Re: Keyboard freezing/repeating endlessly
On Sat, 15 Jan 2011 20:16:16 +0000 (UTC)
Camaleón <email@example.com> wrote:
> On Sat, 15 Jan 2011 09:55:54 -0800, Joe Riel wrote:
> > On Fri, 14 Jan 2011 07:47:59 -0800 Joe Riel wrote:
> >> On Fri, 14 Jan 2011 11:38:43 +0000 (UTC) Camaleón wrote:
> >> > You can:
> >> >
> >> > 1/ Start the system with no xorg (console) and leave it so for
> >> > sometine to check if the problem is still present (even with no x
> >> > server).
> >> I'll attempt to try this. Might be painful working in a virtual
> >> terminal all the time. Kind of limits what I can do. This is my
> >> work machine...
> > Got lucky, sort of. I was just working in a virtual terminal when
> > the lockup occurred. That doesn't rule out X in that X was running
> > (I had just opened the terminal and was reading a man page),
> > however, it suggests the problem might not be with X. That is
> > consistent with the observation that restarting gdm3 does not clear
> > the fault.
> Hum... yes, then it can be something related to the kernel.
> >> > 2/ Review you kernel log, just in case ("cat /var/log/kern.log |
> >> > grep input").
> > Found the following lines (more of same) in kern.log at the time of
> > the lockup. However, I don't see them for other lockups (a lockup
> > occurred while I was typing this email), so it could be a
> > red-herring:
> > Jan 15 09:20:58 gauss kernel: [164858.674186] [drm] nouveau
> > 0000:01:00.0: nouveau_channel_free: freeing fifo 2 Jan 15 09:20:59
> > gauss kernel: [164859.749505] [drm] nouveau 0000:01:00.0:
> > nouveau_channel_free: freeing fifo 1 Jan 15 09:20:59 gauss kernel:
> > [164859.750491] [drm] nouveau 0000:01:00.0: PFIFO_CACHE_ERROR - Ch
> > 1/5 Mthd 0x0000 Data 0x80000013 Jan 15 09:20:59 gauss kernel:
> > [164859.750507] [drm] nouveau 0000:01:00.0: PGRAPH_ERROR - nSource:
> > DATA_ERROR, nStatus: BAD_ARGUMENT Jan 15 09:20:59 gauss kernel:
> > [164859.750511] [drm] nouveau 0000:01:00.0: PGRAPH_ERROR - Ch 1/5
> > Class 0x008a Mthd 0x0300 Data 0x00000000:0x00000000 Jan 15 09:20:59
> > gauss kernel: [164859.750566] [drm] nouveau 0000:01:00.0:
> > PFIFO_CACHE_ERROR - Ch 1/4 Mthd 0x0000 Data 0x8000001a Jan 15
> > 09:20:59 gauss kernel: [164859.750580] [drm] nouveau 0000:01:00.0:
> > PGRAPH_ERROR - nSource: ILLEGAL_MTHD, nStatus: BAD_ARGUMENT
> > PROTECTION_FAULT
> If X server is not running, where are these logs entries coming from?
> (nouveau is the X driver for nvidia cards) >;-)
Well, the X server was running. I was originally in X and switched to
a virtual terminal. The lockup occurred while I was working in the
virtual terminal. I searched google for that error message and found
a few related bug reports in other forums. At least one user was
experiencing very similar behavior (keyboard goes away).
> > Anyway, the behaviour you are experiencing with the ps/2 keyboard
> > is weird enough to fill a bug in Debian BTS, I would go for it.
I'll do so.