups and then actual functional changes.

> 
> Also, this patch is the one that will clash the most with Wei's series'.
> Actually, most of what is being refactored will be either killed or
> moved to libxl. (I'm just mentioning this, as we've already agreed with
> Wei that we will cooperate on taking care of conflicts properly, basing,
> of course, on which series goes in first).
> 
> > > +    if (!strcmp(what, "cpus")) {
> > 
> > Elsewhere you use an "int soft", which was the correct choice (if not a
> > bool_t).
> > 
> > If that is changed: Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>
> > 
> Ok, I'll go for it.
> 
> Thanks and Regards,
> Dario
> 



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