From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Krafft Subject: Re: [Cbe-oss-dev] [Patch] Resending: cell_add_spuaware_cpufreq_governor.diff Date: Wed, 16 Apr 2008 16:53:07 +0200 Message-ID: <20080416165307.1d7a4258@de.ibm.com> References: <20080118171119.3faa4a6a@de.ibm.com> <200804140221.41726.arnd@arndb.de> <20080416134938.01d27dc2@de.ibm.com> <200804161408.26145.arnd@arndb.de> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1400805366==" Return-path: In-Reply-To: <200804161408.26145.arnd@arndb.de> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: cpufreq-bounces@lists.linux.org.uk Errors-To: cpufreq-bounces+glkc-cpufreq=gmane.org+glkc-cpufreq=gmane.org@lists.linux.org.uk To: Arnd Bergmann Cc: parabelboi@bopserverein.de, cpufreq@lists.linux.org.uk, cbe-oss-dev@ozlabs.org --===============1400805366== Content-Type: multipart/signed; boundary="Sig_/kzAr2xAe+9xAlhJopMIrtMj"; protocol="application/pgp-signature"; micalg=PGP-SHA1 --Sig_/kzAr2xAe+9xAlhJopMIrtMj Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Wed, 16 Apr 2008 14:08:25 +0200 Arnd Bergmann wrote: > On Wednesday 16 April 2008, Christian Krafft wrote: > > This patch is also missing a correct load calculation. > > It works pretty well for spu's running at full time or idling, but not = so > > well for mixed load (i.e. each spu running 50 percent of the time we wo= uld > > switch to fullspeed instead of half speed). >=20 > So why should that be something to optimize for? If the SPUs are all idle, > I would assume we'd be better off throwing the tasks off and putting them > into low-power mode, rather than changing the frequency, right? >=20 > What's the point of your driver, really? >=20 > Arnd <>< What we have now is good enough to - switch to full power if there are spus active (waiting or running) and=20 - switch to low power if not. What we have now is not good enough to - switch to low power if there are spus waiting, but none are running As we cannot _yet_ differentiate between spus waiting and spus running means that in _some_ cases we will not save as much power as we could. The point of submitting the driver is to get it working. Further improvements will follow of course. --=20 Mit freundlichen Gruessen, kind regards, Christian Krafft IBM Systems & Technology Group, Linux Kernel Development IT Specialist Vorsitzender des Aufsichtsrats: Martin Jetter Geschaeftsfuehrung: Herbert Kircher Sitz der Gesellschaft: Boeblingen Registriergericht: Amtsgericht Stuttgart, HRB 243294 --Sig_/kzAr2xAe+9xAlhJopMIrtMj Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.7 (GNU/Linux) iD8DBQFIBhLp6rqK4qDx+dcRAm+cAJ9QdOa9aLtoikipRCzjHvXXY9gGRACfbpbV QyHR+LAK6cb33VR1pjtqQmM= =1pMR -----END PGP SIGNATURE----- --Sig_/kzAr2xAe+9xAlhJopMIrtMj-- --===============1400805366== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Cpufreq mailing list Cpufreq@lists.linux.org.uk http://lists.linux.org.uk/mailman/listinfo/cpufreq --===============1400805366==--