[Pkg-gpm-devel] Bug#269161: marked as done (No luck with gpm-1.19.6-{12,12.1,16,17}, X and kernel-image-2.4.{25,26})

Wed, 01 Sep 2004 13:48:16 -0700

Your message dated Wed, 1 Sep 2004 15:39:23 -0500
with message-id <20040901203923.GA1444@p12n.org>
and subject line No luck with gpm-1.19.6-{12,12.1,16,17}, X and kernel-image-2.4.{25,26}
has caused the attached Bug report to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere.  Please contact me immediately.)

Debian bug tracking system administrator
(administrator, Debian Bugs database)

--------------------------------------
Received: (at submit) by bugs.debian.org; 31 Aug 2004 06:13:16 +0000
>From svante.signell@telia.com Mon Aug 30 23:13:16 2004
Return-path: <svante.signell@telia.com>
Received: from av7-1-sn1.fre.skanova.net [81.228.11.113] 
	by spohr.debian.org with esmtp (Exim 3.35 1 (Debian))
	id 1C21tM-0002BT-00; Mon, 30 Aug 2004 23:13:16 -0700
Received: by av7-1-sn1.fre.skanova.net (Postfix, from userid 502)
	id 15FCC37E69; Tue, 31 Aug 2004 08:12:45 +0200 (CEST)
Received: from smtp3-2-sn1.fre.skanova.net (smtp3-2-sn1.fre.skanova.net [81.228.11.164])
	by av7-1-sn1.fre.skanova.net (Postfix) with ESMTP id 0571E37E4C
	for <submit@bugs.debian.org>; Tue, 31 Aug 2004 08:12:45 +0200 (CEST)
Received: from em2.my.own.domain (h5n2fls301o1037.telia.com [81.227.209.5])
	by smtp3-2-sn1.fre.skanova.net (Postfix) with ESMTP id ED9EE37E53;
	Tue, 31 Aug 2004 08:12:44 +0200 (CEST)
Received: from srs by em2.my.own.domain with local (Exim 3.36 #1 (Debian))
	id 1C21sr-0003ge-00; Tue, 31 Aug 2004 08:12:45 +0200
Subject: No luck with gpm-1.19.6-{12,12.1,16,17}, X and
	kernel-image-2.4.{25,26}
From: Svante Signell <svante.signell@telia.com>
To: submit@bugs.debian.org
Cc: svante.signell@telia.com
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1093932764.8591.18.camel@em2.my.own.domain>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.6 
Date: Tue, 31 Aug 2004 08:12:44 +0200
Sender: Svante Signell <svante.signell@telia.com>
Delivered-To: submit@bugs.debian.org
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 
	(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Status: No, hits=-8.0 required=4.0 tests=BAYES_00,HAS_PACKAGE 
	autolearn=no version=2.60-bugs.debian.org_2004_03_25
X-Spam-Level: 

Package: gpm 
Version: 1.19.6-17
Severity: important

I don't know how long this has been broken. I can get the mouse to work
in the console with gpm _or_ in X without gpm, not both any longer.

/var/log/daemon.log:
Aug 30 23:06:39 em2 /usr/sbin/gpm[7306]: write_ps2: 1 0, fe
Aug 30 23:06:39 em2 /usr/sbin/gpm[7306]: PS/2 mouse failed init
Aug 30 23:06:39 em2 /usr/sbin/gpm[7306]: oops() invoked from gpm.c(1019)
Aug 30 23:06:39 em2 /usr/sbin/gpm[7306]: mouse initialization failed:
Inappropriate ioctl for device

/etc/gpm.conf:
device=/dev/psaux
responsiveness=
repeat_type=ms3
type=imps2
#append=""
sample_rate=

---------------------------------------
Received: (at 269161-done) by bugs.debian.org; 1 Sep 2004 20:39:25 +0000
>From ps@p12n.org Wed Sep 01 13:39:25 2004
Return-path: <ps@p12n.org>
Received: from smtp800.mail.sc5.yahoo.com [66.163.168.179] 
	by spohr.debian.org with smtp (Exim 3.35 1 (Debian))
	id 1C2bt7-0002Xx-00; Wed, 01 Sep 2004 13:39:25 -0700
Received: from unknown (HELO wire.p12n.org) (ps.2@sbcglobal.net@67.67.232.175 with login)
  by smtp800.mail.sc5.yahoo.com with SMTP; 1 Sep 2004 20:39:24 -0000
Received: by wire.p12n.org (Postfix, from userid 1001)
	id E87CA9FE8B; Wed,  1 Sep 2004 15:39:23 -0500 (CDT)
Date: Wed, 1 Sep 2004 15:39:23 -0500
From: Peter Samuelson <peter@p12n.org>
To: 269161-done@bugs.debian.org
Subject: Re: No luck with gpm-1.19.6-{12,12.1,16,17}, X and kernel-image-2.4.{25,26}
Message-ID: <20040901203923.GA1444@p12n.org>
References: <1093932764.8591.18.camel@em2.my.own.domain> <20040831205020.GB27482@p12n.org> <1094020488.1312.19.camel@em2.my.own.domain>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="x+6KMIRAuhnl3hBn"
Content-Disposition: inline
In-Reply-To: <1094020488.1312.19.camel@em2.my.own.domain>
User-Agent: Mutt/1.5.6+20040818i
Delivered-To: 269161-done@bugs.debian.org
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 
	(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Status: No, hits=-0.1 required=4.0 tests=BAYES_44 autolearn=no 
	version=2.60-bugs.debian.org_2004_03_25
X-Spam-Level: 

--x+6KMIRAuhnl3hBn
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

[Svante Signell]
> No I don't. The Device is set to /dev/psaux, and the Protocol is
> ImPS/2.

As I said, this is an unsupported configuration.  If it works - great.
If not, use the repeater.

> I've removed the repeat_type=ms3 from gpm.conf and the mouse works OK
> in both X and in the console. Is there any advantage to use the GPM
> repeater functionality, compared to the alternate?

It's strange that the presence or absence of the repeater line would
make a difference here.  I suspect it's a heisenbug - that is, it works
by chance some of the time, and not other times.

Anyway, here is the story on the gpm repeater: a PS/2 mouse, including
yours, keeps an internal state.  Primarily of interest is the protocol
mode - ImPS/2 mice are backward compatible with old PS/2 mice, and they
speak the old protocol until they are sent a magic command sequence to
put them in Intellimouse mode.  In fact, newer models (purchased in,
say, the last 4 years or so) also speak an enhanced protocol which we
know as "exps2" or "ExplorerPS/2", so they actually have *three*
possible protocol modes.

As such, it's a bad idea for two applications to both think they've got
sole control of the mouse.  The mouse will get put in some state
(primarily concerning the mouse protocol) by each application, and if
the two apps don't agree on the exact state they leave the mouse in,
someone will eventually get confused.  GPM and X both try to work
around this difficulty by sending some sort of mouse reset when you
switch from console mode to X and back, but that's not the sort of
thing I like to rely on.

The GPM repeater avoids this problem, by chaining the mouse events.
GPM handles the physical mouse, and X is fed events in a less
complicated protocol such as serial Intellimouse (gpm calls this "ms3",
X calls it "Intellimouse").  Since XFree86 is not talking directly to
the mouse, you don't have the problem of the two apps fighting over
who's controlling the mouse's internal state.

Kernel 2.6 *also* avoids this problem by virtualising the whole mouse
interface.  Basically the kernel takes on the same role as the gpm
repeater - providing an emulated mouse device for anyone who wants one,
and controlling the real mouse/mice internally.

Anyway, I'm closing this bug, as you suggested.  If you've got further
questions about this stuff, feel free to email me directly.

Thanks,
Peter

--x+6KMIRAuhnl3hBn
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)

iD8DBQFBNjN7Xk7sIRPQRh0RAq6MAKCIyS2zyRsihkoGylFIQQVuKXUimgCePUVG
Ce3iJA7lCQCqQzcocRTy0Gs=
=To8M
-----END PGP SIGNATURE-----

--x+6KMIRAuhnl3hBn--