Huge wakeups number from i1915

Huge wakeups number from i1915

[Yves-Alexis Perez]

[-- Attachment #1: Type: text/plain, Size: 992 bytes --]

Hi, I have a problem with 2.6.28-rc6-00209-g6a12141. Not sure if I
should mail to acpi@ or drm.

From time to time (I noticed this resuming from hibernation, but it just
appeared without that), I get huge wakeups from i915 module, and huge
power consumption on my Thinkpad T61 with intel 965:

corsac@hidalgo: sudo powertop -d
PowerTOP 1.10    (C) 2007, 2008 Intel Corporation 

Collecting data for 15 seconds 


Cn	          Avg residency
C0 (cpu running)        (60,9%)
C0		  0,0ms ( 0,0%)
C1		  0,0ms ( 0,0%)
C2		  0,0ms (39,1%)
C4		  0,0ms ( 0,0%)
P-states (frequencies)
Wakeups-from-idle per second : 41769,4	interval: 15,0s
Power usage (ACPI estimate): 33,1W (0,9 hours) 
Top causes for wakeups:
  99,8% (85712,2)       <interrupt> : i915@pci:0000:00:02.0 


Not sure what I can do about this but it's a bit painful when running on
battery :/

If I need to provide more info or ask on another list, please say so :)

Cheers and thanks,
-- 
Yves-Alexis

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: Huge wakeups number from i1915

[Matthew Garrett]

On Tue, Dec 02, 2008 at 05:48:21PM +0100, Yves-Alexis Perez wrote:
> Hi, I have a problem with 2.6.28-rc6-00209-g6a12141. Not sure if I
> should mail to acpi@ or drm.

The interrupt code on i915 has been heavily rewritten. This sounds like 
an interrupt that isn't being acked correctly, so the drm list is 
probably your best bet.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: Huge wakeups number from i1915

[Yves-Alexis Perez]

[-- Attachment #1: Type: text/plain, Size: 449 bytes --]

On mar, 2008-12-02 at 17:11 +0000, Matthew Garrett wrote:
> On Tue, Dec 02, 2008 at 05:48:21PM +0100, Yves-Alexis Perez wrote:
> > Hi, I have a problem with 2.6.28-rc6-00209-g6a12141. Not sure if I
> > should mail to acpi@ or drm.
> 
> The interrupt code on i915 has been heavily rewritten. This sounds
> like 
> an interrupt that isn't being acked correctly, so the drm list is 
> probably your best bet.

done, thanks
-- 
Yves-Alexis

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 197 bytes --]