From mboxrd@z Thu Jan 1 00:00:00 1970 From: Xavier Bru Date: Thu, 21 Oct 2004 14:11:48 +0000 Subject: Re: [PATCH] top level scheduler domain for ia64 Message-Id: <4177C3A4.8080207@bull.net> List-Id: References: <200410191427.27336.jbarnes@engr.sgi.com> In-Reply-To: <200410191427.27336.jbarnes@engr.sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: linux-ia64@vger.kernel.org Hello Nick & all, Nick Piggin wrote: > Luck, Tony wrote: > > >> + .min_interval =3D 80, \ >> + .max_interval =3D 320, \ >> + .busy_factor =3D 320, \ >> + .imbalance_pct =3D 125, \ >> + .cache_hot_time =3D (10*1000000), \ >> + .balance_interval =3D 100*(63+num_online_cpus())/64, \ >> >> That's a lot of magic numbers and formulae ... are they right? >> How would a user know if they are right. >> > > To be honest you really wouldn't. It would take a lot of careful > testing on numerous workloads and systems. I believe SGI is > starting to do a bit of testing... I don't have the resources to > do many "real world" tests. > > At this stage I wouldn't let them worry you too much :P > Hopefully they'll gradually improve. Why should'nt we use the node_distance() function to build in an=20 independant way the Numa hierarchy and compute the right parameters for=20 each level ? --=20 Sinc=E8res salutations. _____________________________________________________________________ =20 Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422 tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE _____________________________________________________________________