Re: [PATCH v6 03/18] xl / libxl: push VCPU affinity pinning down to libxl

From: Dario Faggioli <dario.faggioli@citrix.com>
To: Ian Campbell <Ian.Campbell@citrix.com>
Cc: ian.jackson@eu.citrix.com, Wei Liu <wei.liu2@citrix.com>,
	xen-devel@lists.xen.org
Subject: Re: [PATCH v6 03/18] xl / libxl: push VCPU affinity pinning down to libxl
Date: Tue, 10 Jun 2014 15:09:51 +0200	[thread overview]
Message-ID: <1402405791.16827.29.camel@Solace> (raw)
In-Reply-To: <1402405301.1250.107.camel@kazak.uk.xensource.com>

[-- Attachment #1.1: Type: text/plain, Size: 2690 bytes --]

On mar, 2014-06-10 at 14:01 +0100, Ian Campbell wrote:
> On Tue, 2014-06-10 at 09:09 +0100, Wei Liu wrote:
> > On Tue, Jun 10, 2014 at 08:59:06AM +0200, Dario Faggioli wrote:
> > > On Mon, 2014-06-09 at 13:43 +0100, Wei Liu wrote:
> > > > This patch introduces an array of libxl_bitmap called "vcpu_affinity" in
> > > > libxl IDL to preserve VCPU to PCPU mapping. This is necessary for libxl
> > > > to preserve all information to construct a domain.
> > > > 
> > > > Also define LIBXL_HAVE_AFFINITY_LIST in libxl.h to mark the change in
> > > > API.
> > > > 
> > > > Signed-off-by: Wei Liu <wei.liu2@citrix.com>
> > > > Cc: Dario Faggioli <dario.faggioli@citrix.com>
> > > >
> > > So, as far as this patch is concerned:
> > > 
> > > Acked-by: Dario Faggioli <dario.faggioli@citrix.com>
> > > 
> > > This clashes (not that badly, but it does) with my soft affinity series,
> > > that I just resent yesterday night. It's not a big deal, we just need
> > > two arrays of libxl_bitmap-s instead of just one: one for hard affinity,
> > > and one for soft affinity. Probably that also means that
> > > b_info->vcpu_affinity should have to be renamed to something like
> > > b_info->vcpu_hard_affinity (or b_info->vcpu_hard_aff, or
> > > b_info->hard_affinity, if we want to try keep it a bit shorter).
> > > 
> > 
> > I would go for vcpu_hard_affinity. When it comes to interface I prefer
> > it to be as explicit as possible.
> 
> Renaming fields in the API is problematic, so we should make this be
> correct when we first check it in, whichever order that is in.
> 
Good point.

Well, I may be wrong, but it looks to me that the more logical order
would be for Wei's series to go in first, and me to rebase mine on top
of that. If we do it the other way around, Wei's would have to call the
field vcpu_hard_affinity, without the distinction between hard and soft
affinity being present and explained anywhere in the tree.

Not a big deal, probably, if my series follows quickly enough, but
cetainly confusing for future archaeologists, I guess. :-)

This assuming both the series to be pretty, and almost equally, close to
green light.

In an case, it would still be useful for me to have the current version
(v7) of the soft affinity series reviewed, so that I'll know whether
such rebasing is the only thing I should change in v8. :-)

Let me know what you prefer.

Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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