From aagero@aage.priv.no Sat Apr 6 14:11:05 1996 Received: from birk04.studby.uio.no (birk04.studby.uio.no [129.240.214.13]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id OAA19489 for ; Sat, 6 Apr 1996 14:11:04 -0800 (PST) Received: (from aagero@localhost) by birk04.studby.uio.no (8.7.5/sendmail95) id AAA00933; Sun, 7 Apr 1996 00:10:56 +0200 (MET DST) Message-Id: <199604062210.AAA00933@birk04.studby.uio.no> Date: Sun, 7 Apr 1996 00:10:56 +0200 (MET DST) From: aagero@aage.priv.no Reply-To: aagero@aage.priv.no To: FreeBSD-gnats-submit@freebsd.org Subject: system panics after editing a disklabel X-Send-Pr-Version: 3.2 >Number: 1120 >Category: kern >Synopsis: system panics when writing back the disklabel to the disk >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs >State: closed >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sat Apr 6 14:20:01 PST 1996 >Closed-Date: Sun Apr 21 06:50:35 PDT 1996 >Last-Modified: Sun Apr 21 06:52:59 PDT 1996 >Originator: =C5ge R=F8bekk >Release: FreeBSD 2.2-CURRENT i386 >Organization: =20 >Environment: =20 FreeBSD 2.2-CURRENT, kernel built of sources from april 4 >Description: =20 $ gdb -k kernel /var/crash/vmcore.2=20 GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.13 (i386-unknown-freebsd),=20 Copyright 1994 Free Software Foundation, Inc... IdlePTD 20a000 current pcb at 1fddf4 panic: biodone: buffer not busy #0 boot (howto=3D260) at ../../i386/i386/machdep.c:942 942 dumppcb.pcb_ptd =3D rcr3(); (kgdb) bt #0 boot (howto=3D260) at ../../i386/i386/machdep.c:942 #1 0xf0116c46 in panic (fmt=3D0xf01b9a7c "page fault") at ../../kern/subr_prf.c:133 #2 0xf01ba5d6 in trap_fatal (frame=3D0xf01e6ce0) at = ../../i386/i386/trap.c:740 #3 0xf01ba0c8 in trap_pfault (frame=3D0xf01e6ce0, usermode=3D0) at ../../i386/i386/trap.c:651 #4 0xf01b9dab in trap (frame=3D{tf_es =3D 16, tf_ds =3D -255721456,=20 tf_edi =3D -255363456, tf_esi =3D 17, tf_ebp =3D -266441428,=20 tf_isp =3D -266441464, tf_ebx =3D 0, tf_edx =3D 80, tf_ecx =3D 0, = tf_eax =3D 17,=20 tf_trapno =3D 12, tf_err =3D -266469376, tf_eip =3D -267311499,=20 tf_cs =3D 25624584, tf_eflags =3D 66118, tf_esp =3D -222174484,=20 tf_ss =3D -1073721280}) at ../../i386/i386/trap.c:319 #5 0xf01b2ce1 in calltrap () #6 0xf012af05 in getblk (vp=3D0xf0c77680, blkno=3D80, size=3D8192, = slpflag=3D0,=20 slptimeo=3D0) at ../../kern/vfs_bio.c:982 #7 0xf0129c35 in bread (vp=3D0xf0c77680, blkno=3D80, size=3D8192, = cred=3D0x0,=20 bpp=3D0xf01e6dc8) at ../../kern/vfs_bio.c:205 #8 0xf019a375 in ffs_update (ap=3D0xf01e6df4) at = ../../ufs/ffs/ffs_inode.c:137 #9 0xf019d934 in ffs_fsync (ap=3D0xf01e6e38) at ./vnode_if.h:1031 #10 0xf019c560 in ffs_sync (mp=3D0xf0c26000, waitfor=3D2, = cred=3D0xf0ab6400,=20 p=3D0xf02046b4) at ./vnode_if.h:407 #11 0xf013082b in sync (p=3D0xf02046b4, uap=3D0x0, retval=3D0x0) at ../../kern/vfs_syscalls.c:356 ---Type to continue, or q to quit--- #12 0xf01b531d in boot (howto=3D256) at ../../i386/i386/machdep.c:891 #13 0xf0116c46 in panic (fmt=3D0xf012b875 "biodone: buffer not busy") at ../../kern/subr_prf.c:133 #14 0xf012b9d9 in biodone (bp=3D0xf2c21230) at ../../kern/vfs_bio.c:1359 #15 0xf0192b58 in scsi_done (xs=3D0xf0ab6880) at ../../scsi/scsi_base.c:444 #16 0xf018ab1a in ncr_complete (np=3D0xf0c0e000, cp=3D0xf0c0ac00) at ../../pci/ncr.c:4298 #17 0xf018ab65 in ncr_wakeup (np=3D0xf0c0e000, code=3D0) at = ../../pci/ncr.c:4340 #18 0xf018b68c in ncr_exception (np=3D0xf0c0e000) at ../../pci/ncr.c:4943 #19 0xf0189dfa in ncr_intr (vnp=3D0xf0c0e000) at ../../pci/ncr.c:3524 (kgdb) up 14 #14 0xf012b9d9 in biodone (bp=3D0xf2c21230) at ../../kern/vfs_bio.c:1359 1359 panic("biodone: buffer not busy"); (kgdb) list 1354 { 1355 int s; 1356 1357 s =3D splbio(); 1358 if (!(bp->b_flags & B_BUSY)) 1359 panic("biodone: buffer not busy"); 1360 1361 if (bp->b_flags & B_DONE) { 1362 splx(s); 1363 printf("biodone: buffer already done\n"); The one thing I dont quite understand is the page fault-panic in frame 1. Why another panic? >How-To-Repeat: =20 disklabel -e sd? exit and save >Fix: =20 Not known. >Release-Note: >Audit-Trail: From: Bruce Evans To: FreeBSD-gnats-submit@freebsd.org, aagero@aage.priv.no Cc: Subject: Re: kern/1120: system panics after editing a disklabel Date: Sun, 7 Apr 1996 17:27:58 +1000 >The one thing I dont quite understand is the page fault-panic in frame 1. >Why another panic? The buffer containing the label is screwed up and syncing it causes more problems. Perhaps sync() shouldn't be attempted when the original panic was in disk-related code. Bruce State-Changed-From-To: open->closed State-Changed-By: bde State-Changed-When: Sun Apr 21 06:50:35 PDT 1996 State-Changed-Why: Fixed in rev.1.23 of ufs_disksubr.c. The bug only affected versions of -current between 1996/03/01 and 1996/04/19. >Unformatted: