T400 sluggish after resume from sleep

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

T400 sluggish after resume from sleep

Joachim Breitner-2
Hi,

I am observing a strange pattern here. After I suspend to and resume
from sleep, the system is generally sluggish. It is not always noticable
when typing, but sometime already observable when using the mouse. It is
definitely noticeable when playing, say, frozen-bubble.

CPU usage does not indicate anything running in the background. My blind
guess is that some hardware is misbehaving after the resume and causing
lots of interrupts.

Does anyone else observe similar issues or have a hint what the cause
might be?

Thanks,
Joachim

--
Joachim "nomeata" Breitner
  [hidden email]  |  [hidden email]  |  GPG: 0x4743206C
  xmpp: [hidden email] | http://www.joachim-breitner.de/


signature.asc (205 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: T400 sluggish after resume from sleep

Philipp Kern-6
On Thu, Aug 25, 2011 at 07:29:42PM +0200, Joachim Breitner wrote:
> Does anyone else observe similar issues or have a hint what the cause
> might be?

If it's an interrupt flood, you should see a higher rate in /proc/interrupts
than under normal use?

Kind regards,
Philipp Kern
--
 .''`.  Philipp Kern                        Debian Developer
: :' :  http://philkern.de                         Stable Release Manager
`. `'   xmpp:[hidden email]                         Wanna-Build Admin
  `-    finger pkern/[hidden email]

signature.asc (205 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: T400 sluggish after resume from sleep

Joachim Breitner-2
Hi,

Am Donnerstag, den 25.08.2011, 22:04 +0200 schrieb Philipp Kern:
> On Thu, Aug 25, 2011 at 07:29:42PM +0200, Joachim Breitner wrote:
> > Does anyone else observe similar issues or have a hint what the cause
> > might be?
>
> If it's an interrupt flood, you should see a higher rate in /proc/interrupts
> than under normal use?

thanks for the hint. I’m not sure if my measurements are very accurate,
but I do see some difference. RES seems to be higher, for example, as
well as interrupt 45 (ok, that is eth0, that might just an artefact of
when I am measuring the interrupts).


Attached my script; not very good, but may be useful for others
nevertheless.

Greetings,
Joachim


--
Joachim "nomeata" Breitner
  [hidden email]  |  [hidden email]  |  GPG: 0x4743206C
  xmpp: [hidden email] | http://www.joachim-breitner.de/


itrrate.pl (684 bytes) Download Attachment
signature.asc (205 bytes) Download Attachment