#129215 "gpm: spews noise to /dev/input/mice"

Package:
gpm
Source:
gpm
Description:
General Purpose Mouse interface
Submitter:
Date:
2005-07-18 03:15:29 UTC
Severity:
normal
#129215#5
Date:
2002-01-14 17:22:24 UTC
From:
To:
After the XF86 server configuration is done and the XF86Config-4 file
written, there is (even if a mouse is correctly configured) a section
"InputDevice" with the identifier "Generic Mouse" IN ADDITION to the
"Configured Mouse" entry. This leads (at least on my system) to a problem
when I'm using gpm and X. The mouse works fine with gpm on any terminal until
X is started. After that (when I switch to another terminal) the mouse is
gone.

When I delete the "Generic Mouse" entry (actually the whole Section for the
is entry) everything works perfect. It is now possible to switch from X to a
terminal and back and the mouse works everywhere.


Possible solution for this problem:
The XF86Config-4 file should not include the "Generic Mouse" Section and the
correspondending InputDevice in the "Server Layout" should be deleted.


Nikolaus Regnat
#129215#12
Date:
2002-01-14 23:40:08 UTC
From:
To:
reassign 129215 gpm
retitle 129215 "gpm: spews noise to /dev/input/mice"
thanks

Yes, there is another mouse section IN ADDITION.  This is DELIBERATE.
It is so that people can TAKE ADVANTAGE of multiple pointers attached to
the system via the USB interface.

This is NOT AN ACCIDENT.  You might want to look into fixing your
keyboard; the caps lock key appears to have some sort of intermittent
problem.

This sounds like a bug in gpm.

I suggest dpkg-reconfiguring xserver-xfree86 to not manage the config
file with debconf if this bug cannot be fixed in gpm.

No, that is 100% incorrect and just makes life harder for users who want
their mice to work.
#129215#15
Date:
2002-01-15 15:29:52 UTC
From:
To:
Sorry for getting mad at you; I didn't realize you weren't trying to be
pedantic with me.  If you can reproduce this bug without gpm running,
then it is a bug in the X server.  You should make sure you are using
gpm as a repeater if you want to keep it running while the X server is
running.  The two programs have a long history of non-cooperation unless
special care is taken by the user.
#129215#20
Date:
2002-01-18 09:32:38 UTC
From:
To:
I require the actual gpm.conf and XF86Config-4 in question.

Barring this I will relabel this an obvious X bug with a known fix.

Zephaniah E. Hull.
#129215#25
Date:
2002-01-18 21:34:36 UTC
From:
To:
That's bullshit and you know it.  If your package corrupts the protocol
stream on a device node, that's its fault, not the X server's.