[gm45] i915 causes huge amount of wakeups in powertop in karmic

[Slavius]

Binary package hint: xserver-xorg-video-intel

running fresh installed and updated:
Description: Ubuntu 9.10
Release: 9.10

package details:
xserver-xorg-video-intel:
  Installed: 2:2.9.0-1ubuntu2
  Candidate: 2:2.9.0-1ubuntu2
  Version table:
 *** 2:2.9.0-1ubuntu2 0
        500 http://archive.ubuntu.com karmic/main Packages
        100 /var/lib/dpkg/status

hardware:
laptop HP ProBook 4510s

lspci:
00:00.0 Host bridge [0600]: Intel Corporation Mobile 4 Series Chipset Memory Controller Hub [8086:2a40] (rev 07)
00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07)
00:02.1 Display controller [0380]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a43] (rev 07)
00:1a.0 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4 [8086:2937] (rev 03)
00:1a.1 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5 [8086:2938] (rev 03)
00:1a.2 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #6 [8086:2939] (rev 03)
00:1a.7 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2 [8086:293c] (rev 03)
00:1b.0 Audio device [0403]: Intel Corporation 82801I (ICH9 Family) HD Audio Controller [8086:293e] (rev 03)
00:1c.0 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 1 [8086:2940] (rev 03)
00:1c.1 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 2 [8086:2942] (rev 03)
00:1c.4 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 5 [8086:2948] (rev 03)
00:1c.5 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 6 [8086:294a] (rev 03)
00:1d.0 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1 [8086:2934] (rev 03)
00:1d.1 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2 [8086:2935] (rev 03)
00:1d.2 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3 [8086:2936] (rev 03)
00:1d.7 USB Controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #1 [8086:293a] (rev 03)
00:1e.0 PCI bridge [0604]: Intel Corporation 82801 Mobile PCI Bridge [8086:2448] (rev 93)
00:1f.0 ISA bridge [0601]: Intel Corporation ICH9M LPC Interface Controller [8086:2919] (rev 03)
00:1f.2 SATA controller [0106]: Intel Corporation ICH9M/M-E SATA AHCI Controller [8086:2929] (rev 03)
02:00.0 Network controller [0280]: Broadcom Corporation BCM4312 802.11b/g [14e4:4315] (rev 01)
44:00.0 Ethernet controller [0200]: Marvell Technology Group Ltd. Device [11ab:436c] (rev 10)

problem description:
i915 causes huge amount of wakeups and keeps my cpu in c0 state most of the time making the laptop battery die in less than 2 hours (in contrast to >5 hours in windows on the same machine) while running only gnome-terminal with powertop open on the desktop. I've also disabled secondary core to prevent "Rescheduling interrupts - kernel IPI" appear in powertop.

powertop 1.11:
Cn Avg residency P-states (frequencies)
C0 (cpu running) (75.8%)
C0 0.0ms ( 0.0%)
C1 mwait 0.0ms ( 0.0%)
C3 mwait 1.5ms (24.2%)

Wakeups-from-idle per second : 161.8 interval: 10.0s
Power usage (ACPI estimate): 19.0W (0.6 hours) (long term: 18.2W,/0.7h)

Top causes for wakeups:
  68.7% (528.1) <interrupt> : i915@pci:0000:00:02.0
  11.5% ( 88.5) <kernel core> : hrtimer_start_range_ns (tick_sched_timer)
  11.1% ( 85.6) <interrupt> : PS/2 keyboard/mouse/touchpad
   2.5% ( 19.6) <interrupt> : uhci_hcd:usb4, eth2, HDA Intel
   1.5% ( 11.7) Xorg : hrtimer_start_range_ns (hrtimer_wakeup)
   1.2% ( 9.6) firefox : hrtimer_start_range_ns (hrtimer_wakeup)
   0.8% ( 6.0) <kernel core> : add_timer (wl_timer)
   0.7% ( 5.0) gnome-do : hrtimer_start_range_ns (hrtimer_wakeup)
   0.6% ( 4.9) syndaemon : hrtimer_start_range_ns (hrtimer_wakeup)
   0.3% ( 2.2) <interrupt> : ahci
   0.2% ( 1.5) Xorg : hrtimer_start (it_real_fn)
   0.1% ( 1.1) gnome-panel : hrtimer_start_range_ns (hrtimer_wakeup)
   0.1% ( 1.0) ntpd : hrtimer_start_range_ns (it_real_fn)
   0.1% ( 0.9) Xorg : queue_delayed_work (delayed_work_timer_fn)
   0.1% ( 0.7) <kernel core> : hrtimer_start (tick_sched_timer)
   0.1% ( 0.5) init : neigh_periodic_timer (neigh_periodic_timer)

[Slavius (slavomir-danas)]

Just finished installing Arch Linux. No issue here:

Cn Avg residency P-states (frequencies)
C0 (cpu running) ( 0.3%)
polling 0.0ms ( 0.0%)
C1 mwait 0.0ms ( 0.0%)
C3 mwait 38.4ms (99.7%)

Wakeups-from-idle per second : 26.0 interval: 20.0s
Power usage (5 minute ACPI estimate) : 0.2 W (263.2 hours left)

Top causes for wakeups:
  27.5% ( 7.7) <interrupt> : acpi
  18.8% ( 5.2) <kernel IPI> : Rescheduling interrupts
  13.2% ( 3.7) <interrupt> : extra timer interrupt
   7.0% ( 1.9) <interrupt> : ahci
   5.9% ( 1.6) gnome-terminal : schedule_hrtimeout_range (hrtimer_wakeup)
   4.3% ( 1.2) clock-applet : schedule_hrtimeout_range (hrtimer_wakeup)
   3.6% ( 1.0) <interrupt> : i915@pci:0000:00:02.0
   3.6% ( 1.0) <kernel core> : run_timer_softirq (sky2_watchdog)
   3.6% ( 1.0) ntpd : do_setitimer (it_real_fn)
   3.4% ( 0.9) Xorg : i915_add_request (delayed_work_timer_fn)
   1.6% ( 0.5) <kernel IPI> : TLB shootdowns
   1.4% ( 0.4) <kernel core> : neigh_periodic_timer (neigh_periodic_timer)
   0.9% ( 0.2) chrome : sys_epoll_wait (process_timeout)
   0.7% ( 0.2) <kernel core> : dev_watchdog (dev_watchdog)
   0.7% ( 0.2) gnome-panel : schedule_hrtimeout_range (hrtimer_wakeup)
   0.7% ( 0.2) nautilus : schedule_hrtimeout_range (hrtimer_wakeup)

package info:
xf86-video-intel 2.9.1-1
xorg-server 1.7.3-1

Linux slavius-laptop 2.6.31-ARCH #1 SMP PREEMPT Tue Nov 10 19:01:40 CET 2009 x86_64 Celeron(R) Dual-Core CPU T3000 @ 1.80GHz GenuineIntel GNU/Linux

[Slavius (slavomir-danas)]

I managed to simulate the same behaviour also in Arch. It happened after I've installed gnome-extra stuff. I also noticed that udev is eating like 5-15% of CPU. I run the `udevadm monitor` and saw this:
KERNEL[1259129717.200097] change /devices/pci0000:00/0000:00:02.0/drm/card0 (drm)
UDEV [1259129717.209746] change /devices/pci0000:00/0000:00:02.0/drm/card0 (drm)
KERNEL[1259129717.218808] change /devices/pci0000:00/0000:00:02.0/drm/card0 (drm)
UDEV [1259129717.225593] change /devices/pci0000:00/0000:00:02.0/drm/card0 (drm)
...
repeating fast.

After some googling I found a workaround by using Option "NoDRI" in Device section of xorg.conf, however that's not a solution.
Source: http://www.lesswatts.org/projects/powertop/known.php#intelgfx

[Slavius (slavomir-danas)]

This bug is present also in Lucid Alpha1. Using Option "NoDRI" in xorg.conf does not help.
Still 341 wake-ups from [i915] interrupt

[Slavius (slavomir-danas)]

I've marked this bug as duplicate of 440411.