From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nick Piggin Subject: Re: [RFC][PATCH] Runtime switching of the idle function [take 2] Date: Tue, 29 Nov 2005 17:55:00 +1100 Message-ID: <438BFB44.2080208@yahoo.com.au> References: <20051118220755.GA3029@elte.hu> <1132353689.4735.43.camel@cmn3.stanford.edu> <1132367947.5706.11.camel@localhost.localdomain> <20051124150731.GD2717@elte.hu> <1132952191.24417.14.camel@localhost.localdomain> <20051126130548.GA6503@elte.hu> <1133232503.6328.18.camel@localhost.localdomain> <20051128190253.1b7068d6.akpm@osdl.org> <1133235740.6328.27.camel@localhost.localdomain> <20051128200108.068b2dcd.akpm@osdl.org> <20051129064420.GA15374@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20051129064420.GA15374@elte.hu> Sender: linux-kernel-owner@vger.kernel.org To: Ingo Molnar Cc: Andrew Morton , Steven Rostedt , acpi-devel@lists.sourceforge.net, len.brown@intel.com, nando@ccrma.Stanford.EDU, rlrevell@joe-job.com, linux-kernel@vger.kernel.org, paulmck@us.ibm.com, kr@cybsft.com, tglx@linutronix.de, pluto@agmk.net, john.cooper@timesys.com, bene@linutronix.de, dwalker@mvista.com, trini@kernel.crashing.org, george@mvista.com List-Id: linux-acpi@vger.kernel.org Ingo Molnar wrote: > * Andrew Morton wrote: > > >>>The way to solve this was to set >>> idle=poll. The original patch I sent was to allow the user to change to >>> idle=poll dynamically. This way they could switch to the poll_idle and >>> run there tests (requiring tsc not to drift) and then switch back to the >>> default idle to save on electricity. >> >>Use gettimeofday()? >> >>If it's just for some sort of instrumentation, run NR_CPUS instances >>of a niced-down busyloop, pin each one to a different CPU? That way >>the idle function doesn't get called at all.. > > > idle=poll is also frequently done for performance reasons [it reduces > idle wakeup latency by 10 usecs] - while it could be turned off if the > system has been idle for some time. E.g. cpufreqd could sample idle time > and turn on/off idle=poll. High-performance setups could enable it all > the time. > > as long as it can be done with zero-cost, i dont see why Steven's patch > wouldnt be a plus for us. It's a performance thing, and having runtime > switches for seemless performance features cannot be bad. > Why not just slightly cleanup and extend (eg. to ACPI) the hlt_counter thingy that many architectures already have? Nick -- SUSE Labs, Novell Inc. Send instant messages to your online friends http://au.messenger.yahoo.com