NetBSD Problem Report #33241

From ignatios@fourier.cs.uni-bonn.de  Tue Apr 11 17:49:22 2006
Return-Path: <ignatios@fourier.cs.uni-bonn.de>
Received: from postfix.iai.uni-bonn.de (postfix.iai.uni-bonn.de [131.220.8.4])
	by narn.netbsd.org (Postfix) with ESMTP id 39D0C63B896
	for <gnats-bugs@gnats.NetBSD.org>; Tue, 11 Apr 2006 17:49:17 +0000 (UTC)
Message-Id: <20060411174915.GF2469@cs.uni-bonn.de>
Date: Tue, 11 Apr 2006 19:49:15 +0200
From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
Reply-To: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Subject: Nokia 6230i umodem device stopped to work in 3.0
X-Send-Pr-Version: 3.95

>Number:         33241
>Category:       kern
>Synopsis:       Nokia 6230i umodem device stopped to work in 3.0
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Apr 11 17:50:00 +0000 2006
>Closed-Date:    Tue Nov 27 19:45:34 +0000 2018
>Last-Modified:  Tue Nov 27 19:45:34 +0000 2018
>Originator:     Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
>Release:        NetBSD 3.0.0_STABLE
>Organization:

>Environment:


System: NetBSD rock 3.0.0_STABLE NetBSD 3.0.0_STABLE (VAIO0) #10: Tue Apr 11 19:17:19 CEST 2006 ignatios@tangens:/usr/local/sources/3.0/oi/sys/arch/i386/compile/VAIO0 i386
Architecture: i386
Machine: i386

umodem0 at uhub0 port 1 configuration 1 interface 1
umodem0: Nokia Nokia 6230i, rev 1.10/3.50, addr 2, iclass 2/2
umodem0: data interface 2, has CM over data, has break
umodem0: status change notification available
umodem_common_attach: sc=0xc0de5f00
ucom0 at umodem0
>Description:

The above described phone worked with 1.6ish userland and nearly-2.0 kernel.
When using NetBSD-3.0, cu hangs in ttyin. even SIGINFO seems to be blocked.

However, I can log in to the machine on other wscons consoles or via the
network, I can connect an USB mouse and use it with X, etc.

I can connect the phone as an UMASS device (if I tell it to allow this).
dmesg, mount, read from the mounted filesystem all work.

>How-To-Repeat:
	Upgrade to 3.0. Go to test session with boss and his phone.

>Fix:
	unknown.

>Release-Note:

>Audit-Trail:

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Wed, 12 Apr 2006 14:59:17 +0200

 Reference point: a 2.99.16 kernel binary (built on 20050226) works.

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Wed, 23 Aug 2006 13:13:41 +0200

 The problem persists in a kernel compiled from the source code of
 NetBSD-3.0.1. I hope I did mention that this is a critical problem,
 and IMO, the change that broke this (from 2.99.16) should be backed out?

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Wed, 23 Aug 2006 15:41:35 +0200

 Another datapoint: 4.99.1 (cvs of today) works.
 Whoever fixed it: please backport the fix to the netbsd-3 branch. Or at 
 least to the netbsd-4 branch, if not already there.

 Regards,
 	-is

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Thu, 24 Aug 2006 11:40:08 +0200

 This is fixed in 4.0_BETA :

 NetBSD rock 4.0_BETA NetBSD 4.0_BETA (VAIO0) #0: Wed Aug 23 18:10:11 CEST 2006  ignatios@tangens:/usr/local/sources/4.0/oi/sys/arch/i386/compile/VAIO0 i386

 if possible, please pull up the umodem fix to the netbsd-3 branch, too!

 Regards,
 	-is

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Fri, 1 Sep 2006 21:09:32 +0200

 I just noticed that I didn't add a few observations to this PR:

 I tried to pull up the few post-3.0 trunk changes to the 3.0.1 sources that
 seemed to be relevant to the umodem code one at a time; none of them fixes
 the problem.

 I suspect that maybe something is missing in the the pullup that changed
 umodem on the netbsd-3 branch.

From: itohy@netbsd.org (ITOH Yasufumi)
To: ignatios@cs.uni-bonn.de
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Sun, 3 Sep 2006 22:59:40 +0900 (JST)

 The umodem changes had been pulled up to the netbsd-2 and netbsd-1-6 branches.
 How does your device work on the branches?
 -- 
 ITOH Yasufumi

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Mon, 4 Sep 2006 09:02:23 +0200

 On Sun, Sep 03, 2006 at 03:05:13PM +0000, ITOH Yasufumi wrote:

 >  The umodem changes had been pulled up to the netbsd-2 and netbsd-1-6 branches.
 >  How does your device work on the branches?

 I can't tell. 

 I might be able to set up a test machine of nearly the same type, and
 abuse my partner's phone once I buy an adapter, but

 - this will take a while and 
 - see "nearly same type".

 If I can reproduce the problem with that one, I'll test the netbsd-2 
 and maybe netbsd-1.6 branch.

 	-is

From: Ignatios Souvatzis <ignatios@cs.uni-bonn.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/33241: Nokia 6230i umodem device stopped to work in 3.0
Date: Thu, 21 Sep 2006 11:31:22 +0200

 On Sun, Sep 03, 2006 at 03:05:13PM +0000, ITOH Yasufumi wrote:

 >  The umodem changes had been pulled up to the netbsd-2 and netbsd-1-6
 >  branches. How does your device work on the branches?

 I didn't test netbsd-1-6.

 I did test said umodem with a slightly different model of Sony VAIO, and
 a netbsd-2.1 userland installation, using a quick cu -l $DEVICE :

 * netbsd-3.0.1 and
 * netbsd-3.1 RC3 kern: hang as described in the PR, with 
  /dev/ttyU0 as well as /dev/dtyU0.

 Additional observations:
 - ~. (from cu) needs a long time.
 - The kernel complains about short transfers on aue0: when I try to ftp
   while this happens.

 * netbsd-4_BETA,
 * netbsd-2.1 and
 * netbsd-2.0.2 all are fine. 

 All kernels where GENERIC_LAPTOP from ftp.netbsd.org (from NetBSD-daily
 where applicable).

 I did an additional test on an AMD64 desktop running netbsd-i386 3.0.1;
 same result (hangs). Here's the debug code from that machine:

 umodem0 at uhub0 port 2 configuration 1 interface 1
 umodem0: Nokia Nokia 6230i, rev 1.10/3.50, addr 2, iclass 2/2
 umodem0: data interface 2, has CM over data, has break
 is: no cm over data forced.
 umodem0: status change notification available
 umodem_common_attach: sc=0xc1d8ed00
 ucom0 at umodem0
 umodem_open: sc=0xc1d8ed00
 umodem_get_status:
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem0: notify bytes = 0200
 umodem_get_status:
 umodem_dtr: onoff=1
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem_set_line_coding: already set
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem_set_line_coding: already set
 umodem_ioctl: cmd=0x20007471
 umodem_ioctl: unknown
 umodem_dtr: onoff=0
 umodem_close: sc=0xc1d8ed00
 umodem_open: sc=0xc1d8ed00
 umodem_get_status:
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem_set_line_coding: already set
 umodem_dtr: onoff=1
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem_set_line_coding: already set
 umodem_param: sc=0xc1d8ed00
 umodem_set_line_coding: rate=9600 fmt=0 parity=0 bits=8
 umodem_set_line_coding: already set
 umodem_ioctl: cmd=0x20007471
 umodem_ioctl: unknown
 umodem_dtr: onoff=0
 umodem_close: sc=0xc1d8ed00
 umodem0: at uhub0 port 2 (addr 2) disconnected
 umodem_common_detach: sc=0xc1d8ed00 flags=1
 ucom0 detached
 umodem0 detached

 Regards,
 	-is

State-Changed-From-To: open->closed
State-Changed-By: maya@NetBSD.org
State-Changed-When: Tue, 27 Nov 2018 19:45:34 +0000
State-Changed-Why:
Report says this was fixed in netbsd-4, a backport to netbsd-3 is not going to happen now.


>Unformatted:

Home	PR Database Search