All of lore.kernel.org
 help / color / mirror / Atom feed
From: Juergen Gross <juergen.gross@ts.fujitsu.com>
To: Ian Campbell <Ian.Campbell@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>,
	"Keir (Xen.org)" <keir@xen.org>,
	Stefano Stabellini <Stefano.Stabellini@eu.citrix.com>,
	"Tim (Xen.org)" <tim@xen.org>,
	Ian Jackson <Ian.Jackson@eu.citrix.com>,
	Jan Beulich <JBeulich@suse.com>
Subject: Re: RFC: Still TODO for 4.2? xl domain numa memory allocation vs xm/xend
Date: Mon, 23 Jan 2012 10:59:40 +0100	[thread overview]
Message-ID: <4F1D2F8C.6040203@ts.fujitsu.com> (raw)
In-Reply-To: <1327050108.17599.113.camel@zakaz.uk.xensource.com>

On 01/20/2012 10:01 AM, Ian Campbell wrote:
> On Fri, 2012-01-20 at 08:15 +0000, Pasi Kärkkäinen wrote:
>> On Fri, Jan 20, 2012 at 07:59:28AM +0000, Ian Campbell wrote:
>>> On Thu, 2012-01-19 at 21:14 +0000, Pasi Kärkkäinen wrote:
>>>> On Wed, Jan 04, 2012 at 04:29:22PM +0000, Ian Campbell wrote:
>>>>> Has anybody got anything else? I'm sure I've missed stuff. Are there any
>>>>> must haves e.g. in the paging/sharing spaces?
>>>>>
>>>> Something that I just remembered:
>>>> http://wiki.xen.org/xenwiki/Xen4.1
>>>>
>>>> "NUMA-aware memory allocation for VMs. xl in Xen 4.1 will allocate
>>>> equal amount of memory from every NUMA node for the VM. xm/xend
>>>> allocates all the memory from the same NUMA node."
>>> I'm not that familiar with the NUMA support but my understanding was
>>> that memory was allocated by libxc/the-hypervisor and not by the
>>> toolstack and that the default was to allocate from the same numa nodes
>>> as domains the processor's were pinned to i.e. if you pin the processors
>>> appropriately the Right Thing just happens. Do you believe this is not
>>> the case and/or not working right with xl?
>>>
>>> CCing Juergen since he added the cpupool support and in particular the
>>> cpupool-numa-split option so I'm hoping he knows something about NUMA
>>> more generally.
>>>
>>>> Is this something that should be looked at?
>>>
>>> Probably, but is anyone doing so?
>>>
>>>> Should the numa memory allocation be an option so it can be controlled
>>>> per domain?
>>> What options did xm provide in this regard?
>>>
>>> Does xl's cpupool (with the cpupool-numa-split option) server the same
>>> purpose?
>>>
>>>> The default libxl behaviour might cause unexpected performance issues
>>>> on multi-socket systems?
>>> I'm not convinced libxl is behaving any different to xend but perhaps
>>> someone can show me the error of my ways.
>>>
>>
>> See this thread:
>> http://old-list-archives.xen.org/archives/html/xen-devel/2011-07/msg01423.html
>>
>> where Stefano wrote:
>> "I think we forgot about this feature but it is important and hopefully
>> somebody will write a patch for it before 4.2 is out."
> Is anyone looking into this?
>
> Does cpupool-numa-split solve this same problem?
>
> I think I forgot to actually CC Juergen when I said, doing that now.

I've just sent a patch which should do the job.
I just have no NUMA machine to test it on, I just tested the patch
doesn't break booting dom0...


Juergen

-- 
Juergen Gross                 Principal Developer Operating Systems
PDG ES&S SWE OS6                       Telephone: +49 (0) 89 3222 2967
Fujitsu Technology Solutions              e-mail: juergen.gross@ts.fujitsu.com
Domagkstr. 28                           Internet: ts.fujitsu.com
D-80807 Muenchen                 Company details: ts.fujitsu.com/imprint.html


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

  parent reply	other threads:[~2012-01-23  9:59 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-04 16:29 RFC: Still TODO for 4.2? Ian Campbell
2012-01-04 16:47 ` Konrad Rzeszutek Wilk
2012-01-04 16:51   ` Stefano Stabellini
2012-01-16 13:42     ` Ian Campbell
2012-01-04 16:55 ` Jan Beulich
2012-01-16 13:39   ` Ian Campbell
2012-01-16 14:48     ` Jan Beulich
2012-01-16 15:00       ` Stefano Stabellini
2012-01-04 17:25 ` Pasi Kärkkäinen
2012-01-04 17:36   ` George Dunlap
2012-01-04 18:20   ` Tim Deegan
2012-01-05 10:39     ` Ian Campbell
2012-01-06 15:24       ` RFC: Still TODO for 4.2? Nested Paging for Intel Nested Virt Pasi Kärkkäinen
2012-01-04 19:21   ` RFC: Still TODO for 4.2? Wei Huang
2012-01-04 19:43     ` Pasi Kärkkäinen
2012-01-04 19:57       ` Wei Huang
2012-01-05  7:27         ` Pasi Kärkkäinen
2012-01-06 15:37         ` Konrad Rzeszutek Wilk
2012-01-06 19:08           ` Wei Huang
2012-02-06 17:57           ` Pasi Kärkkäinen
2012-02-13 17:52             ` Extracting ATI/AMD Radeon VBIOS ROM Pasi Kärkkäinen
2012-01-05 13:19       ` Re : RFC: Still TODO for 4.2? David TECHER
2012-01-05 13:25         ` Ian Campbell
2012-01-05 13:41           ` Re : " David TECHER
2012-01-05 16:18             ` Ian Campbell
2012-01-16 13:28   ` Ian Campbell
2012-01-16 14:39     ` Re : " David TECHER
2012-01-04 17:39 ` Roger Pau Monné
2012-01-05 18:07   ` Driver domains and hotplug scripts, redux Ian Jackson
2012-01-06 12:01     ` Stefano Stabellini
2012-01-09 10:08     ` Roger Pau Monné
2012-01-09 12:26       ` Stefano Stabellini
2012-01-09 17:39         ` Ian Jackson
     [not found]           ` <alpine.DEB.2.00.1201101448030.3150@kaball-desktop>
     [not found]             ` <20236.23822.715733.455559@mariner.uk.xensource.com>
     [not found]               ` <alpine.DEB.2.00.1201101547540.3150@kaball-desktop>
     [not found]                 ` <20236.24780.865152.458124@mariner.uk.xensource.com>
     [not found]                   ` <alpine.DEB.2.00.1201101619460.3150@kaball-desktop>
     [not found]                     ` <20236.27158.706017.813195@mariner.uk.xensource.com>
     [not found]                       ` <alpine.DEB.2.00.1201101655390.3150@kaball-desktop>
     [not found]                         ` <20236.28931.127139.752426@mariner.uk.xensource.com>
2012-01-11 11:50                           ` Roger Pau Monné
2012-01-11 12:17                             ` Ian Campbell
2012-01-11 14:26                               ` Dave Scott
2012-01-12 16:50                                 ` Ian Jackson
2012-01-12 18:07                                   ` Dave Scott
2012-01-11 14:44                               ` Roger Pau Monné
2012-01-12 16:48                               ` Ian Jackson
2012-01-16 17:52                                 ` Roger Pau Monné
2012-01-16 17:58                                   ` Ian Jackson
2012-01-17  9:17                                     ` Ian Campbell
2012-01-17  9:30                                       ` Roger Pau Monné
2012-01-17  9:43                                         ` Ian Campbell
2012-01-17  9:40                                       ` Roger Pau Monné
2012-01-17  9:52                                         ` Ian Campbell
2012-01-17 10:00                                           ` Roger Pau Monné
2012-01-17 10:39                                             ` Ian Campbell
2012-01-23 11:40                                               ` Roger Pau Monné
2012-01-27  8:43                                                 ` Roger Pau Monné
2012-01-27 10:57                                                   ` Stefano Stabellini
2012-01-31  9:57                                                     ` Roger Pau Monné
2012-01-31 10:06                                                       ` Tim Deegan
2012-01-31 13:47                                                         ` Stefano Stabellini
2012-01-31 13:51                                                           ` Ian Campbell
2012-01-31 13:51                                                       ` Stefano Stabellini
2012-01-31 20:02                                                       ` Ian Campbell
2012-01-17  9:22                                     ` Roger Pau Monné
2012-01-11 12:50                             ` Stefano Stabellini
2012-01-05 17:49 ` RFC: Still TODO for 4.2? Ian Jackson
2012-01-06 13:37   ` Ian Campbell
2012-01-10 16:06     ` Ian Jackson
2012-01-16 11:55 ` George Dunlap
2012-01-19 21:14 ` RFC: Still TODO for 4.2? xl domain numa memory allocation vs xm/xend Pasi Kärkkäinen
2012-01-20  7:59   ` Ian Campbell
2012-01-20  8:15     ` Pasi Kärkkäinen
2012-01-20  9:01       ` Ian Campbell
2012-01-20  9:47         ` Dario Faggioli
2012-01-20 11:56           ` Ian Campbell
2012-01-23  9:59         ` Juergen Gross [this message]
2012-01-20 10:55     ` Stefano Stabellini
2012-01-20 11:22       ` Ian Campbell
2012-01-20 11:25         ` Stefano Stabellini
2012-01-20 11:44         ` Dario Faggioli
2012-01-20 11:54           ` Ian Campbell
2012-01-20 12:04             ` Dario Faggioli
2012-01-20 12:33               ` Ian Campbell
2012-01-20 13:11                 ` Ian Campbell
2012-01-20 15:06                   ` Ian Campbell
2012-01-20 16:02                     ` Dario Faggioli
2012-01-20 16:21                       ` Ian Campbell
2012-01-20 16:28                         ` Ian Campbell
2012-01-20 16:31                           ` George Dunlap
2012-01-20 16:39                             ` Ian Campbell
2012-01-20 16:43                               ` George Dunlap
2012-01-20 16:54                                 ` Ian Campbell
2012-01-20 17:32                                   ` Dario Faggioli
2012-01-23 10:19                                     ` Ian Campbell
2012-01-23 13:14                                       ` Dario Faggioli
2012-01-23 13:20                                         ` Ian Campbell
2012-01-20 16:55                                 ` Ian Campbell
2012-01-20 16:59                                   ` George Dunlap
2012-01-20 16:58                         ` Dario Faggioli
2012-01-20 17:23                           ` Ian Campbell
2012-01-20 17:28                             ` Dario Faggioli
2012-01-20 11:26       ` Dario Faggioli

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=4F1D2F8C.6040203@ts.fujitsu.com \
    --to=juergen.gross@ts.fujitsu.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=Ian.Jackson@eu.citrix.com \
    --cc=JBeulich@suse.com \
    --cc=Stefano.Stabellini@eu.citrix.com \
    --cc=keir@xen.org \
    --cc=tim@xen.org \
    --cc=xen-devel@lists.xensource.com \
    /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.