From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH v2] xen: sched: introduce hard and soft affinity in credit 2 scheduler Date: Fri, 17 Jan 2014 23:18:46 +0100 Message-ID: <1389997126.16457.339.camel@Solace> References: <1387044943-5325-1-git-send-email-jtweaver@hawaii.edu> <1387334265.3880.87.camel@Solace> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0988506268020692131==" Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Justin Weaver Cc: George Dunlap , Marcus.Granado@eu.citrix.com, xen-devel@lists.xen.org, Henri Casanova List-Id: xen-devel@lists.xenproject.org --===============0988506268020692131== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-xAcJLPyoZlSBTJoi3JVZ" --=-xAcJLPyoZlSBTJoi3JVZ Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On gio, 2014-01-16 at 10:23 -1000, Justin Weaver wrote: > Dario, >=20 Hey! :-) > Sorry for disappearing for so long ... I'm back and ready to continue wor= king. >=20 NP at all. > Other functions will need to change, but currently with only one run > queue, only runq_candidate needed to change. I'll look through the > others again with the mindset that we (or maybe I) will fix the issue > that is causing only one run queue to be created despite having > multiple cores/sockets available. >=20 > >> Function now chooses the vCPU with the most credit that has hard affin= ity > >> and maybe soft affinity for the given pCPU. If it does not have soft a= ffinity > >> and there is another vCPU that prefers to run on the given pCPU, then = as long > >> as it has at least a certain amount of credit (currently defined as ha= lf of > >> CSCHED_CREDIT_INIT, but more testing is needed to determine the best v= alue) > >> then it is chosen instead. > >> > > Ok, so, why this 'certain amount of credit' thing? I got the technical > > details of it from the code below, but can you spend a few words on why > > and how you think something like this would be required and/or useful? >=20 Allow me to comment only on the 'only one runqueue on multiple socket issue' thing. I honestly think that that one is a bug, so you shouldn't base your work on that behavior. To try facilitate you doing this, I'll try to put together a patch for fixing such issue early next week. I'm not sure wheter it will be accepted in Xen right now or when 4.5 development cycle opens, but at least you can apply that and work on top of it. Would that make sense and be of any help to you? Regards, Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-xAcJLPyoZlSBTJoi3JVZ 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 iEYEABECAAYFAlLZrEYACgkQk4XaBE3IOsQ7iQCfX7kT+szKDfALSeJXiNBEdL9k GasAn1BBQVyPwmmJpSIRAK1K3mbPyWIW =Dzpe -----END PGP SIGNATURE----- --=-xAcJLPyoZlSBTJoi3JVZ-- --===============0988506268020692131== 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 --===============0988506268020692131==--