NetBSD Problem Report #30398
From dogcow@mercury.babymeat.com Thu Jun 2 05:47:50 2005
Return-Path: <dogcow@mercury.babymeat.com>
Received: from a.mail.sonic.net (a.mail.sonic.net [64.142.16.245])
by narn.netbsd.org (Postfix) with ESMTP id 604FB63B104
for <gnats-bugs@gnats.NetBSD.org>; Thu, 2 Jun 2005 05:47:50 +0000 (UTC)
Message-Id: <20050602053919.1BFA016674@mercury.babymeat.com>
Date: Wed, 1 Jun 2005 22:39:19 -0700 (PDT)
From: dogcow@babymeat.com
Reply-To: dogcow@babymeat.com
To: gnats-bugs@netbsd.org
Subject: panic in ohci_softintr
X-Send-Pr-Version: 3.95
>Number: 30398
>Category: kern
>Synopsis: panic in ohci_softintr
>Confidential: no
>Severity: serious
>Priority: low
>Responsible: kern-bug-people
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Thu Jun 02 05:48:00 +0000 2005
>Closed-Date: Sat Aug 13 18:34:34 +0000 2011
>Last-Modified: Sun Nov 13 18:45:02 +0000 2011
>Originator: Tom Spindler
>Release: NetBSD 3.99.5
>Organization:
>Environment:
System: NetBSD mercury.babymeat.com 3.99.5 NetBSD 3.99.5 (MERCURY) #19: Mon May 23 22:37:22 PDT 2005 dogcow@mercury.babymeat.com:/media/tmp/obj/obj/usr/src/sys/arch/i386/compile/MERCURY i386
Architecture: i386
Machine: i386
>Description:
While trying to get gphoto2 to recognize my camera, I ended up
(apparently) unplugging something at the wrong time. I have no idea
if having simultaneous ehci and ohci devices active exacerbated
the problem.
panic: ohci_softintr: addr 0x%08lx not found
#0 0xc0450000 in ?? ()
(gdb) bt
#0 0xc0450000 in ?? ()
#1 0xc026d822 in cpu_reboot (howto=256, bootstr=0x0)
at /usr/src/sys/arch/i386/i386/machdep.c:752
#2 0xc01eed8b in panic (
fmt=0xc034a600 "ohci_softintr: addr 0x%08lx not found")
at /usr/src/sys/kern/subr_prf.c:245
#3 0xc015969a in ohci_softintr (v=0xc11c5000)
at /usr/src/sys/dev/usb/ohci.c:1287
#4 0xc0267b6c in softintr_dispatch (which=1)
at /usr/src/sys/arch/x86/x86/softintr.c:104
(gdb) up
#1 0xc026d822 in cpu_reboot (howto=256, bootstr=0x0)
at /usr/src/sys/arch/i386/i386/machdep.c:752
752 dumpsys();
(gdb) up
#2 0xc01eed8b in panic (
fmt=0xc034a600 "ohci_softintr: addr 0x%08lx not found")
at /usr/src/sys/kern/subr_prf.c:245
245 cpu_reboot(bootopt, NULL);
(gdb) up
#3 0xc015969a in ohci_softintr (v=0xc11c5000)
at /usr/src/sys/dev/usb/ohci.c:1287
1287 panic("ohci_softintr: addr 0x%08lx not found", (u_long)done);
(gdb) print done
$1 = 1097136
(gdb) l
1282 done = le32toh(sitd->itd.itd_nextitd);
1283 sidone = sitd;
1284 DPRINTFN(5,("add ITD %p\n", sitd));
1285 continue;
1286 }
1287 panic("ohci_softintr: addr 0x%08lx not found", (u_long)done);
1288 }
1289
1290 DPRINTFN(10,("ohci_softintr: sdone=%p sidone=%p\n", sdone, sidone));
1291
(gdb) print sitd
$2 = (ohci_soft_itd_t *) 0xcb58bd54
(gdb) print *sitd
$3 = {itd = {itd_flags = 239044224, itd_bp0 = 0, itd_nextitd = 3411590852,
itd_be = 3411590664, itd_offset = {1, 0, 54204, 49546, 48556, 52056,
54635, 49180}}, nextitd = 0xcb58bdb4, dnext = 0x206,
physaddr = 3228564800, hnext = {le_next = 0x2, le_prev = 0xcb58bdb4},
xfer = 0x0, flags = 65280, isdone = 7 '\a'}
(gdb) up
#4 0xc0267b6c in softintr_dispatch (which=1)
at /usr/src/sys/arch/x86/x86/softintr.c:104
104 (*sih->sih_fn)(sih->sih_arg);
(gdb) print *sih
$4 = {sih_q = {tqe_next = 0x0, tqe_prev = 0xc0391f70},
sih_intrhead = 0xc0391f70, sih_fn = 0xc01591e0 <ohci_softintr>,
sih_arg = 0xc11c5000, sih_pending = 0}
>How-To-Repeat:
Unknown.
>Fix:
>Release-Note:
>Audit-Trail:
From: Karl Janmar <karl@utopiafoundation.org>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org, dogcow@babymeat.com
Subject: Re: kern/30398: panic in ohci_softintr
Date: Wed, 24 Aug 2005 15:11:43 +0200
This is a multi-part message in MIME format.
--------------070803040104010807080506
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
(This PR refers to the same problem as PR:
kern/30331 [serious/medium]: kernel panic somethimes with USB printer)
I have posted a message to tech-kern@ some time ago, but I didn't got any reply.
http://mail-index.netbsd.org/tech-kern/2005/07/13/0007.html
Since the USB guys havn't answered my question: "what happens if we just ignore
this condition and not issuing a panic?"
I guess they don't think this is a general problem that is prioritized.
As I write in my post to tech-kern I think Linux just print some error message
when this condition occur and continue. I have made a patch for netbsd that do
the same. Now my machine doesn't reboot ever 3 day or so, but I a lot of error
messages....but anyway it's still better than having a constantly panic'ing system.
I don't claim this is the best fix for the problem, but at least it stop the
system from constantly panic'ing. And it looks like I'm not the only one running
in to this problem. Why prevent people using netbsd on the same hardware that
they could run linux on (if this isn't a sw bug). I have been running linux on
this particular machine for about 1 year without any USB related panics.
Patch against netbsd-3 appended.
Here is the kernel messages from my system:
Jul 16 18:16:01 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c7d0 not found
Jul 27 09:01:17 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c7a0 not found
Jul 27 10:50:54 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c620 not found
Jul 27 14:09:20 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c620 not found
Jul 27 20:21:08 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c7d0 not found
Jul 27 22:13:46 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306c5f0 not found
Jul 28 17:35:17 ngong /netbsd: ohci_softintr: WARNING!! addr 0x0306b7d0 not found
Jul 29 22:17:24 ngong /netbsd: ohci_softintr: WARNING!! addr 0x030697d0 not found
Jul 29 22:43:46 ngong /netbsd: ohci_softintr: WARNING!! addr 0x03069740 not found
Jul 30 13:20:35 ngong /netbsd: ohci_softintr: WARNING!! addr 0x03069680 not found
Jul 30 13:39:40 ngong /netbsd: ohci_softintr: WARNING!! addr 0x030697d0 not
found
Regards,
Karl Janmar
karl@utopiafoundation.org
dogcow@babymeat.com wrote:
>>Number: 30398
>>Category: kern
>>Synopsis: panic in ohci_softintr
>>Confidential: no
>>Severity: serious
>>Priority: low
>>Responsible: kern-bug-people
>>State: open
>>Class: sw-bug
>>Submitter-Id: net
>>Arrival-Date: Thu Jun 02 05:48:00 +0000 2005
>>Originator: Tom Spindler
>>Release: NetBSD 3.99.5
>>Organization:
>
>
>
>>Environment:
>
>
>
> System: NetBSD mercury.babymeat.com 3.99.5 NetBSD 3.99.5 (MERCURY) #19: Mon May 23 22:37:22 PDT 2005 dogcow@mercury.babymeat.com:/media/tmp/obj/obj/usr/src/sys/arch/i386/compile/MERCURY i386
> Architecture: i386
> Machine: i386
>
>>Description:
>
>
> While trying to get gphoto2 to recognize my camera, I ended up
> (apparently) unplugging something at the wrong time. I have no idea
> if having simultaneous ehci and ohci devices active exacerbated
> the problem.
>
> panic: ohci_softintr: addr 0x%08lx not found
> #0 0xc0450000 in ?? ()
> (gdb) bt
> #0 0xc0450000 in ?? ()
> #1 0xc026d822 in cpu_reboot (howto=256, bootstr=0x0)
> at /usr/src/sys/arch/i386/i386/machdep.c:752
> #2 0xc01eed8b in panic (
> fmt=0xc034a600 "ohci_softintr: addr 0x%08lx not found")
> at /usr/src/sys/kern/subr_prf.c:245
> #3 0xc015969a in ohci_softintr (v=0xc11c5000)
> at /usr/src/sys/dev/usb/ohci.c:1287
> #4 0xc0267b6c in softintr_dispatch (which=1)
> at /usr/src/sys/arch/x86/x86/softintr.c:104
> (gdb) up
> #1 0xc026d822 in cpu_reboot (howto=256, bootstr=0x0)
> at /usr/src/sys/arch/i386/i386/machdep.c:752
> 752 dumpsys();
> (gdb) up
> #2 0xc01eed8b in panic (
> fmt=0xc034a600 "ohci_softintr: addr 0x%08lx not found")
> at /usr/src/sys/kern/subr_prf.c:245
> 245 cpu_reboot(bootopt, NULL);
> (gdb) up
> #3 0xc015969a in ohci_softintr (v=0xc11c5000)
> at /usr/src/sys/dev/usb/ohci.c:1287
> 1287 panic("ohci_softintr: addr 0x%08lx not found", (u_long)done);
> (gdb) print done
> $1 = 1097136
> (gdb) l
> 1282 done = le32toh(sitd->itd.itd_nextitd);
> 1283 sidone = sitd;
> 1284 DPRINTFN(5,("add ITD %p\n", sitd));
> 1285 continue;
> 1286 }
> 1287 panic("ohci_softintr: addr 0x%08lx not found", (u_long)done);
> 1288 }
> 1289
> 1290 DPRINTFN(10,("ohci_softintr: sdone=%p sidone=%p\n", sdone, sidone));
> 1291
> (gdb) print sitd
> $2 = (ohci_soft_itd_t *) 0xcb58bd54
> (gdb) print *sitd
> $3 = {itd = {itd_flags = 239044224, itd_bp0 = 0, itd_nextitd = 3411590852,
> itd_be = 3411590664, itd_offset = {1, 0, 54204, 49546, 48556, 52056,
> 54635, 49180}}, nextitd = 0xcb58bdb4, dnext = 0x206,
> physaddr = 3228564800, hnext = {le_next = 0x2, le_prev = 0xcb58bdb4},
> xfer = 0x0, flags = 65280, isdone = 7 '\a'}
> (gdb) up
> #4 0xc0267b6c in softintr_dispatch (which=1)
> at /usr/src/sys/arch/x86/x86/softintr.c:104
> 104 (*sih->sih_fn)(sih->sih_arg);
> (gdb) print *sih
> $4 = {sih_q = {tqe_next = 0x0, tqe_prev = 0xc0391f70},
> sih_intrhead = 0xc0391f70, sih_fn = 0xc01591e0 <ohci_softintr>,
> sih_arg = 0xc11c5000, sih_pending = 0}
>
>
>>How-To-Repeat:
>
> Unknown.
>
>
>>Fix:
>
>
>
>
>>Unformatted:
>
>
>
--------------070803040104010807080506
Content-Type: text/plain;
name="ohci_nopanic.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="ohci_nopanic.patch"
Index: dev/usb/ohci.c
===================================================================
RCS file: /cvsroot/src/sys/dev/usb/ohci.c,v
retrieving revision 1.157
diff -u -r1.157 ohci.c
--- dev/usb/ohci.c 11 Mar 2005 19:25:22 -0000 1.157
+++ dev/usb/ohci.c 24 Aug 2005 13:00:29 -0000
@@ -1275,7 +1275,8 @@
DPRINTFN(5,("add ITD %p\n", sitd));
continue;
}
- panic("ohci_softintr: addr 0x%08lx not found", (u_long)done);
+ printf("ohci_softintr: WARNING!! addr 0x%08lx not found\n", (u_long)done);
+ break;
}
DPRINTFN(10,("ohci_softintr: sdone=%p sidone=%p\n", sdone, sidone));
--------------070803040104010807080506--
From: Paul Goyette <paul@whooppee.com>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: kern/30398: panic in ohci_softintr
Date: Tue, 5 Apr 2011 05:02:50 -0700 (PDT)
This rather old PR (vintage 2005) has been idle for quite some time,
although there is actually some analysis and a patch is available. I've
been running the patch for many months now, and just as noted by the
originator, the only downside is the occassional logging of the error
message, which IMHO is much better than having the machine panic!
Is there any chance our USB experts might have a look at this, and maybe
commit the patch?
-------------------------------------------------------------------------
| Paul Goyette | PGP Key fingerprint: | E-mail addresses: |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net |
| Kernel Developer | | pgoyette at netbsd.org |
-------------------------------------------------------------------------
From: Jukka Ruohonen <jruohonen@iki.fi>
To: gnats-bugs@NetBSD.org
Cc: Paul Goyette <paul@whooppee.com>
Subject: Re: kern/30398: panic in ohci_softintr
Date: Wed, 6 Apr 2011 19:30:01 +0300
On Tue, Apr 05, 2011 at 01:25:02PM +0000, Paul Goyette wrote:
> This rather old PR (vintage 2005) has been idle for quite some time,
> although there is actually some analysis and a patch is available. I've
> been running the patch for many months now, and just as noted by the
> originator, the only downside is the occassional logging of the error
> message, which IMHO is much better than having the machine panic!
>
> Is there any chance our USB experts might have a look at this, and maybe
> commit the patch?
I am no USB expert, but given the state of the USB stack, I would believe
it is reasonable to change a panic(9) to a aprint_error(9)/_debug(9) as an
immediate "fix" to a nearly five year old bug that is supposedly reproducable.
State-Changed-From-To: open->pending-pullups
State-Changed-By: jmcneill@NetBSD.org
State-Changed-When: Sun, 07 Aug 2011 13:46:19 +0000
State-Changed-Why:
Patch applied, thanks!
From: "Jared D. McNeill" <jmcneill@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/30398 CVS commit: src/sys/dev/usb
Date: Sun, 7 Aug 2011 13:45:46 +0000
Module Name: src
Committed By: jmcneill
Date: Sun Aug 7 13:45:46 UTC 2011
Modified Files:
src/sys/dev/usb: ohci.c
Log Message:
PR# kern/30398: panic in ohci_softintr
If the host controller has a TD that the driver doesn't know about, ignore it
instead of calling panic. Patch from Karl Janmar.
To generate a diff of this commit:
cvs rdiff -u -r1.217 -r1.218 src/sys/dev/usb/ohci.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
From: "Jeff Rizzo" <riz@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/30398 CVS commit: [netbsd-5] src/sys/dev/usb
Date: Fri, 12 Aug 2011 20:52:25 +0000
Module Name: src
Committed By: riz
Date: Fri Aug 12 20:52:25 UTC 2011
Modified Files:
src/sys/dev/usb [netbsd-5]: ohci.c
Log Message:
Pull up following revision(s) (requested by jmcneill in ticket #1655):
sys/dev/usb/ohci.c: revision 1.218
PR# kern/30398: panic in ohci_softintr
If the host controller has a TD that the driver doesn't know about, ignore it
instead of calling panic. Patch from Karl Janmar.
To generate a diff of this commit:
cvs rdiff -u -r1.196 -r1.196.4.1 src/sys/dev/usb/ohci.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
State-Changed-From-To: pending-pullups->closed
State-Changed-By: jmcneill@NetBSD.org
State-Changed-When: Sat, 13 Aug 2011 18:34:34 +0000
State-Changed-Why:
Pulled up to netbsd-5 branch.
From: Pierre Pronchery <khorben@defora.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: Re: PR/30398 CVS commit: [netbsd-5] src/sys/dev/usb
Date: Mon, 15 Aug 2011 01:28:20 +0200
Hi,
I believe the pull-up to the netbsd-5 branch doesn't compile:
/usr/src/sys/dev/usb/ohci.c: In function 'ohci_softintr':
/usr/src/sys/dev/usb/ohci.c:1333: warning: implicit declaration of
function 'device_printf'
Turning the call to device_printf() to printf() fixed it for me.
HTH,
--
khorben
From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: PR/30398 CVS commit: [netbsd-5] src/sys/dev/usb
Date: Sun, 13 Nov 2011 18:41:43 +0000
On Sun, Aug 14, 2011 at 11:30:06PM +0000, Pierre Pronchery wrote:
> The following reply was made to PR kern/30398; it has been noted by GNATS.
>
> From: Pierre Pronchery <khorben@defora.org>
> To: gnats-bugs@gnats.NetBSD.org
> Cc:
> Subject: Re: PR/30398 CVS commit: [netbsd-5] src/sys/dev/usb
> Date: Mon, 15 Aug 2011 01:28:20 +0200
>
> Hi,
>
> I believe the pull-up to the netbsd-5 branch doesn't compile:
>
> /usr/src/sys/dev/usb/ohci.c: In function 'ohci_softintr':
> /usr/src/sys/dev/usb/ohci.c:1333: warning: implicit declaration of
> function 'device_printf'
>
> Turning the call to device_printf() to printf() fixed it for me.
I trust this got fixed?
--
David A. Holland
dholland@netbsd.org
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.
Home