Mouse error

eklirmr at wanadoo.fr eklirmr at wanadoo.fr
Wed Feb 18 05:58:19 PST 2004


On Tue Feb 17 2004 at 18:14:13 (-0600), Kevin Alm wrote:
> >
> > Dagmar d'Surreal <dagmar.wants at nospam.com> wrote:On Mon, 2004-02-16 at 
> 14:24, eklirmr at wanadoo.fr wrote:
> > > In the last few days I've built LFS 5.0 on two different PCs, a Dell
> > > Dimension H266 and a more recent Dell Dimension D4100. There were no
> > > problems on either machine. Now I'm installing some libraries and
> > > utilities, following the BLFS book; no problems on the H266, and just
> > > one small problem on the D4100.
> > >
> > > I've installed gpm on both machines. On the D4100, when gpm runs at
> > > startup, I get the following:
> > >
> > > error [mice.c (18770]: imps2: PS/2 mouse failed to read id, assuming
> > > standard PS/2.
> > >
> > > The mouse behaves very strangely and so does the screen. But if I
> > > then do ./gpm restart, there are no error messages, and everything
> > > works fine.
> > >
> > > Both PCs do indeed have a PS/2 mouse, and I've altered the boot
> > > scripts to echo the values of $MDEVICE and $PROTOCOL to be sure that
> > > they're passed correctly. This doesn't happen on the H266, but it
> > > happens every time on the D4100, and it also happens when I install
> > > gpm on the old LFS 3.0 system on the D4100.
> > >
> > > >From looking in Google and the gpm man page, I've seen that there can
> > >
> > > be a conflict between gpm and X, but I haven't installed X on the
> > > D4100 yet.
> > >
> > > OK, it's a minor problem and it's easily cured, but I'd like to know
> > > why it happens. Does anyone have any ideas?
> >
> > There are multiple types of PS/2 mice. It looks to me like you need to
> > have a look at being more specific to gpm with -t help. You can switch
> > the mice around between them to verify this. You might also want to
> > check that /dev/psaux is the same on all the machines, and double-check
> > that you have the appropriate kernel modules loaded.
> > 

# Mice
#
# CONFIG_BUSMOUSE is not set
CONFIG_MOUSE=y
CONFIG_PSMOUSE=y
# CONFIG_82C710_MOUSE is not set
# CONFIG_PC110_PAD is not set
# CONFIG_MK712_MOUSE is not set

in both kernels.  Neither has modules.

> 
> I have seen the "failed to read id" error many times, usually exiting x to rl 
> 3. I haven't taken the time to track down a solution. I suspect a time delay 
> is needed somewhere in the gpm scripts, though I can't be sure. Anyway, as 
> mentioned gpm -k followed by gpm -m <device> -t <type> generally works.
> 

Thanks for the gpm -t help suggestion, I have to admit that I'd
overlooked the difference between ps2 and imps2.

Anyway, if I specify the mouse type as just ps2, I get the same crazy
behaviour as when I start up the mouse for the first time using the
correct configuration.  Which is logical, because the error message
says clearly that gpm is treating the mouse as plain ps2 and not
imps2.

Kevin must be right, it looks as if gpm isn't waiting long enough for
a reply from the mouse.  But it's odd that I have this problem on the
_faster_ machine rather than the _slower_ one.

I'll just start and restart gpm by hand.  Thanks to everyone who
replied.

Paul

-- 
http://perso.wanadoo.fr/paul.bailey/
--



More information about the blfs-support mailing list