NetBSD Problem Report #56209

From www@netbsd.org  Tue May 25 14:55:13 2021
Return-Path: <www@netbsd.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 34D791A9277
	for <gnats-bugs@gnats.NetBSD.org>; Tue, 25 May 2021 14:55:13 +0000 (UTC)
Message-Id: <20210525145511.9CDB01A92A2@mollari.NetBSD.org>
Date: Tue, 25 May 2021 14:55:11 +0000 (UTC)
From: jsarkes@gmail.com
Reply-To: jsarkes@gmail.com
To: gnats-bugs@NetBSD.org
Subject: erratic usb operation
X-Send-Pr-Version: www-1.0

>Number:         56209
>Category:       kern
>Synopsis:       erratic usb operation
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue May 25 15:00:00 +0000 2021
>Last-Modified:  Mon Jun 07 03:05:02 +0000 2021
>Originator:     Joseph Sarkes
>Release:        9.1
>Organization:
none
>Environment:
unavailable atm, just the amd64 download for 9.1
>Description:
usb has irregular timeout messages with multiple mice of different manufacturers on console. the mice operate properly it seems. there ends up a lot of green messages on the console for connect / disconnect of the mice though.

a possibly related problem is a usb prologix gpib adapter has erratic operation where much corruption takes place semi randomly. using the prologix ++help command in minicom or a c program written to communicate with it results in around 1/3 of the help message displayed, the rest lost. this prologix device seems to work fine displaying the full help message in minicom on linux (i finally broke down and tried linux for troubleshooting.) i suspect the same thing that is messing with the mice is breaking the prologix (some type of timeout). my eyes cross when i try to trace through and find the driver stuff in kernel source.

if you need more info (if this isn't known about already, i tried to search but found nothing on my attempts) feel free to email for more info and to try stuff/report messages, etc. i remember netbsd serial being rock solid so this seems weird to me but it IS usb.
>How-To-Repeat:
for mice just plug in a mouse. both a cheap targus and something else i have that i can't find atm timeout.

if you have a prologix gpib adapter just run minicom or what have you and use the ++help command and likely it will be truncated. if it is working elsewhere it is a hardware issue with my toshiba laptop.

the prologix device is what i really need working. it is an ftdi serial / usb adapter chip talking to another programmed logic chip.
>Fix:
run linux? if i could get through the drivers properly i'd give it a try but i just go slackjawed. honestly i have always preferred netbsd with the standard loadout of unix apps and then some packages. linux is too different for me and bloated with all sorts of stuff loaded in. 

>Audit-Trail:
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/56209: erratic usb operation
Date: Tue, 25 May 2021 17:44:10 +0200

 If you want any serious help, you need to at least quote some of the 
 messages you see (from /var/log/messages or via the dmesg command).

 There is nothing wrong with the mice timeouts - the mice themselves do
 that. Typical fix:

  - run X (so the X server opens the mouse device)
  - run wsmoused (like: add wsmoused=YES to /etc/rc.conf)

 Please do not mix totaly unrelated issues in one PR.

 Martin

From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@netbsd.org
Cc: jsarkes@gmail.com
Subject: Re: kern/56209: erratic usb operation
Date: Tue, 25 May 2021 08:57:15 -0700 (PDT)

 On Tue, 25 May 2021, Martin Husemann wrote:

 > There is nothing wrong with the mice timeouts - the mice themselves do
 > that. Typical fix:
 >
 >  - run X (so the X server opens the mouse device)
 >  - run wsmoused (like: add wsmoused=YES to /etc/rc.conf)

 I can copnfirm that this fixes the problem - many of us have seen
 and reported this before.

 > Please do not mix totaly unrelated issues in one PR.

 Also please try not to mischaracterise the problem.  It's not
 really "erratic" - the detach/disconnect happens approximately
 62 seconds after the previous attach;  once disconnected, it
 will re-attach in a couple of seconds.  The time intervals are
 quite consistent.


 +--------------------+--------------------------+-----------------------+
 | Paul Goyette       | PGP Key fingerprint:     | E-mail addresses:     |
 | (Retired)          | FA29 0E3B 35AF E8AE 6651 | paul@whooppee.com     |
 | Software Developer | 0786 F758 55DE 53BA 7731 | pgoyette@netbsd.org   |
 +--------------------+--------------------------+-----------------------+

From: Christos Zoulas <christos@zoulas.com>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org,
 gnats-admin@netbsd.org,
 netbsd-bugs@netbsd.org,
 jsarkes@gmail.com
Subject: Re: kern/56209: erratic usb operation
Date: Tue, 25 May 2021 12:25:59 -0400

 --Apple-Mail=_C76984C6-5FF8-4874-AE9C-AD14E6395E50
 Content-Transfer-Encoding: 7bit
 Content-Type: text/plain;
 	charset=us-ascii

 Poor guy did not know :-)

 christos


 --Apple-Mail=_C76984C6-5FF8-4874-AE9C-AD14E6395E50
 Content-Transfer-Encoding: 7bit
 Content-Disposition: attachment;
 	filename=signature.asc
 Content-Type: application/pgp-signature;
 	name=signature.asc
 Content-Description: Message signed with OpenPGP

 -----BEGIN PGP SIGNATURE-----
 Comment: GPGTools - http://gpgtools.org

 iF0EARECAB0WIQS+BJlbqPkO0MDBdsRxESqxbLM7OgUCYK0lFwAKCRBxESqxbLM7
 OnIUAKChOsWxBEcjnTSceVw6Fv0Gw15gngCgmhlgEeiIQ0PjwQPG0tL70gO76ok=
 =QbLF
 -----END PGP SIGNATURE-----

 --Apple-Mail=_C76984C6-5FF8-4874-AE9C-AD14E6395E50--

From: Joseph Sarkes <jsarkes@gmail.com>
To: Paul Goyette <paul@whooppee.com>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/56209: erratic usb operation
Date: Fri, 28 May 2021 10:24:28 -0400

 ok, just for my edification here, you seem to have run with the mouse
 timeouts with this pr, which was intended to be regarding the prologix
 usb-gpib device. the disconnects were related info that i thought
 might apply, and not my intended pr.

 do i need to create a different pr with just the prologix devce
 mentioned and nothing else? i've been searching for my old usr modem
 that had a usb connection along with serial to see if the help screen
 on that was also corrupted but can't locate it atm.

 i apologize if i offended the sensibilities of people here. i just
 managed to figure out it was netbsd and not the device that was
 causing my problem as linux works with the device and i never expected
 a problem with the netbsd serial stuff. I spent a couple of weeks
 fighting with the damn thing and then started looking at the
 improbable.

 the upshot is that the prologix usb-gpib device doesn't work somehow
 with netbsd. not sure if it is some flow control issue or corruption,
 and i personally don't have the ability any more to hold onto enough
 details to have a chance of fixing it. I generally fight with things
 until i get them working but established this was beyond my
 capabilities.

 I'd like to reiterate that i included the timeout info as amplifying
 details, as i considered that it was not an issue for others, but
 might help point in some sort of crack one might pry into. I never
 expected to have multiple whacks with a stick when sending the pr. My
 assumption was that the problem report would be appreciated as a
 pointer to something perhaps needing attention, and not as a reason to
 strike swiftly with a stick.

 feel free to delete this mail once it is absorbed. no need to perturb
 people perpetually with a bit of feedback.

 On 5/25/21, Paul Goyette <paul@whooppee.com> wrote:
 > On Tue, 25 May 2021, Martin Husemann wrote:
 >
 >> There is nothing wrong with the mice timeouts - the mice themselves do
 >> that. Typical fix:
 >>
 >>  - run X (so the X server opens the mouse device)
 >>  - run wsmoused (like: add wsmoused=YES to /etc/rc.conf)
 >
 > I can copnfirm that this fixes the problem - many of us have seen
 > and reported this before.
 >
 >> Please do not mix totaly unrelated issues in one PR.
 >
 > Also please try not to mischaracterise the problem.  It's not
 > really "erratic" - the detach/disconnect happens approximately
 > 62 seconds after the previous attach;  once disconnected, it
 > will re-attach in a couple of seconds.  The time intervals are
 > quite consistent.
 >
 >
 > +--------------------+--------------------------+-----------------------+
 > | Paul Goyette       | PGP Key fingerprint:     | E-mail addresses:     |
 > | (Retired)          | FA29 0E3B 35AF E8AE 6651 | paul@whooppee.com     |
 > | Software Developer | 0786 F758 55DE 53BA 7731 | pgoyette@netbsd.org   |
 > +--------------------+--------------------------+-----------------------+
 >

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/56209: erratic usb operation
Date: Mon, 7 Jun 2021 03:03:16 +0000

 On Fri, May 28, 2021 at 02:25:01PM +0000, Joseph Sarkes wrote:
  >  ok, just for my edification here, you seem to have run with the mouse
  >  timeouts with this pr, which was intended to be regarding the prologix
  >  usb-gpib device. the disconnects were related info that i thought
  >  might apply, and not my intended pr.

 Well, that was what you said first, and it has a well-known
 explanation (though apparently not well-enough known; can you think of
 anywhere this behavior should have been documented that you would have
 seen it while investigating?)

 I suppose ums(4) would be a good starting point. The current ums(4)
 seems to be about two sentences...

  >  do i need to create a different pr with just the prologix devce
  >  mentioned and nothing else?

 No. One's enough, no reason it can't be this one.

 However, we do need to know more about the device in order to do
 anything about it, since we don't have one. A copy of the messages
 that get printed when you plug in the device (or at boot time) is
 a good starting point -- both the messages for the device itself and
 the usb plumbing (uhub, *hci, etc.) underneath it.

 Sorry everyone seems to have piled on you; that was unfortunate.

 -- 
 David A. Holland
 dholland@netbsd.org

From: "David A. Holland" <dholland@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/56209 CVS commit: src/share/man/man4
Date: Mon, 7 Jun 2021 03:04:50 +0000

 Module Name:	src
 Committed By:	dholland
 Date:		Mon Jun  7 03:04:50 UTC 2021

 Modified Files:
 	src/share/man/man4: ums.4

 Log Message:
 Mention in ums(4) that a lot of usb mice detach/reattcah when not in use.

 Prompted by PR 56209 but not related to the actual problem there.


 To generate a diff of this commit:
 cvs rdiff -u -r1.11 -r1.12 src/share/man/man4/ums.4

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

Home	PR Database Search