Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754713Ab0BVVuM (ORCPT ); Mon, 22 Feb 2010 16:50:12 -0500 Received: from mail.gmx.net ([213.165.64.20]:52602 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1754170Ab0BVVuJ (ORCPT ); Mon, 22 Feb 2010 16:50:09 -0500 X-Authenticated: #20450766 X-Provags-ID: V01U2FsdGVkX1+ZHMoU8bKZTHAyEfNaOeOVg3OUeBsn7240mhoZyZ q9CadT7tyuhjne Date: Mon, 22 Feb 2010 22:50:00 +0100 (CET) From: Guennadi Liakhovetski To: Arjan van de Ven cc: "Rafael J. Wysocki" , linux-pm@lists.linux-foundation.org, LKML , ACPI Devel Maling List , Venki Pallipadi , Len Brown Subject: Re: What's the right value for idle= (was: Re: [linux-pm] regression on P-II SMP) In-Reply-To: <20100221131144.34a9b8e1@infradead.org> Message-ID: References: <201002212136.03783.rjw@sisk.pl> <20100221131144.34a9b8e1@infradead.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Y-GMX-Trusted: 0 X-FuHaFi: 0.46000000000000002 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5219 Lines: 141 On Sun, 21 Feb 2010, Arjan van de Ven wrote: > can you get us /proc/cpuinfo ? > (only one of the cpus, the other ones are just duplicates) They are a bit different (different stepping): processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 5 model name : Pentium II (Deschutes) stepping : 2 cpu MHz : 399.077 cache size : 512 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr bogomips : 799.35 clflush size : 32 processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 5 model name : Pentium II (Deschutes) stepping : 1 cpu MHz : 399.077 cache size : 512 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr bogomips : 798.19 clflush size : 32 > if the CPU advertizes mwait, we normally ought to use it, > but there's some exceptions to that due to buggy-as-**** bioses. > > can you also run powertop -d and get that output ? > (preferably on a fully known working kernel) on 2.6.25 with "root=/dev/sdb1 acpi=noirq scsi_mod.scan=sync netconsole=..." kernel command line (dmesg follows in a reply to another mail) # powertop -d PowerTOP 1.10 (C) 2007, 2008 Intel Corporation Collecting data for 15 seconds < Detailed C-state information is not available.> P-states (frequencies) Wakeups-from-idle per second : 40.3 interval: 15.0s no ACPI power usage estimate available Top causes for wakeups: 41.2% ( 26.1) : ohci_hcd:usb3, tmscsim 40.2% ( 25.5) : ohci_hcd:usb1, bttv0 6.3% ( 4.0) : usb_hcd_poll_rh_status (rh_timer_func) 3.2% ( 2.0) : clocksource_register (clocksource_watchdog) 3.2% ( 2.0) : sym_timer (sym53c8xx_timer) 0.9% ( 0.6) : eth0 0.8% ( 0.5) hald-addon-stor : schedule_timeout (process_timeout) 0.8% ( 0.5) : __netdev_watchdog_up (dev_watchdog) 0.8% ( 0.5) : e100_up (e100_watchdog) 0.7% ( 0.5) : neigh_table_init_no_netlink (neigh_periodic_timer) 0.3% ( 0.2) : sym53c8xx 0.3% ( 0.2) init : schedule_timeout (process_timeout) 0.3% ( 0.2) : page_writeback_init (wb_timer_fn) 0.3% ( 0.2) hald-addon-acpi : do_nanosleep (hrtimer_wakeup) 0.1% ( 0.1) sshd : sk_reset_timer (tcp_write_timer) 0.1% ( 0.1) : neigh_add_timer (neigh_timer_handler) 0.1% ( 0.1) sh : start_this_handle (commit_timeout) 0.1% ( 0.1) cron : do_nanosleep (hrtimer_wakeup) 0.1% ( 0.1) syslogd : do_setitimer (it_real_fn) A USB device is active 0.0% of the time: /sys/bus/usb/devices/3-2.3 Suggestion: increase the VM dirty writeback time from 5.00 to 15 seconds with: echo 1500 > /proc/sys/vm/dirty_writeback_centisecs This wakes the disk up less frequenty for background VM activity Suggestion: enable the noatime filesystem option by executing the following command: mount -o remount,noatime / or by pressing the T key noatime disables persistent access time of file accesses, which causes lots of disk IO. Suggestion: Disable 'hal' from polling your cdrom with: hal-disable-polling --device /dev/cdrom 'hal' is the component that auto-opens a window if you plug in a CD but disables SATA power saving from kicking in. Suggestion: Enable the CONFIG_CPU_FREQ_STAT kernel configuration option. This option allows PowerTOP to show P-state percentages P-states correspond to CPU frequencies. Suggestion: Enable the CONFIG_ACPI_BATTERY kernel configuration option. This option is required to get power estimages from PowerTOP Suggestion: Enable the CONFIG_CPU_FREQ_GOV_ONDEMAND kernel configuration option. The 'ondemand' CPU speed governor will minimize the CPU power usage while giving you performance when it is needed. Suggestion: Enable the CONFIG_USB_SUSPEND kernel configuration option. This option will automatically disable UHCI USB when not in use, and may save approximately 1 Watt of power. Recent USB suspend statistics Active Device name 0.0% USB device usb5 : EHCI Host Controller (Linux 2.6.25 ehci_hcd) 0.0% USB device usb4 : UHCI Host Controller (Linux 2.6.25 uhci_hcd) 0.0% USB device 3-2.4 : USB Multimedia Keyboard (BTC) 0.0% /sys/bus/usb/devices/3-2.3 0.0% USB device 3-2 : Generic USB Hub (ALCOR) 0.0% USB device usb3 : OHCI Host Controller (Linux 2.6.25 ohci_hcd) 0.0% USB device usb2 : OHCI Host Controller (Linux 2.6.25 ohci_hcd) 0.0% USB device usb1 : OHCI Host Controller (Linux 2.6.25 ohci_hcd) Thanks Guennadi --- Guennadi Liakhovetski, Ph.D. Freelance Open-Source Software Developer http://www.open-technology.de/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/