NetBSD Problem Report #8699
Received: (qmail 3454 invoked by uid 1000); 29 Oct 1999 07:26:22 -0000
Message-Id: <19991029072622.3453.qmail@mail.netbsd.org>
Date: 29 Oct 1999 07:26:22 -0000
From: cgd@NetBSD.ORG
Reply-To: cgd@NetBSD.ORG
To: gnats-bugs@gnats.netbsd.org
Subject: i386 kernel crash (trap) with out of control makes run as root?
X-Send-Pr-Version: 3.95
>Number: 8699
>Category: kern
>Synopsis: i386 kernel crashed when make ran out of control as root.
>Confidential: no
>Severity: serious
>Priority: high
>Responsible: mycroft
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Fri Oct 29 00:27:00 +0000 1999
>Closed-Date: Fri Sep 21 07:09:23 +0000 2001
>Last-Modified: Fri Sep 21 07:09:23 +0000 2001
>Originator: Chris G. Demetriou
>Release: 1.4.1
>Organization:
Kernel Hackers 'r' Us
>Environment:
System: NetBSD speedy.int.demetriou.com 1.4.1 NetBSD 1.4.1 (SPEEDY) #16: Tue Sep 21 23:35:27 PDT 1999 cgd@speedy.int.demetriou.com:/a/src/src-1-4-branch/sys/arch/i386/compile/SPEEDY i386
(but "1.4.1" just means 1.4-branch as of 19990921)
>Description:
[filed as a kern bug, rather than port-i386 since the smoking
gun seems to have been held by tsleep...]
I ran a 'make' in pkgsrc as root, and it went nuts doing
an apparent recursive make because of some broken xpkgwedge
dependency goop. lots and lots of make processes.
after running it once and having it croak because the system
ran out of processes, I made the mistake of running it again.
After almost maxing out the process table again (ps on the dead
kernel said 501 processes, kern.maxproc on the rebooted kernel
is 532) the kernel crashed with:
fatal page fault in supervisor mode
trap type 6 code 0 eip f0133b1c cs 8 eflags 10246 cr2 a8 cpl 0
panic: trap
syncing disks... 12 12 9 done
(There are earlier relatively bad things; running out of mbuf
clusters, out of space in proc table, file system full, etc.,
but they happened a while before the crash.)
Unfortunately I don't have a copy of this kernel with debugging
symbols. That EIP is:
0xf0133b0a <tsleep+110>: je 0xf0133d2b <tsleep+655>
0xf0133b10 <tsleep+116>: call 0xf0101d1c <eintrnames>
0xf0133b15 <tsleep+121>: jmp 0xf0133d2b <tsleep+655>
0xf0133b1a <tsleep+126>: movl %esi,%esi
0xf0133b1c <tsleep+128>: movl 0xa8(%ebx),%eax
0xf0133b22 <tsleep+134>: andl $0x20000040,%eax
0xf0133b27 <tsleep+139>: cmpl $0x40,%eax
0xf0133b2a <tsleep+142>: jne 0xf0133b3e <tsleep+162>
I can provide the kernel and crash dump, at least for a while...
>How-To-Repeat:
as root, run an out of control recursive make?
>Fix:
unknown.
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->analyzed
State-Changed-By: mycroft
State-Changed-When: Fri Oct 29 02:14:22 PDT 1999
State-Changed-Why:
This is fixed in i386/locore.s 1.210 and vm_machdep.c 1.83.
Responsible-Changed-From-To: kern-bug-people->mycroft
Responsible-Changed-By: mycroft
Responsible-Changed-When: Fri Oct 29 02:15:28 PDT 1999
Responsible-Changed-Why:
State-Changed-From-To: analyzed->closed
State-Changed-By: chs
State-Changed-When: Fri Sep 21 00:08:50 PDT 2001
State-Changed-Why:
audit trail says it's been fixed for almost 2 years.
>Unformatted:
(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.