NetBSD Problem Report #10158

Received: (qmail 13721 invoked from network); 20 May 2000 05:47:07 -0000
Message-Id: <200005200542.e4K5gcg00364@rumolt.teuto.de>
Date: Sat, 20 May 2000 07:42:38 +0200 (MEST)
From: Martin Husemann <martin@rumolt.teuto.de>
Reply-To: martin@rumolt.teuto.de
To: gnats-bugs@gnats.netbsd.org
Subject: softdeps panic my machine
X-Send-Pr-Version: 3.95

>Number:         10158
>Category:       kern
>Synopsis:       softdeps panic my machine
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat May 20 05:48:00 +0000 2000
>Closed-Date:    Fri Jun 23 21:29:47 +0000 2000
>Last-Modified:  Fri Jun 23 21:29:47 +0000 2000
>Originator:     Martin Husemann
>Release:        cvs update from May 6
>Organization:
>Environment:

System: NetBSD rumolt.teuto.de 1.4Y NetBSD 1.4Y (RUMOLT) #0: Sun May 7 17:10:35 MEST 2000 martin@rumolt.teuto.de:/usr/src/sys-i4b/arch/i386/compile/RUMOLT i386


>Description:

While I have softdeps turned on on several machines which are happy with
it, this machine locks up with it regularily. Always when I'm in X and
doing something, so I hooked up a serial debugger and got a backtrace.

I can reproduce this, but not fast (this time it took about a week,
last time without debugger setup yet it was only minutes).

If someone can tell me how to get a crash dump from the remote gdb,
I'll be able to get that and set it up on the net. If I need to 
examine something, please tell me.

Here is the backtrace:

Program received signal SIGTRAP, Trace/breakpoint trap.
kgdb_panic () at ../../../../arch/i386/i386/kgdb_machdep.c:244
244             if (verbose)
(gdb) bt
#0  kgdb_panic () at ../../../../arch/i386/i386/kgdb_machdep.c:244
#1  0xc0137719 in panic (
    fmt=0xc0296b80 "softdep_disk_write_complete: lock is held")
    at ../../../../kern/subr_prf.c:208
#2  0xc025f238 in softdep_disk_write_complete (bp=0xc4cdcf88)
    at ../../../../../gnu/sys/ufs/ffs/ffs_softdep.c:3069
#3  0xc014e4d7 in biodone (bp=0xc4cdcf88) at ../../../../kern/vfs_bio.c:954
#4  0xc01f754a in wddone (v=0xc075e000) at ../../../../dev/ata/wd.c:622
#5  0xc01fa2cf in wdc_ata_bio_done (chp=0xc075c4a0, xfer=0xc0761000)
    at ../../../../dev/ata/ata_wdc.c:656
#6  0xc01f9bb4 in wdc_ata_bio_intr (chp=0xc075c4a0, xfer=0xc0761000, irq=1)
    at ../../../../dev/ata/ata_wdc.c:597
#7  0xc0109a95 in wdcintr (arg=0xc075c4a0) at ../../../../dev/ic/wdc.c:693
#8  0xc0203e62 in pciide_compat_intr (arg=0xc075c4a0)
    at ../../../../dev/pci/pciide.c:651
#9  0xc01018c0 in Xintr14 ()
#10 0xc025c9bf in softdep_freefile (v=0xd219ceb4)
    at ../../../../../gnu/sys/ufs/ffs/ffs_softdep.c:1856
#11 0xc01c87c4 in ffs_vfree (v=0xd219ceb4)
    at ../../../../ufs/ffs/ffs_alloc.c:1481
#12 0xc01ce399 in ufs_inactive (v=0xd219cee0)
    at ../../../../sys/vnode_if.h:1005
#13 0xc0152dcb in vput (vp=0xd239cbc0) at ../../../../sys/vnode_if.h:717
#14 0xc025e777 in handle_workitem_remove (dirrem=0xc084bcc0)
    at ../../../../../gnu/sys/ufs/ffs/ffs_softdep.c:2683
#15 0xc025a150 in softdep_process_worklist (matchmnt=0x0)
    at ../../../../../gnu/sys/ufs/ffs/ffs_softdep.c:544
#16 0xc015d681 in sched_sync (v=0x0)
    at ../../../../miscfs/syncfs/sync_subr.c:192
#17 0xc010031b in proc_trampoline ()
#18 0xc02d9efc in ?? ()
Cannot access memory at address 0xf9000000.


>How-To-Repeat:

Do "make cleandir" or "cvs update" on this machine with softdeps
enabled

>Fix:

n/a
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback 
State-Changed-By: mycroft 
State-Changed-When: Tue May 30 19:15:31 PDT 2000 
State-Changed-Why:  
Should be fixed in ffs_softdep.c 1.14. 

From: "Charles M. Hannum" <root@ihack.net>
To: Martin Husemann <martin@rumolt.teuto.de>
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/10158: softdeps panic my machine
Date: Tue, 30 May 2000 20:17:46 -0400 (EDT)

 I believe this bug is fixed with my latest commit:

 Index: ffs_softdep.c
 ===================================================================
 RCS file: /cvsroot/gnusrc/gnu/sys/ufs/ffs/ffs_softdep.c,v
 retrieving revision 1.13
 retrieving revision 1.14
 diff -c -2 -r1.13 -r1.14
 *** ffs_softdep.c	2000/05/31 00:16:16	1.13
 --- ffs_softdep.c	2000/05/31 02:06:57	1.14
 ***************
 *** 1,3 ****
 ! /*	$NetBSD: ffs_softdep.c,v 1.13 2000/05/31 00:16:16 mycroft Exp $	*/

   /*
 --- 1,3 ----
 ! /*	$NetBSD: ffs_softdep.c,v 1.14 2000/05/31 02:06:57 mycroft Exp $	*/

   /*
 ***************
 *** 1058,1064 ****
   	 * the cylinder group map from which it was allocated.
   	 */
 - 	ACQUIRE_LOCK(&lk);
   	if (newblk_lookup(fs, newblkno, DEPALLOC, &newblk) != 0)
   		panic("softdep_setup_blkmapdep: found block");
   	newblk->nb_bmsafemap = bmsafemap = bmsafemap_lookup(bp);
   	LIST_INSERT_HEAD(&bmsafemap->sm_newblkhd, newblk, nb_deps);
 --- 1058,1064 ----
   	 * the cylinder group map from which it was allocated.
   	 */
   	if (newblk_lookup(fs, newblkno, DEPALLOC, &newblk) != 0)
   		panic("softdep_setup_blkmapdep: found block");
 + 	ACQUIRE_LOCK(&lk);
   	newblk->nb_bmsafemap = bmsafemap = bmsafemap_lookup(bp);
   	LIST_INSERT_HEAD(&bmsafemap->sm_newblkhd, newblk, nb_deps);


From: Martin Husemann <martin@rumolt.teuto.de>
To: root@ihack.net (Charles M. Hannum)
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: kern/10158: softdeps panic my machine
Date: Fri, 23 Jun 2000 20:57:50 +0200 (MEST)

 > I believe this bug is fixed with my latest commit:

 OK, I couldn't reproduce it with that version while bashing on the system
 real hard for some weeks. Seems to be OK now.

 Thank's for fixing it!


 Martin
State-Changed-From-To: feedback->closed 
State-Changed-By: mycroft 
State-Changed-When: Fri Jun 23 14:29:17 PDT 2000 
State-Changed-Why:  
User indicates problem is fixed. 
>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.