From mboxrd@z Thu Jan 1 00:00:00 1970 From: Martin Steigerwald Subject: Re: [REGRESSION] fan always on with 3.10-rc2 Date: Sun, 09 Jun 2013 20:34:10 +0200 Message-ID: <1461538.eEAU3Zp7hk@merkaba> References: <15568618.SsOOjPSAq8@merkaba> <1825381.bJOGDrYABG@merkaba> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mondschein.lichtvoll.de ([194.150.191.11]:56394 "EHLO mail.lichtvoll.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751706Ab3FISeQ convert rfc822-to-8bit (ORCPT ); Sun, 9 Jun 2013 14:34:16 -0400 In-Reply-To: <1825381.bJOGDrYABG@merkaba> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Linux Kernel Mailing List Cc: Dirk Brandewie , linux-pm@vger.kernel.org, Len Brown , Frederic Weisbecker , Thomas Gleixner , Ingo Molnar , Michael Karcher Am Samstag, 8. Juni 2013, 22:34:44 schrieb Martin Steigerwald: > Am Freitag, 24. Mai 2013, 13:03:18 schrieb Martin Steigerwald: > > Hi! > >=20 > > With 3.10-rc2 I see fan always or almost always on, even during ext= ended > > periods of basically idling around. I did not notice this with 3.9.= This is > > on an ThinkPad T520 with Intel Sandybridge i5-2520M dual core with > > hyperthreading at regularily 2,5 GhZ and Intel graphics (no nvidia)= =2E > >=20 > > I am using full hz: > >=20 > > martin@merkaba:~/Linux/Kernel/Mainline/Bugs/fan always on with 3.10= =2E2-rc2> xzgrep NO_HZ config-3.10.0-rc2-tp520.xz > > CONFIG_NO_HZ_COMMON=3Dy > > # CONFIG_NO_HZ_IDLE is not set > > CONFIG_NO_HZ_FULL=3Dy > > CONFIG_NO_HZ_FULL_ALL=3Dy > > CONFIG_NO_HZ=3Dy > > CONFIG_RCU_FAST_NO_HZ=3Dy > >=20 > > And P-State driver (which I used in 3.9 already as well). > >=20 > > Kernel config attached as xz. Use xzless or xzcat to display. > >=20 > >=20 > > What puzzles is output of powertop, especially: >=20 > Still present in 3.10-rc4. >=20 > I disabled P-State driver but then rpm seems to be even worse. >=20 > Around 2800 rpm all the time, was about 2650 with Intel P State drive= r. >=20 > Next I will try without CONFIG_NO_HZ_FULL and CONFIG_NO_HZ_FULL_ALL. Okay, without these both the behavior is better. Fan still does not see= m to get into non running state, which it was able to with earlier kernel= s (or userspaces?), but its down to a quiet 1950 rpm with Intel P-State d= river enabled again. CC=B4d to people from thread "power management regressio= n in linux 3.10? (related to CONFIG_NO_HZ_FULL)" which I think is related= =2E Powertop values make a lot more sense as well: PowerTOP v2.0 Overview Idle stats Frequency stats Device stat= s Tunables =20 Package | Core | CPU 0 CPU = 1 | | C0 active 1,4% 0,9= % | | POLL 0,0% 0,0 ms = 0,0% 0,0 ms | | C1E-SNB 7,7% 3,0 ms = 2,9% 5,3 ms C2 (pc2) 1,1% | | C3 (pc3) 7,2% | C3 (cc3) 8,4% | C3-SNB 8,2% 3,1 ms = 3,0% 15,2 ms C6 (pc6) 4,9% | C6 (cc6) 9,3% | C6-SNB 7,0% 3,5 ms = 6,2% 13,0 ms C7 (pc7) 9,9% | C7 (cc7) 64,3% | C7-SNB 73,4% 6,8 ms = 84,0% 21,6 ms | Core | CPU 2 CPU = 3 | | C0 active 1,1% 0,9= % | | POLL 0,0% 0,0 ms = 0,0% 0,0 ms | | C1E-SNB 11,0% 6,9 ms = 1,4% 4,8 ms | | | C3 (cc3) 9,3% | C3-SNB 7,7% 5,8 ms = 1,9% 6,0 ms | C6 (cc6) 10,0% | C6-SNB 8,5% 7,4 ms = 4,1% 4,9 ms | C7 (cc7) 64,9% | C7-SNB 71,4% 8,2 ms = 91,1% 7,1 ms PowerTOP v2.0 Overview Idle stats Frequency stats Device stat= s Tunables =20 Package | Core | CPU 0 CPU = 1 | | Actual 2,0 GHz 2,4 G= Hz Idle 54,4% | Idle 97,4% | Idle 98,9% 98,= 2% 2,90 GHz 0,6% | 2,90 GHz 0,1% | 2,90 GHz 0,1% 0,= 1% 2,81 GHz 2,6% | 3,10 GHz 0,2% | 3,10 GHz 0,1% 0,= 1% 3,10 GHz 1,2% | 2,10 GHz 0,1% | 2,10 GHz 0,0% 0,= 0% 3,21 GHz 7,3% | 3,21 GHz 0,9% | 3,21 GHz 0,1% 0,= 9% 2,10 GHz 2,3% | 2,81 GHz 0,2% | 2,81 GHz 0,1% 0,= 1% 3,00 GHz 3,5% | 3,00 GHz 0,0% | 3,00 GHz 0,0% 0,= 0% 2,40 GHz 3,5% | 2,40 GHz 0,1% | 2,40 GHz 0,1% 0,= 1% 1,91 GHz 1,0% | 1,91 GHz 0,1% | 1,91 GHz 0,0% 0,= 0% 2,21 GHz 0,6% | 2,21 GHz 0,1% | 2,21 GHz 0,1% 0,= 1% | Core | CPU 2 CPU = 3 | | Actual 2,4 GHz 2,5 G= Hz | Idle 56,3% | Idle 98,4% 57,= 7% | 2,90 GHz 0,5% | 2,90 GHz 0,1% 0,= 9% | 2,81 GHz 2,5% | 2,81 GHz 0,1% 6,= 4% | 3,10 GHz 1,1% | 3,10 GHz 0,2% 2,= 2% | 3,21 GHz 6,7% | 3,00 GHz 0,0% 2,= 4% | 3,00 GHz 3,5% | 3,21 GHz 0,5% 3,= 4% | 2,71 GHz 2,4% | 2,50 GHz 0,1% 3,= 3% | 2,50 GHz 1,1% | 2,31 GHz 0,1% 0,= 4% | 2,31 GHz 0,2% | 2,71 GHz 0,1% 1,= 0% | 2,60 GHz 1,0% | 2,60 GHz 0,1% 2,= 7% If I haven=B4t stated it. This is on a ThinkPad T520 with dual core Sandybridge i5-2620M at 2.5 GHz (overclockable to 3.2 GhZ), now with Kernel 3.10-rc5. I didn=B4t test this one with the full hz stuff enable= d, only rc2 and rc4. > According to powertop CPU 0 is never idle >=20 > PowerTOP v2.0 Overview Idle stats Frequency stats Device st= ats Tunables =20 >=20 >=20 > Package | Core | CPU 0 CP= U 1 > | | Actual 843 MHz 1= 148 MHz > Turbo Mode 2,2% | Turbo Mode 2,1% | Turbo Mode 2,1% = 0,8% > 2,50 GHz 0,9% | 2,50 GHz 0,9% | 2,50 GHz 0,9% = 0,0% > 2,00 GHz 0,0% | 2,00 GHz 0,0% | 2,00 GHz 0,0% = 0,0% > 1,80 GHz 0,1% | 1,80 GHz 0,1% | 1,80 GHz 0,1% = 0,0% > 1,60 GHz 0,0% | 1,60 GHz 0,0% | 1,60 GHz 0,0% = 0,0% > 1400 MHz 0,0% | 1400 MHz 0,0% | 1400 MHz 0,0% = 0,0% > 1200 MHz 0,0% | 1200 MHz 0,0% | 1200 MHz 0,0% = 0,0% > 1000 MHz 0,0% | 1000 MHz 0,0% | 1000 MHz 0,0% = 0,0% > 800 MHz 87,9% | 800 MHz 87,3% | 800 MHz 87,1% = 3,5% > Idle 9,0% | Idle 9,6% | Idle 9,9% 9= 5,7% >=20 > | Core | CPU 2 CP= U 3 > | | Actual 953 MHz = 905 MHz > | Turbo Mode 0,8% | Turbo Mode 0,7% = 0,1% > | 2,50 GHz 0,7% | 2,50 GHz 0,7% = 0,0% > | 2,00 GHz 0,0% | 2,00 GHz 0,0% = 0,0% > | 1,80 GHz 0,1% | 1,80 GHz 0,1% = 0,0% > | 1,60 GHz 0,0% | 1,60 GHz 0,0% = 0,0% > | 1400 MHz 0,0% | 1400 MHz 0,0% = 0,0% > | 1200 MHz 0,0% | 1200 MHz 0,0% = 0,0% > | 1000 MHz 0,0% | 1000 MHz 0,0% = 0,0% > | 800 MHz 10,3% | 800 MHz 7,4% = 3,7% > | Idle 88,1% | Idle 91,0% 9= 6,2% >=20 >=20 >=20 >=20 > PowerTOP v2.0 Overview Idle stats Frequency stats Device st= ats Tunables =20 >=20 >=20 > Package | Core | CPU 0 CP= U 1 > | | C0 active 32,5% 0= ,3% > | | POLL 96,4% 0,9 m= s 0,0% 0,0 ms > | | C1E-SNB 0,0% 0,0 m= s 0,0% 0,2 ms > C2 (pc2) 0,0% | | > C3 (pc3) 0,0% | C3 (cc3) 0,0% | C3-SNB 0,0% 0,0 m= s 0,0% 0,1 ms > C6 (pc6) 0,0% | C6 (cc6) 0,0% | C6-SNB 0,0% 0,0 m= s 0,0% 0,0 ms > C7 (pc7) 0,0% | C7 (cc7) 0,0% | C7-SNB 0,0% 0,0 m= s 99,2% 10,9 ms >=20 > | Core | CPU 2 CP= U 3 > | | C0 active 1,2% 0= ,7% > | | POLL 0,0% 0,0 m= s 0,0% 0,0 ms > | | C1E-SNB 0,0% 0,1 m= s 0,1% 0,2 ms > | | > | C3 (cc3) 0,1% | C3-SNB 0,0% 0,8 m= s 0,0% 0,2 ms > | C6 (cc6) 0,0% | C6-SNB 0,0% 0,0 m= s 0,0% 0,0 ms > | C7 (cc7) 95,0% | C7-SNB 97,7% 14,5 m= s 97,7% 12,2 ms >=20 >=20 > Old outputs with rc2 and P-State driver for comparison: >=20 > >=20 > > PowerTOP v2.0 Overview Idle stats Frequency stats Device = stats Tunables =20 > >=20 > >=20 > > Package | Core | CPU 0 = CPU 1 > > | | C0 active 127,4% = 2,3% > > ^^^^^^ > > | | POLL 98,1% 0,9= ms 0,0% 0,0 ms > > ^^^^^ > > | | C1E-SNB 0,0% 0,0= ms 0,4% 0,9 ms > > C2 (pc2) 0,0% | | > > C3 (pc3) 0,0% | C3 (cc3) 0,0% | C3-SNB 0,0% 0,0= ms 0,1% 0,8 ms > > C6 (pc6) 0,0% | C6 (cc6) 0,0% | C6-SNB 0,0% 0,0= ms 0,0% 0,8 ms > > C7 (pc7) 0,0% | C7 (cc7) 0,0% | C7-SNB 0,0% 0,0= ms 97,5% 7,4 ms > >=20 > > | Core | CPU 2 = CPU 3 > > | | C0 active 1,9% = 0,8% > > | | POLL 0,0% 0,0= ms 0,0% 0,0 ms > > | | C1E-SNB 1,1% 1,3= ms 0,0% 0,1 ms > > | | > > | C3 (cc3) 0,3% | C3-SNB 0,3% 1,0= ms 0,0% 0,2 ms > > | C6 (cc6) 0,1% | C6-SNB 0,1% 2,5= ms 0,0% 0,3 ms > > | C7 (cc7) 96,2% | C7-SNB 97,0% 7,2= ms 99,3% 17,2 ms > >=20 > >=20 > > PowerTOP v2.0 Overview Idle stats Frequency stats Device = stats Tunables =20 > >=20 > >=20 > > Package | Core | CPU 0 = CPU 1 > > | | Actual 3,2 GHz 3= ,1 GHz > > Idle 100,0% | Idle 100,0% | Idle 100,0% = 100,0% > >=20 > > | Core | CPU 2 = CPU 3 > > | | Actual 3,0 GHz 3= ,0 GHz > > | Idle 100,0% | Idle 100,0% = 100,0% > >=20 > >=20 > > It seems the kernel is overbusying one core completely, if the outp= ut of > > powertop is correct. And why is Actual frequencing of CPUs that hig= h? > > I saw that kernel tends to overtact quickly. I thought this was due= to > > getting work done quickly and then let it idle. But the idle stats > > seem bogus to, maybe powertop is not up to date with current kernel= s? > >=20 > > I see no reason for busying one core. This happens when CPU usage i= s > > below 20%. The fan is consistently around 2640 rpm: > >=20 > >=20 > > PowerTOP v2.0 Overview Idle stats Frequency stats Device = stats Tunables =20 > >=20 > > Summary: 399,8 wakeups/second, 0,0 GPU ops/second, 0,0 VFS ops/sec= and 3,0% CPU use > >=20 > > Usage Events/s Category Description > > 2644 rpm Device Laptop fan > > 100,0% Device Audio codec = hwC0D1: Conexant > > 100,0% Device Audio codec = hwC0D3: Intel > > 100,0% Device Audio codec = hwC0D0: Conexant > > 30,7 =B5s/s 112,3 Process [ksoftirqd= /0] > > 4,2 ms/s 83,5 Process kwin -sessio= n 10cec7d36b000136265311700000023930000_1369322581_349094 > > 0,8 ms/s 67,3 Process [irq/42-i915= @pci] > > 5,5 ms/s 53,6 Process /usr/bin/X := 0 vt7 -br -nolisten tcp -auth /var/run/xauth/A:0-eXsirc > > 698,3 =B5s/s 13,3 Process [irq/16-mm= c0] > > 85,8 =B5s/s 9,1 Process [rcu_preem= pt] > > 297,1 =B5s/s 8,7 Process /usr/sbin/= mysqld --defaults-file=3D/home/martin/.local/share/akonadi/mysql.conf -= -datadir > > 78,6 =B5s/s 8,1 Process [ksoftirqd= /2] > > 6,7 ms/s 3,7 Process /usr/bin/pla= sma-desktop > > 218,9 =B5s/s 5,0 Interrupt [1] timer(= softirq) > > 2,8 ms/s 2,9 Process /usr/bin/kon= sole -session 10cec7d36b000135326160200000249810027_1369322580_899575 > > 65,9 =B5s/s 3,6 Process [ksoftirqd= /1] > > 69,8 =B5s/s 3,5 Process [ksoftirqd= /3] > > 38,1 =B5s/s 3,4 Process [rcuop/3] > > 35,3 =B5s/s 2,4 Process [rcuop/2] > > 6,0 ms/s 0,00 Process atop > > 21,3 =B5s/s 1,8 Process [irq/43-ah= ci] > > 33,3 =B5s/s 1,7 Process [rcuop/0] > > 122,3 =B5s/s 1,7 Process [btrfs-tra= nsacti] > > 66,4 =B5s/s 1,5 Process /usr/bin/d= irmngr --daemon --sh > > 61,2 =B5s/s 1,0 Process [rcuop/1] > > 29,2 =B5s/s 1,0 Process /usr/lib/g= vfs/gvfs-afc-volume-monitor > > 40,8 =B5s/s 0,9 Timer process_ti= meout > > 25,5 =B5s/s 0,9 Process /usr/bin/p= ython /usr/bin/hp-systray -x > > 32,2 =B5s/s 0,7 Process /usr/bin/d= bus-daemon --fork --print-pid 5 --print-address 7 --session > > 5,1 =B5s/s 0,7 Process [irq/44-et= h0] > > 612,1 =B5s/s 0,4 Process ksysguardd > > 137,6 =B5s/s 0,5 Interrupt [7] sched(= softirq) > > 5,6 =B5s/s 0,5 Timer clocksourc= e_watchdog > > 71,2 =B5s/s 0,4 kWork disk_event= s_workfn > > 57,5 =B5s/s 0,4 Process kdeinit4: = kded4 [kdeinit] > > 68,3 =B5s/s 0,3 Process akonadiser= ver > > 34,6 =B5s/s 0,3 Process /usr/bin/v= irtuoso-t +foreground +configfile /tmp/virtuoso_fB2599.ini +wait > > 14,9 =B5s/s 0,3 Process /usr/bin/g= pg-agent --daemon --sh --write-env-file=3D/home/martin/.gnupg/gpg-agent= -info-me > > 39,4 =B5s/s 0,30 Process /usr/bin/n= epomukservicestub nepomukfileindexer > >=20 > >=20 > > PowerTOP v2.0 Overview Idle stats Frequency stats Device = stats Tunables =20 > >=20 > >=20 > > Usage Device name > > 2644 rpm Laptop fan > > 4,3% CPU use > > 100,0% Audio codec hwC0D3: Intel > > 100,0% Audio codec hwC0D0: Conexant > > 100,0% Audio codec hwC0D1: Conexant > > 20,9 pkts/s Network interface: eth0 (e1000e) > > 100,0% USB device: usb-device-8087-0024 > > 100,0% USB device: usb-device-8087-0024 > > 100,0% Display backlight > > 100,0% Display backlight > > 100,0% USB device: EHCI Host Controller > > 100,0% USB device: Biometric Coprocessor (UPEK) > > 100,0% USB device: Integrated Smart Card Reader = (Lenovo) > > 100,0% USB device: EHCI Host Controller > > 0,0 ops/s GPU > > 100,0% USB device: usb-device-17ef-100a > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family PCI Express Root Port 5 > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family PCI Express Root Port 4 > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family PCI Express Root Port 2 > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family High Definition Audio Controller > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family MEI Controller #1 > > 100,0% USB device: PS/2+USB Mouse > > 100,0% PCI Device: Intel Corporation QM67 Expres= s Chipset Family LPC Controller > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family 6 port SATA AHCI Controller > > 100,0% PCI Device: Ricoh Co Ltd MMC/SD Host Cont= roller > > 100,0% PCI Device: Silicon Image, Inc. SiI 3531 = [SATALink/SATARaid] Serial ATA Controller > > 100,0% PCI Device: Ricoh Co Ltd R5C832 PCIe IEEE= 1394 Controller > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family USB Enhanced Host Controller #2 > > 100,0% PCI Device: Intel Corporation 82579LM Gig= abit Network Connection > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family SMBus Controller > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family PCI Express Root Port 1 > > 100,0% PCI Device: Intel Corporation 6 Series/C2= 00 Series Chipset Family USB Enhanced Host Controller #1 > > 100,0% PCI Device: Intel Corporation 2nd Generat= ion Core Processor Family Integrated Graphics Controller > > 100,0% PCI Device: Intel Corporation 2nd Generat= ion Core Processor Family DRAM Controller > > 100,0% PCI Device: Intel Corporation Centrino Ad= vanced-N 6205 [Taylor Peak] > > 0,0 pkts/s Network interface: wlan0 (iwlwifi) > > 0,0% Thinkpad light > > 0,0% Radio device: thinkpad_acpi > > 0,0% Radio device: iwlwifi > >=20 > >=20 > > I saw nothing outstanding in dmesg or kern.log, just > >=20 > > merkaba:~#2> grep "Intel pstate controlling: cpu" /var/log/kern.log > > May 20 10:03:31 merkaba kernel: [264852.930994] Intel pstate contro= lling: cpu 1 > > May 20 10:03:31 merkaba kernel: [264852.944479] Intel pstate contro= lling: cpu 2 > > May 20 10:03:31 merkaba kernel: [264852.957879] Intel pstate contro= lling: cpu 3 > > May 21 09:14:03 merkaba kernel: [315800.710623] Intel pstate contro= lling: cpu 1 > > May 21 09:14:03 merkaba kernel: [315800.724087] Intel pstate contro= lling: cpu 2 > > May 21 09:14:03 merkaba kernel: [315800.737493] Intel pstate contro= lling: cpu 3 > > May 21 20:01:20 merkaba kernel: [344305.122543] Intel pstate contro= lling: cpu 1 > > May 21 20:01:20 merkaba kernel: [344305.135953] Intel pstate contro= lling: cpu 2 > > May 21 20:01:20 merkaba kernel: [344305.149398] Intel pstate contro= lling: cpu 3 > > May 21 21:52:18 merkaba kernel: [ 1.441043] Intel pstate control= ling: cpu 0 > > May 21 21:52:18 merkaba kernel: [ 1.441084] Intel pstate control= ling: cpu 1 > > May 21 21:52:18 merkaba kernel: [ 1.441123] Intel pstate control= ling: cpu 2 > > May 21 21:52:18 merkaba kernel: [ 1.441162] Intel pstate control= ling: cpu 3 > > May 23 17:22:29 merkaba kernel: [ 1.860224] Intel pstate control= ling: cpu 0 > > May 23 17:22:29 merkaba kernel: [ 1.860267] Intel pstate control= ling: cpu 1 > > May 23 17:22:29 merkaba kernel: [ 1.860306] Intel pstate control= ling: cpu 2 > > May 23 17:22:29 merkaba kernel: [ 1.860345] Intel pstate control= ling: cpu 3 > >=20 > > But these seem to be regular and I had them before as well. > >=20 > > Thanks, > >=20 >=20 --=20 Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7