From: George Dunlap <george.dunlap@eu.citrix.com>
To: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Marcus Granado <Marcus.Granado@eu.citrix.com>,
Keir Fraser <keir@xen.org>,
Ian Campbell <Ian.Campbell@citrix.com>,
Li Yechen <lccycc123@gmail.com>,
Andrew Cooper <Andrew.Cooper3@citrix.com>,
Dario Faggioli <dario.faggioli@citrix.com>,
xen-devel@lists.xen.org, Jan Beulich <JBeulich@suse.com>,
Justin Weaver <jtweaver@hawaii.edu>, Matt Wilson <msw@amazon.com>,
Juergen Gross <juergen.gross@ts.fujitsu.com>,
Elena Ufimtseva <ufimtseva@gmail.com>
Subject: Re: [PATCH v2 12/16] libxl: get and set soft affinity
Date: Fri, 15 Nov 2013 12:02:50 +0000 [thread overview]
Message-ID: <52860D6A.30901@eu.citrix.com> (raw)
In-Reply-To: <21124.63867.762364.416870@mariner.uk.xensource.com>
On 14/11/13 16:25, Ian Jackson wrote:
> George Dunlap writes ("Re: [PATCH v2 12/16] libxl: get and set soft affinity"):
>> On 14/11/13 15:11, Ian Jackson wrote:
>>> So perhaps we should have a singe function that can change the
>>> cpupool, hard affinity, and soft affinity, all at once ?
>> I think this is probably a good idea. Would it make sense to basically
>> have libxl_[gs]et_vcpuaffinity2(), which takes a parameter that can
>> specify that the mask is for either hard, soft, or both?
> No, it needs to take at least two separate parameters for the hard and
> soft masks, because the toolstack might want to set the hard affinity
> and soft affinity to different values, without having to think
> carefully about ordering these calls to avoid generating spurious
> warnings and incompatible arrangements.
>
> I think this probably applies to cpupool changes too. So perhaps the
> function should look a bit like setresuid or something: it should take
> an optional new cpupool (is there a sentinel value that can be used to
> mean "no change"?), an optional new soft mask, and an optional new
> hard mask.
>
>>> What if the application makes a call to change the cpupool, without
>>> touching the affinities ? Should the affinities be reset
>>> automatically ?
>> I think whatever happens for hard affinities right now should be carried on.
> Maybe it is a bug that it doesn't do anything. I think it depends how
> we expect people to use this. If a caller sets the hard affinities
> and then changes the cpupool, are they supposed to always then set the
> hard affinities again to a new suitable value ?
Well in fact, as far as I can tell, it *does* do something. When moving
a vcpu to a new pool, it unconditionally calls
cpumask_setall(v->cpu_affinity) for each vcpu, which will effectively
erase the hard affinity. (xen/common/schedule.c:sched_move_domain()).
And, when unplugging cpus, if it unplugs the last cpu a vcpu can run on,
it also resets the affinity to "all".
-George
next prev parent reply other threads:[~2013-11-15 12:02 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-13 19:10 [PATCH v2 00/16] Implement vcpu soft affinity for credit1 Dario Faggioli
2013-11-13 19:11 ` [PATCH v2 01/16] xl: match output of vcpu-list with pinning syntax Dario Faggioli
2013-11-14 10:50 ` George Dunlap
2013-11-14 11:11 ` Dario Faggioli
2013-11-14 11:14 ` George Dunlap
2013-11-14 11:13 ` Dario Faggioli
2013-11-14 12:44 ` Ian Jackson
2013-11-14 14:19 ` Ian Jackson
2013-11-13 19:11 ` [PATCH v2 02/16] xl: allow for node-wise specification of vcpu pinning Dario Faggioli
2013-11-14 11:02 ` George Dunlap
2013-11-14 14:24 ` Ian Jackson
2013-11-14 14:37 ` Dario Faggioli
2013-11-13 19:11 ` [PATCH v2 03/16] xl: implement and enable dryrun mode for `xl vcpu-pin' Dario Faggioli
2013-11-13 19:11 ` [PATCH v2 04/16] xl: test script for the cpumap parser (for vCPU pinning) Dario Faggioli
2013-11-13 19:11 ` [PATCH v2 05/16] xen: fix leaking of v->cpu_affinity_saved Dario Faggioli
2013-11-14 11:11 ` George Dunlap
2013-11-14 11:58 ` Dario Faggioli
2013-11-14 14:25 ` Ian Jackson
2013-11-13 19:11 ` [PATCH v2 06/16] xen: sched: make space for cpu_soft_affinity Dario Faggioli
2013-11-14 15:03 ` George Dunlap
2013-11-14 16:14 ` Dario Faggioli
2013-11-15 10:07 ` George Dunlap
2013-11-13 19:12 ` [PATCH v2 07/16] xen: sched: rename v->cpu_affinity into v->cpu_hard_affinity Dario Faggioli
2013-11-14 14:17 ` George Dunlap
2013-11-13 19:12 ` [PATCH v2 08/16] xen: derive NUMA node affinity from hard and soft CPU affinity Dario Faggioli
2013-11-14 15:21 ` George Dunlap
2013-11-14 16:30 ` Dario Faggioli
2013-11-15 10:52 ` George Dunlap
2013-11-15 14:17 ` Dario Faggioli
2013-11-13 19:12 ` [PATCH v2 09/16] xen: sched: DOMCTL_*vcpuaffinity works with hard and soft affinity Dario Faggioli
2013-11-14 14:42 ` George Dunlap
2013-11-14 16:21 ` Dario Faggioli
2013-11-13 19:12 ` [PATCH v2 10/16] xen: sched: use soft-affinity instead of domain's node-affinity Dario Faggioli
2013-11-14 15:30 ` George Dunlap
2013-11-15 0:39 ` Dario Faggioli
2013-11-15 11:23 ` George Dunlap
2013-11-13 19:12 ` [PATCH v2 11/16] libxc: get and set soft and hard affinity Dario Faggioli
2013-11-14 14:58 ` Ian Jackson
2013-11-14 16:18 ` Dario Faggioli
2013-11-14 15:38 ` George Dunlap
2013-11-14 16:41 ` Dario Faggioli
2013-11-13 19:12 ` [PATCH v2 12/16] libxl: get and set soft affinity Dario Faggioli
2013-11-13 19:16 ` Dario Faggioli
2013-11-14 15:11 ` Ian Jackson
2013-11-14 15:55 ` George Dunlap
2013-11-14 16:25 ` Ian Jackson
2013-11-15 5:13 ` Dario Faggioli
2013-11-15 12:02 ` George Dunlap [this message]
2013-11-15 17:29 ` Dario Faggioli
2013-11-15 3:45 ` Dario Faggioli
2013-11-13 19:12 ` [PATCH v2 13/16] xl: show soft affinity in `xl vcpu-list' Dario Faggioli
2013-11-14 15:12 ` Ian Jackson
2013-11-13 19:13 ` [PATCH v2 14/16] xl: enable setting soft affinity Dario Faggioli
2013-11-13 19:13 ` [PATCH v2 15/16] xl: enable for specifying node-affinity in the config file Dario Faggioli
2013-11-14 15:14 ` Ian Jackson
2013-11-14 16:12 ` Dario Faggioli
2013-11-13 19:13 ` [PATCH v2 16/16] libxl: automatic NUMA placement affects soft affinity Dario Faggioli
2013-11-14 15:17 ` Ian Jackson
2013-11-14 16:11 ` Dario Faggioli
2013-11-14 16:03 ` George Dunlap
2013-11-14 16:48 ` Dario Faggioli
2013-11-14 17:49 ` George Dunlap
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=52860D6A.30901@eu.citrix.com \
--to=george.dunlap@eu.citrix.com \
--cc=Andrew.Cooper3@citrix.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=JBeulich@suse.com \
--cc=Marcus.Granado@eu.citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=jtweaver@hawaii.edu \
--cc=juergen.gross@ts.fujitsu.com \
--cc=keir@xen.org \
--cc=lccycc123@gmail.com \
--cc=msw@amazon.com \
--cc=ufimtseva@gmail.com \
--cc=xen-devel@lists.xen.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).