hel[ tracking down two pesky problems

Kyle O'Donnell kyleodonnell-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org
Mon Sep 25 18:05:23 UTC 2006


Hi Matt,

I've found the same problem with /dev/dsp and have narrowed it down to
either firefox and/or mplayerplug-in.

--kyleo

On 9/25/06, Matt Price <matt.price-H217xnMUJC0sA/PxXw9srA at public.gmane.org> wrote:
> Hey,
>
> I run Ubuntu Dapper on my Dell Latitude D820, pretty much a vanilla
> distro except that I use the suspend2-enabled kernels which can be found
> at various places on the net.
>
> Mostly it works great, but I have a couple of irritating problems with
> I'm now ready to try and deal with.  they are:
>
> 1) strange screen flickering when screensaver turns on.  When the
> screensaver activates, there are 5-10 seconds of flashing light, suring
> which the desktop is intermittently visible and can sometimes be seen in
> inverted colors (white baps to black, and various other colors undergo
> unsettling trnasformations).  About half the time, these colors stay
> inverted when I deactivate the screensaver.
>
> I'm not sure what exactly this bug is related to.  It occurs when I run
> either the free nv or the nonfree nvidia drivers, and whether I choose a
> blank screen or any of the specificable gnome screensavers.  Any ideas
> what kinds of tests I might run to diagnose the problem?
>
> 2) some process grabs /dev/dsp when I suspend/resume.  Often (but not
> always) when I resume, /dev/dsp becomes unavailable.  At first I thought
> this was a driver problem, because there are several bugs in the
> intel_hda driver that my sound card uses -- it doesn't like to switch
> back and forth between speakers and headphones, and apparently hte
> microphone doesn't work that well (haven't tried it).  However, I have
> discovered that access to /dev/dsp is usallky resotred if I log out and
> log back in again.  SO I assume that some process is
> monopolizing /dev/dsp and is hung.  Logging out and in kills the
> process, so /dev/dsp is available.  Since ubuntu uses esd by default,
> that would be the obvious candidate; but theproblem occurs even when no
> instance of esd is running.
>
> Again, I don't know how to go about diagnosing this.  ANy suggestions?
> Is there some way to check whether a device is being accessed by a
> process?
>
> Thanks as usual,
>
> matt
>
>
>
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists





More information about the Legacy mailing list