From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH RFC 1/2] linux/vnuma: vnuma support for pv guest Date: Wed, 28 Aug 2013 15:08:31 +0200 Message-ID: <1377695311.5255.34.camel@Abyss> References: <1377593580-10040-1-git-send-email-ufimtseva@gmail.com> <1377593580-10040-2-git-send-email-ufimtseva@gmail.com> <20130828012713.GA2840@u109add4315675089e695.ant.amazon.com> <20130828013733.GB2840@u109add4315675089e695.ant.amazon.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============6038571321525763986==" Return-path: In-Reply-To: <20130828013733.GB2840@u109add4315675089e695.ant.amazon.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: Matt Wilson Cc: "konrad.r.wilk" , David Vrabel , Elena Ufimtseva , xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org --===============6038571321525763986== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-4XKgVbpV/bXuFrTtrtEN" --=-4XKgVbpV/bXuFrTtrtEN Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On mar, 2013-08-27 at 18:37 -0700, Matt Wilson wrote: > On Tue, Aug 27, 2013 at 06:27:15PM -0700, Matt Wilson wrote: > > On Tue, Aug 27, 2013 at 04:52:59AM -0400, Elena Ufimtseva wrote: > > > Uses subop hypercall to request XEN about vnuma topology. > > > Sets the memory blocks (aligned by XEN), cpus, distance table > > > on boot. NUMA support should be compiled in kernel. > >=20 > > Are we *really sure* that we want to go this route for PV vNUMA? > > Couldn't we build just enough(tm) of the ACPI tables to express the > > NUMA topology when constructing the domain? That's what we do for the > > e820 map. >=20 > Ignore me somewhat, since the e820 information is retrieved via > hypercall similar to what you're proposing. >=20 :-) > Still, if there's some way that we can reuse existing Linux code > rather than bolting on a completely parallel mechanism to set this up > under PV I think it'd be better. >=20 Well, it looks to me that Elena is reusing quite a bit of it, isn't she? All she's providing is a new initialization function ( xen_numa_init() ), as it is happening already for ACPI NUMA, NUMAQ, and other NUMA implementations. In practice, while ACPI based NUMA code parses the ACPI tables in acpi_numa_init(), PV vNUMA parses the information coming from an hypercall xen_numa_init(). From that point on, Linux that steps-in and do everything else "as usual". Isn't that enough sharing? Thanks and Regards, Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-4XKgVbpV/bXuFrTtrtEN 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 v1.4.14 (GNU/Linux) iEYEABECAAYFAlId9k8ACgkQk4XaBE3IOsQw1ACgobzvG/bgDdtvA1gC+sxGmYz9 eLQAn2KjhdIKJ9kN8m27XredjT5do3ap =o2FC -----END PGP SIGNATURE----- --=-4XKgVbpV/bXuFrTtrtEN-- --===============6038571321525763986== 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 --===============6038571321525763986==--