From: Dario Faggioli <dario.faggioli@citrix.com>
To: Wei Liu <wei.liu2@citrix.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>,
Andrew Cooper <Andrew.Cooper3@citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
"JBeulich@suse.com" <JBeulich@suse.com>,
"ufimtseva@gmail.com" <ufimtseva@gmail.com>,
Ian Jackson <Ian.Jackson@citrix.com>
Subject: Re: [PATCH v4 21/21] xl: vNUMA support
Date: Tue, 24 Feb 2015 16:15:17 +0000 [thread overview]
Message-ID: <1424794515.4742.117.camel@citrix.com> (raw)
In-Reply-To: <20150129174651.GI20229@zion.uk.xensource.com>
[-- Attachment #1.1: Type: text/plain, Size: 1968 bytes --]
On Thu, 2015-01-29 at 17:46 +0000, Wei Liu wrote:
> On Thu, Jan 29, 2015 at 11:10:39AM +0000, Ian Campbell wrote:
> > On Wed, 2015-01-28 at 22:52 +0000, Wei Liu wrote:
Hey,
While reviewing v5, I realized I always wanted to, but never replied to
this email! :-(
Very quickly...
> > > > > +For example, for a guest with 2 virtual nodes, user can specify:
> > > > > +
> > > > > + vnuma_vdistance = [ [10, 20], [20, 10] ]
> > > >
> > > > Any guidance on how a user should choose these numbers?
> > > >
> > >
> > > I think using the number from numactl is good enough.
> >
> > Worth mentioning in the docs I think.
> >
> > > > Do we support a mode where something figures this out based on the
> > > > underlying distances between the pnode to which a vnode is assigned?
> > > >
> > >
> > > Dario is working on that.
> >
> > I thought he was working on automatic numa placement (i.e. figuring out
> > the best set of pnodes to map the vnodes to), whereas what I was
> > suggesting was that given the user has specified a vnode->pnode mapping
> > it should be possible to construct a distances table pretty trivially
> > from that. Is Dario working on that too?
> >
>
> Right. That's trivial inside libxl.
>
> What I meant was Dario was about to touch all these automation stuffs it
> might be trivial for him to just do it all in one go. Of course if he
> has not done that I can add the logic myself.
>
Yes, Ian, that is a valid way of coming up with a default set of
distance values.
And as Wei says, our agreement is that he would make things work when
all the parameters are explicitly specified. Then, I'll take care of the
automation required to deal with situations where some parameters are
missing.
That include triggering automatic placement, in a way that it properly
accounts for vNUMA, and other things, among which default values for
distances, if nothing is specified.
Regards,
Dario
[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
[-- Attachment #2: Type: text/plain, Size: 126 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
prev parent reply other threads:[~2015-02-24 16:15 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-23 11:13 [PATCH v4 00/21] Virtual NUMA for PV and HVM Wei Liu
2015-01-23 11:13 ` [PATCH v4 01/21] xen: dump vNUMA information with debug key "u" Wei Liu
2015-01-23 13:03 ` Jan Beulich
2015-01-23 13:23 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 02/21] xen: make two memory hypercalls vNUMA-aware Wei Liu
2015-01-23 13:16 ` Jan Beulich
2015-01-23 14:46 ` Wei Liu
2015-01-23 15:37 ` Jan Beulich
2015-01-23 15:43 ` Wei Liu
2015-01-23 16:06 ` Wei Liu
2015-01-23 16:17 ` Jan Beulich
2015-01-23 11:13 ` [PATCH v4 03/21] libxc: allocate memory with vNUMA information for PV guest Wei Liu
2015-01-28 16:02 ` Ian Campbell
2015-01-23 11:13 ` [PATCH v4 04/21] libxl: introduce vNUMA types Wei Liu
2015-01-28 16:04 ` Ian Campbell
2015-01-28 21:51 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 05/21] libxl: add vmemrange to libxl__domain_build_state Wei Liu
2015-01-28 16:05 ` Ian Campbell
2015-01-23 11:13 ` [PATCH v4 06/21] libxl: introduce libxl__vnuma_config_check Wei Liu
2015-01-28 16:13 ` Ian Campbell
2015-01-28 21:51 ` Wei Liu
2015-01-29 11:04 ` Ian Campbell
2015-01-29 16:01 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 07/21] libxl: x86: factor out e820_host_sanitize Wei Liu
2015-01-28 16:14 ` Ian Campbell
2015-01-23 11:13 ` [PATCH v4 08/21] libxl: functions to build vmemranges for PV guest Wei Liu
2015-01-28 16:27 ` Ian Campbell
2015-01-28 21:59 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 09/21] libxl: build, check and pass vNUMA info to Xen " Wei Liu
2015-01-28 16:29 ` Ian Campbell
2015-01-23 11:13 ` [PATCH v4 10/21] xen: handle XENMEM_get_vnumainfo in compat_memory_op Wei Liu
2015-01-23 11:13 ` [PATCH v4 11/21] hvmloader: retrieve vNUMA information from hypervisor Wei Liu
2015-01-23 13:27 ` Jan Beulich
2015-01-23 14:17 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 12/21] hvmloader: construct SRAT Wei Liu
2015-01-23 11:13 ` [PATCH v4 13/21] hvmloader: construct SLIT Wei Liu
2015-01-23 11:13 ` [PATCH v4 14/21] libxc: indentation change to xc_hvm_build_x86.c Wei Liu
2015-01-28 16:30 ` Ian Campbell
2015-01-23 11:13 ` [PATCH v4 15/21] libxc: allocate memory with vNUMA information for HVM guest Wei Liu
2015-01-28 16:36 ` Ian Campbell
2015-01-28 22:07 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 16/21] libxl: build, check and pass vNUMA info to Xen " Wei Liu
2015-01-28 16:41 ` Ian Campbell
2015-01-28 22:14 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 17/21] libxl: disallow memory relocation when vNUMA is enabled Wei Liu
2015-01-28 16:41 ` Ian Campbell
2015-01-28 22:22 ` Wei Liu
2015-01-29 11:06 ` Ian Campbell
2015-01-29 16:04 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 18/21] libxlu: rework internal representation of setting Wei Liu
2015-01-28 16:41 ` Ian Campbell
2015-02-11 16:12 ` Ian Jackson
2015-02-12 10:58 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 19/21] libxlu: nested list support Wei Liu
2015-02-11 16:13 ` Ian Jackson
2015-01-23 11:13 ` [PATCH v4 20/21] libxlu: introduce new APIs Wei Liu
2015-02-11 16:17 ` Ian Jackson
2015-02-12 10:57 ` Wei Liu
2015-01-23 11:13 ` [PATCH v4 21/21] xl: vNUMA support Wei Liu
2015-01-28 16:46 ` Ian Campbell
2015-01-28 22:52 ` Wei Liu
2015-01-29 11:10 ` Ian Campbell
2015-01-29 17:46 ` Wei Liu
2015-02-24 16:15 ` Dario Faggioli [this message]
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=1424794515.4742.117.camel@citrix.com \
--to=dario.faggioli@citrix.com \
--cc=Andrew.Cooper3@citrix.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@citrix.com \
--cc=JBeulich@suse.com \
--cc=ufimtseva@gmail.com \
--cc=wei.liu2@citrix.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.