From: Wei Liu <wei.liu2@citrix.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Wei Liu <wei.liu2@citrix.com>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
Ian Jackson <ian.jackson@eu.citrix.com>,
Ian Campbell <ian.campbell@citrix.com>,
xen-devel@lists.xen.org
Subject: Re: [PATCH RFC 5/8] libxl: introduce smt field
Date: Thu, 25 Feb 2016 16:29:06 +0000 [thread overview]
Message-ID: <20160225162906.GI5694@citrix.com> (raw)
In-Reply-To: <1456174934-22973-6-git-send-email-joao.m.martins@oracle.com>
On Mon, Feb 22, 2016 at 09:02:11PM +0000, Joao Martins wrote:
> By setting it to true, it enables the vcpus set by the guest
> to be seen as a SMT-enabled topology. It uses then
> libxl__cpuid_set_topology() to change the cpuid accordingly.
> This setting is made *before* the cpuid is set so that
> any changes could be overwritten. The number of SMT threads
> are the ones supported by the host. And for that it adds
> another helper routine libxl__count_threads_per_core to fetch
> it.
>
> This feature is useful in HT-enabled hosts so that hard
> pinned domains can see the right topology (i.e. SMT cores exposed
> as SMT cores) which in return would lead guest scheduler making
> better decisions.
>
I think we need more sophisticated configuration options then just a
bool. If we're doing this, there is no reason why we shouldn't allow
administrator to set arbitrary topology for the guest.
Wei.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-02-25 16:29 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-22 21:02 [PATCH RFC 0/8] x86/hvm, libxl: HVM SMT topology support Joao Martins
2016-02-22 21:02 ` [PATCH RFC 1/8] x86/hvm: set initial apicid to vcpu_id Joao Martins
2016-02-25 17:03 ` Jan Beulich
2016-03-02 18:49 ` Joao Martins
2016-02-22 21:02 ` [PATCH RFC 2/8] libxl: remove whitespace on libxl_types.idl Joao Martins
2016-02-25 16:28 ` Wei Liu
2016-03-02 19:14 ` Joao Martins
2016-02-22 21:02 ` [PATCH RFC 3/8] libxl: cpuid: add cache core count support Joao Martins
2016-02-22 21:02 ` [PATCH RFC 4/8] libxl: cpuid: add guest topology support Joao Martins
2016-02-25 16:29 ` Wei Liu
2016-03-02 19:14 ` Joao Martins
2016-02-22 21:02 ` [PATCH RFC 5/8] libxl: introduce smt field Joao Martins
2016-02-25 16:29 ` Wei Liu [this message]
2016-02-22 21:02 ` [PATCH RFC 6/8] xl: introduce smt option Joao Martins
2016-02-22 21:02 ` [PATCH RFC 7/8] libxl: introduce topology fields Joao Martins
2016-02-25 16:29 ` Wei Liu
2016-03-02 19:16 ` Joao Martins
2016-02-22 21:02 ` [PATCH RFC 8/8] xl: introduce topology options Joao Martins
2016-02-25 17:21 ` [PATCH RFC 0/8] x86/hvm, libxl: HVM SMT topology support Andrew Cooper
2016-02-26 15:03 ` Dario Faggioli
2016-02-26 15:27 ` Konrad Rzeszutek Wilk
2016-02-26 15:42 ` Dario Faggioli
2016-02-26 15:48 ` Andrew Cooper
2016-03-02 19:18 ` Joao Martins
2016-03-02 20:03 ` Andrew Cooper
2016-03-03 9:52 ` Joao Martins
2016-03-03 10:24 ` Andrew Cooper
2016-03-03 12:23 ` Joao Martins
2016-03-03 12:48 ` Andrew Cooper
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=20160225162906.GI5694@citrix.com \
--to=wei.liu2@citrix.com \
--cc=ian.campbell@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=joao.m.martins@oracle.com \
--cc=stefano.stabellini@eu.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 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).