From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH v8] Some automatic NUMA placement documentation Date: Thu, 26 Jul 2012 18:09:59 +0200 Message-ID: <1343318999.26725.7.camel@Abyss> References: <3f5e9cf6d2931a7968ab.1343317587@Solace> <20497.27009.964251.991072@mariner.uk.xensource.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============8403511395814569508==" Return-path: In-Reply-To: <20497.27009.964251.991072@mariner.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Jackson Cc: Andre Przywara , Ian Campbell , Stefano Stabellini , George Dunlap , Juergen Gross , xen-devel List-Id: xen-devel@lists.xenproject.org --===============8403511395814569508== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-IQvlQL4ikEue7Tu4uf21" --=-IQvlQL4ikEue7Tu4uf21 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Thu, 2012-07-26 at 17:00 +0100, Ian Jackson wrote:=20 > Dario Faggioli writes ("[PATCH v8] Some automatic NUMA placement document= ation"): > > About rationale, usage and (some small bits of) API. >=20 > > +Also, if `numa_placement` is set to `true`, the domain must not > > +have any cpu affinity (i.e., `domain_build_info->cpumap` must > > +have all its bits set, as it is by default), or domain creation > > +will fail returning `ERROR_INVAL`. >=20 > This is no longer true, is it ? >=20 I think it is. Actually, it was you that suggested turning the code into that... :-P It's this thing I'm trying to document: if (libxl_defbool_val(info->numa_placement)) { int rc; if (!libxl_bitmap_is_full(&info->cpumap)) { LOG(ERROR, "Can run NUMA placement only if no vcpu " "affinity is specified"); return ERROR_INVAL; } rc =3D numa_place_domain(gc, domid, info); if (rc) return rc; } Am I missing something? > > +Analyzing various possible placement solutions is what makes the > > +algorithm flexible and quite effective. However, that also means > > +it won't scale well to systems with arbitrary number of nodes. > > +For this reason, if machines with more than 16 NUMA nodes will > > +ever exist, placement will automatically disables itself when > > +running libxl on them. >=20 > This last is very awkwardly phrased. How about: >=20 > For this reason, automatic numa placement is disabled (with > a warning) if it is requested on a host with more than 16 NUMA > nodes. >=20 Fine with it. Will change it. Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://retis.sssup.it/people/faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-IQvlQL4ikEue7Tu4uf21 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEABECAAYFAlARa9cACgkQk4XaBE3IOsRVaQCeLMrQPPBJ3VIgcNSV5S9DRuXy CNMAoJ23i030m5gyvla6dlJyOWmSfSDr =ZtY3 -----END PGP SIGNATURE----- --=-IQvlQL4ikEue7Tu4uf21-- --===============8403511395814569508== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============8403511395814569508==--