Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Thu, Feb 25, 2010 at 06:16:51PM +, Tom Parker wrote: On 25 February 2010 18:15, Moritz Muehlenhoff j...@inutil.org wrote: Did you report this at http://bugzilla.kernel.org ? Haven't yet. Somewhere on the todo list... What's the status? Cheers, Moritz -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100501201129.ga17...@galadriel.inutil.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On 25 February 2010 18:15, Moritz Muehlenhoff j...@inutil.org wrote: Did you report this at http://bugzilla.kernel.org ? Haven't yet. Somewhere on the todo list... Tom -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/243a9cc1002251016ya4a1e22oa2b84d229aab0...@mail.gmail.com
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Tue, Feb 16, 2010 at 04:46:31AM +, Ben Hutchings wrote:
On Sat, 2010-02-06 at 09:31 +, Tom Parker wrote:
On 2 February 2010 21:16, maximilian attems m...@stro.at wrote:
Squeeze will release with 2.6.32 can anyone of you still
reproduce this?
also please make sure to use latest powertop, aka
ii powertop 1.13~pre201001 Linux tool to find out what is using
power
palf...@drone:[~] dpkg -l |grep powertop
ii powertop 1.13~pre20100125-1
Linux tool to find out what is using power
o
palf...@drone:[~] uname -a
Linux drone 2.6.32-1-686 #1 SMP Mon Feb 1 01:37:26 UTC 2010 i686 GNU/Linux
Attached is powertop.log ('powertop -d') and top.log('top -b -n 1')
As you can see, the major wakeup is still Load balancing tick, which
I'm guessing is a translated name for hrtimer_start_expires? I think
the ACPI estimate of 1.1W is a little off though...
hrtimer_start_expires() is a generic function in the kernel to schedule
a wakeup by the high-resolution timer. In 2.6.32 it is an inline
function that calls hrtimer_start_range_ns(). Powertop recognises and
relabels some specific calling sequences, including this one.
If I understand correctly, these are scheduler ticks that periodically
interrupt the running task (preemption). However, on a kernel compiled
with the NOHZ option (as Debian's kernel images are) this should never
wake the system up - scheduler ticks are disabled when there are no
tasks ready to run and no other timers due to expire before the next
tick.
I can only suggest you report this upstream at
http://bugzilla.kernel.org. Use product 'Process Management',
component 'Scheduler'. Let us know the bug number so we can track it.
Did you report this at http://bugzilla.kernel.org ?
Cheers,
Moritz
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100225181542.ga5...@galadriel.inutil.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Sat, 2010-02-06 at 09:31 +, Tom Parker wrote:
On 2 February 2010 21:16, maximilian attems m...@stro.at wrote:
Squeeze will release with 2.6.32 can anyone of you still
reproduce this?
also please make sure to use latest powertop, aka
ii powertop 1.13~pre201001 Linux tool to find out what is using power
palf...@drone:[~] dpkg -l |grep powertop
ii powertop 1.13~pre20100125-1
Linux tool to find out what is using power
o
palf...@drone:[~] uname -a
Linux drone 2.6.32-1-686 #1 SMP Mon Feb 1 01:37:26 UTC 2010 i686 GNU/Linux
Attached is powertop.log ('powertop -d') and top.log('top -b -n 1')
As you can see, the major wakeup is still Load balancing tick, which
I'm guessing is a translated name for hrtimer_start_expires? I think
the ACPI estimate of 1.1W is a little off though...
hrtimer_start_expires() is a generic function in the kernel to schedule
a wakeup by the high-resolution timer. In 2.6.32 it is an inline
function that calls hrtimer_start_range_ns(). Powertop recognises and
relabels some specific calling sequences, including this one.
If I understand correctly, these are scheduler ticks that periodically
interrupt the running task (preemption). However, on a kernel compiled
with the NOHZ option (as Debian's kernel images are) this should never
wake the system up - scheduler ticks are disabled when there are no
tasks ready to run and no other timers due to expire before the next
tick.
I can only suggest you report this upstream at
http://bugzilla.kernel.org. Use product 'Process Management',
component 'Scheduler'. Let us know the bug number so we can track it.
So, doesn't look fixed. I'm currently at FOSDEM if anyone with more
knowledge wants to borrow my laptop to do more debugging on this.
Sorry I didn't have the time to do this. This is not really my area of
expertise though.
Ben.
--
Ben Hutchings
Humour is the best antidote to reality.
signature.asc
Description: This is a digitally signed message part
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On 2 February 2010 21:16, maximilian attems m...@stro.at wrote:
Squeeze will release with 2.6.32 can anyone of you still
reproduce this?
also please make sure to use latest powertop, aka
ii powertop 1.13~pre201001 Linux tool to find out what is using power
palf...@drone:[~] dpkg -l |grep powertop
ii powertop 1.13~pre20100125-1
Linux tool to find out what is using power
o
palf...@drone:[~] uname -a
Linux drone 2.6.32-1-686 #1 SMP Mon Feb 1 01:37:26 UTC 2010 i686 GNU/Linux
Attached is powertop.log ('powertop -d') and top.log('top -b -n 1')
As you can see, the major wakeup is still Load balancing tick, which
I'm guessing is a translated name for hrtimer_start_expires? I think
the ACPI estimate of 1.1W is a little off though...
So, doesn't look fixed. I'm currently at FOSDEM if anyone with more
knowledge wants to borrow my laptop to do more debugging on this.
Tom
PowerTOP 1.12 (C) 2007, 2008 Intel Corporation
Collecting data for 15 seconds
Cn Avg residency
C0 (cpu running)( 6.1%)
polling 0.0ms ( 0.0%)
C1 halt 0.0ms ( 0.0%)
C2 7.4ms (20.7%)
C3 5.2ms (73.2%)
P-states (frequencies)
1.60 Ghz 4.2%
1333 Mhz 0.6%
1066 Mhz 0.1%
800 Mhz95.0%
Wakeups-from-idle per second : 168.0 interval: 15.0s
Power usage (ACPI estimate): 1.1W (6.5 hours)
Top causes for wakeups:
46.8% ( 85.1) [kernel scheduler] Load balancing tick
23.4% ( 42.7) [uhci_hcd:usb4, ath] interrupt
7.7% ( 13.9) avahi-daemon
7.3% ( 13.3) chromium-browse
3.6% ( 6.6) [ata_piix] interrupt
1.7% ( 3.1) Xorg
1.5% ( 2.7) [kernel core] hrtimer_start (tick_sched_timer)
1.2% ( 2.1) [acpi] interrupt
1.2% ( 2.1) [ehci_hcd:usb1, uhci_hcd:usb2, HDA Intel, i...@pci::00:02.0] interrupt
1.1% ( 2.0) multiload-apple
0.9% ( 1.6) gnome-terminal
0.7% ( 1.2) nautilus
0.5% ( 1.0) ntpd
0.5% ( 1.0) cpufreq-applet
0.5% ( 0.9) lighttpd
0.2% ( 0.4) NetworkManager
0.1% ( 0.3) PS/2 keyboard/mouse/touchpad interrupt
0.1% ( 0.3) gnome-settings-
0.1% ( 0.3) [kernel core] enqueue_task_rt (sched_rt_period_timer)
0.1% ( 0.2) sleep
0.1% ( 0.2) gnome-panel
0.1% ( 0.2) init
0.1% ( 0.1) mono
0.1% ( 0.1) ssh-agent
0.1% ( 0.1) gnome-power-man
0.1% ( 0.1) hald
0.0% ( 0.1) metacity
0.0% ( 0.1) [kernel core] mod_timer (fib6_gc_timer_cb)
0.0% ( 0.1) rsyslogd
0.0% ( 0.1) kblockd/1
0.0% ( 0.1) [kernel core] mod_timer (inet_twdr_hangman)
Suggestion: Enable wireless power saving mode by executing the following command:
iwconfig wlan0 power timeout 500ms
This will sacrifice network performance slightly to save power.
Recent USB suspend statistics
Active Device name
0.0% USB device 1-5 : USB 2.0 Camera (Sonix Technology Co., Ltd.)
0.0% USB device usb5 : UHCI Host Controller (Linux 2.6.32-1-686 uhci_hcd)
0.0% USB device usb4 : UHCI Host Controller (Linux 2.6.32-1-686 uhci_hcd)
0.0% USB device usb3 : UHCI Host Controller (Linux 2.6.32-1-686 uhci_hcd)
0.0% USB device usb2 : UHCI Host Controller (Linux 2.6.32-1-686 uhci_hcd)
0.0% USB device usb1 : EHCI Host Controller (Linux 2.6.32-1-686 ehci_hcd)
Recent audio activity statistics
Active Device name
Recent SATA AHCI link activity statistics
Active Partial Slumber Device name
top - 09:24:47 up 5 min, 2 users, load average: 0.31, 0.50, 0.27
Tasks: 132 total, 1 running, 131 sleeping, 0 stopped, 0 zombie
Cpu(s): 12.4%us, 6.3%sy, 0.0%ni, 68.3%id, 12.3%wa, 0.1%hi, 0.4%si, 0.0%st
Mem: 1022564k total, 412120k used, 610444k free,21788k buffers
Swap: 1052248k total,0k used, 1052248k free, 236704k cached
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
2354 root 20 0 310m 23m 19m S 11 2.4 0:16.70 Xorg
2562 palfrey 20 0 21588 10m 8780 S4 1.0 0:01.61 metacity
2589 palfrey 20 0 45736 17m 13m S4 1.8 0:04.65 gnome-panel
3666 palfrey 20 0 2444 1060 792 R4 0.1 0:00.04 top
2736 palfrey 20 0 23432 8532 7240 S2 0.8 0:02.18 multiload-apple
3122 palfrey 20 0 226m 48m 22m S2 4.9 0:21.33 chromium-browse
1 root 20 0 2040 712 616 S0 0.1 0:01.12 init
2 root 20 0 000 S0 0.0 0:00.00 kthreadd
3 root RT 0 000 S0 0.0 0:00.00 migration/0
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Tue, Feb 2, 2010 at 11:16 PM, maximilian attems m...@stro.at wrote: currently this bug looks like a bug that has been fixed since. Seconded. I was following unstable all this time, and this bug didn't resurface since at least 2.6.31, and I definitely don't see it now. ii linux-image-2.6.32-trunk-686 2.6.32-5 Linux 2.6.32 for modern PCs ii powertop 1.13~pre20100125-1 Linux tool to find out what is using power on a laptop -- Dmitry Borodaenko -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Wed, Feb 03, 2010 at 04:37:48AM +, Ben Hutchings wrote: On Wed, 2010-02-03 at 00:13 -0200, Renato S. Yamane wrote: Em 02-02-2010 19:16, maximilian attems escreveu: Squeeze will release with 2.6.32 can anyone of you still reproduce this? also please make sure to use latest powertop, aka ii powertop 1.13~pre201001 Linux tool to find out what is using power $ uname -a Linux mandachuva 2.6.30-2-686 #1 SMP Fri Dec 4 00:53:20 UTC 2009 i686 GNU/Linux Where is Linux-Kernel 2.6.32? It's in unstable and testing (but not backports yet). $ dpkg -l | grep powertop ii powertop 1.11-1 And yes, this bug is still present. Powertop show me 210W power used (maybe ten times higher??) and my This figure is not believable. battery goes from 96% to 82% in only 17min (I use a 9 cell battery, wich be enough by ~4h, but using Debian Lenny it's fully eated in only ~2h). Please send the output of 'top -b -n 1' and 'powertop -d'. this powertop is outdated, you need to use the updated powertop, which is in squeeze or the sid. the old one will provide bogus values on current linux images. as we already got an expected report that this bug is fixed in 2.6.32, will wait a day or so before closing. -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Processed: Re: Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
Processing commands for cont...@bugs.debian.org: tags 521944 moreinfo Bug #521944 [linux-2.6] linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires Added tag(s) moreinfo. stop Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
tags 521944 moreinfo stop hello, Squeeze will release with 2.6.32 can anyone of you still reproduce this? also please make sure to use latest powertop, aka ii powertop 1.13~pre201001 Linux tool to find out what is using power currently this bug looks like a bug that has been fixed since. thanks for your feedback maks -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
Em 02-02-2010 19:16, maximilian attems escreveu: Squeeze will release with 2.6.32 can anyone of you still reproduce this? also please make sure to use latest powertop, aka ii powertop 1.13~pre201001 Linux tool to find out what is using power $ uname -a Linux mandachuva 2.6.30-2-686 #1 SMP Fri Dec 4 00:53:20 UTC 2009 i686 GNU/Linux Where is Linux-Kernel 2.6.32? $ dpkg -l | grep powertop ii powertop 1.11-1 And yes, this bug is still present. Powertop show me 210W power used (maybe ten times higher??) and my battery goes from 96% to 82% in only 17min (I use a 9 cell battery, wich be enough by ~4h, but using Debian Lenny it's fully eated in only ~2h). -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
On Wed, 2010-02-03 at 00:13 -0200, Renato S. Yamane wrote: Em 02-02-2010 19:16, maximilian attems escreveu: Squeeze will release with 2.6.32 can anyone of you still reproduce this? also please make sure to use latest powertop, aka ii powertop 1.13~pre201001 Linux tool to find out what is using power $ uname -a Linux mandachuva 2.6.30-2-686 #1 SMP Fri Dec 4 00:53:20 UTC 2009 i686 GNU/Linux Where is Linux-Kernel 2.6.32? It's in unstable and testing (but not backports yet). $ dpkg -l | grep powertop ii powertop 1.11-1 And yes, this bug is still present. Powertop show me 210W power used (maybe ten times higher??) and my This figure is not believable. battery goes from 96% to 82% in only 17min (I use a 9 cell battery, wich be enough by ~4h, but using Debian Lenny it's fully eated in only ~2h). Please send the output of 'top -b -n 1' and 'powertop -d'. Ben. -- Ben Hutchings I'm not a reverse psychological virus. Please don't copy me into your sig. signature.asc Description: This is a digitally signed message part
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
So sorry, but someone taking a looking here? This bug make me lost an important time, because battery is drained very faster. Regards, Renato S. Yamane -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
I have seem this bug on my Lenovo 3000-V200 laptop too! Battery is drained very fast! With Lenny, my battery go to 10% after ~3,5h but with Squeeze the battery go to 10% after 2,5h. And powertop show me a strange value: Uso de energia (estimado por ACPI): 191,6 W (0,2 horas) 191,6W?? $ dpkg -l | grep linux-image ii linux-image-2.6-686 2.6.30+20 ii linux-image-2.6.30-1-686 2.6.30-6 -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#521944: linux-image-2.6.29-1-686: Excessive number of interrupts from hrtimer_start_expires
Package: linux-image-2.6.29-1-686 Version: 2.6.29-1 Severity: normal Running powertop on the 2.6.29-1 kernel gets me the following Top causes for wakeups: 50.8% (126.3) kernel core : hrtimer_start_expires (tick_sched_timer) 12.8% ( 31.7) kernel IPI : Rescheduling interrupts 7.4% ( 18.5) interrupt : uhci_hcd:usb3, ath 5.3% ( 13.1) interrupt : ata_piix 4.4% ( 10.9) interrupt : extra timer interrupt Previously, the major cause of interrupts had been the kernel IPI, and hence one reason for upgrading to the latest kernel version. Now there's hrtimer_start_expires, and quite frankly this is a median case - I've seen it hit as high as 500 interrupts/second occasionally. This is causing my battery to get drained at an accelerated rate and I don't think this is correct behaviour. -- Package-specific info: ** Version: Linux version 2.6.29-1-686 (Debian 2.6.29-1) (wa...@debian.org) (gcc version 4.3.3 (Debian 4.3.3-5) ) #1 SMP Tue Mar 24 23:05:52 UTC 2009 ** Command line: BOOT_IMAGE=/boot/vmlinuz-2.6.29-1-686 root=UUID=1d2ed566-a39a-423f-8a96-59206e41336e ro quiet ** Not tainted ** Kernel log: [ 913.817857] pci :04:00.3: restoring config space at offset 0x1 (was 0x10, writing 0x17) [ 913.817857] pci :04:00.4: restoring config space at offset 0xf (was 0x1ff, writing 0x10b) [ 913.817857] pci :04:00.4: restoring config space at offset 0x4 (was 0x0, writing 0x5410) [ 913.817857] pci :04:00.4: restoring config space at offset 0x1 (was 0x10, writing 0x17) [ 913.820542] Enabling non-boot CPUs ... [ 913.821353] SMP alternatives: switching to SMP code [ 914.363428] Booting processor 1 APIC 0x1 ip 0x6000 [ 913.816617] Initializing CPU#1 [ 913.816617] Calibrating delay using timer specific routine.. 3192.03 BogoMIPS (lpj=6384069) [ 913.816617] CPU: L1 I cache: 32K, L1 D cache: 24K [ 913.816617] CPU: L2 cache: 512K [ 913.816617] [ds] using Core 2/Atom configuration [ 913.816617] CPU: Physical Processor ID: 0 [ 913.816617] CPU: Processor Core ID: 0 [ 913.816617] Intel machine check architecture supported. [ 913.816617] Intel machine check reporting enabled on CPU#1. [ 914.452394] CPU1: Intel(R) Atom(TM) CPU N270 @ 1.60GHz stepping 02 [ 914.452572] CPU0 attaching NULL sched-domain. [ 914.453038] Switched to high resolution mode on CPU 1 [ 914.464581] CPU0 attaching sched-domain: [ 914.464587] domain 0: span 0-1 level SIBLING [ 914.464591] groups: 0 1 [ 914.464600] CPU1 attaching sched-domain: [ 914.464604] domain 0: span 0-1 level SIBLING [ 914.464607] groups: 1 0 [ 914.469032] CPU1 is up [ 914.469037] ACPI: Waking up from system sleep state S3 [ 914.500538] pci :00:02.0: PME# disabled [ 914.500560] pci :00:02.0: PCI INT A - GSI 16 (level, low) - IRQ 16 [ 914.500571] pci :00:02.0: setting latency timer to 64 [ 914.500581] pci :00:02.1: PME# disabled [ 914.500659] HDA Intel :00:1b.0: PCI INT A - GSI 16 (level, low) - IRQ 16 [ 914.500671] HDA Intel :00:1b.0: setting latency timer to 64 [ 914.500712] pcieport-driver :00:1c.0: setting latency timer to 64 [ 914.500728] pcieport-driver :00:1c.1: setting latency timer to 64 [ 914.500744] pcieport-driver :00:1c.2: setting latency timer to 64 [ 914.500759] pcieport-driver :00:1c.3: setting latency timer to 64 [ 914.500813] uhci_hcd :00:1d.0: PCI INT A - GSI 16 (level, low) - IRQ 16 [ 914.500825] uhci_hcd :00:1d.0: setting latency timer to 64 [ 914.500859] usb usb1: root hub lost power or was reset [ 914.500919] uhci_hcd :00:1d.1: PCI INT B - GSI 17 (level, low) - IRQ 17 [ 914.500932] uhci_hcd :00:1d.1: setting latency timer to 64 [ 914.500964] usb usb2: root hub lost power or was reset [ 914.501047] uhci_hcd :00:1d.2: PCI INT C - GSI 18 (level, low) - IRQ 18 [ 914.501063] uhci_hcd :00:1d.2: setting latency timer to 64 [ 914.501104] usb usb3: root hub lost power or was reset [ 914.501165] uhci_hcd :00:1d.3: PCI INT D - GSI 19 (level, low) - IRQ 19 [ 914.501177] uhci_hcd :00:1d.3: setting latency timer to 64 [ 914.501210] usb usb4: root hub lost power or was reset [ 914.501288] ehci_hcd :00:1d.7: PME# disabled [ 914.501299] ehci_hcd :00:1d.7: PCI INT A - GSI 16 (level, low) - IRQ 16 [ 914.501311] ehci_hcd :00:1d.7: setting latency timer to 64 [ 914.501324] ehci_hcd :00:1d.7: PME# disabled [ 914.501368] pci :00:1e.0: setting latency timer to 64 [ 914.501432] ata_piix :00:1f.2: PCI INT B - GSI 17 (level, low) - IRQ 17 [ 914.501443] ata_piix :00:1f.2: setting latency timer to 64 [ 914.501651] r8169 :02:00.0: PME# disabled [ 914.501843] ath5k :03:00.0: PCI INT A - GSI 18 (level, low) - IRQ 18 [ 914.529635] sdhci-pci :04:00.0: restoring config space at offset 0xc (was 0x0, writing 0x8000) [ 914.529729] sdhci-pci :04:00.0: PCI INT A - GSI 19 (level, low) - IRQ 19 [ 914.529756] sdhci-pci :04:00.0: setting latency timer to 64
