From owner-freebsd-current Sat Jun 1 05:25:29 1996
Return-Path: owner-current
Received: (from root@localhost)
by freefall.freebsd.org (8.7.5/8.7.3) id FAA02038
for current-outgoing; Sat, 1 Jun 1996 05:25:29 -0700 (PDT)
Received: from cbgw2.att.com (cbgw2.att.com [192.20.239.134])
by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id FAA02025;
Sat, 1 Jun 1996 05:25:26 -0700 (PDT)
From: dob@nasvr1.cb.att.com
Received: from nasvr1.cb.att.com by cbig3.att.att.com (SMI-8.6/EMS-1.2 sol2)
id IAA25633; Sat, 1 Jun 1996 08:18:39 -0400
Received: by nasvr1.cb.att.com (5.x/EMS-1.1 Sol2)
id AA20267; Sat, 1 Jun 1996 08:20:47 -0400
Cc: ejc@nasvr1.cb.att.com
Received: from cbsky.cb.att.com by nasvr1.cb.att.com (5.x/EMS-1.1 Sol2)
id AA20263; Sat, 1 Jun 1996 08:20:44 -0400
Received: by cbsky.cb.att.com (5.x/EMS-1.1 Sol2)
id AA17037; Sat, 1 Jun 1996 08:20:20 -0400
Date: Sat, 1 Jun 1996 08:20:20 -0400
Message-Id: <9606011220.AA17037@cbsky.cb.att.com>
To: dyson@freebsd.org, freebsd-current@freebsd.org
Original-Cc: ejc@nasvr1
Subject: Re: Vm fixes NG
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk
John,
Interim report. Still NG. Sup from Friday, 7PM EDT.
Same conditions (btw: /usr/src/usr.bin/xlint/lint1/tree.c was ``broken'' by
including y.tab.h instead of cgram.tab.h, so make world stumbled a bit...).
Anyway, panic was ``panic: freeing held page, count = 4...''
Here are the date/times of the pertinent vm files:
-rw-r--r-- 1 root wheel 55171 May 30 22:52 /sys/i386/i386/pmap.c
-rw-r--r-- 1 root wheel 4405 May 29 04:51 /sys/vm/default_pager.c
-rw-r--r-- 1 root wheel 1912 May 30 22:53 /sys/vm/default_pager.h
-rw-r--r-- 1 root wheel 7935 May 18 04:51 /sys/vm/device_pager.c
-rw-r--r-- 1 root wheel 2222 Dec 14 04:58 /sys/vm/device_pager.h
-rw-r--r-- 1 root wheel 8323 Dec 7 10:56 /sys/vm/kern_lock.c
-rw-r--r-- 1 root wheel 4658 Oct 4 1995 /sys/vm/lock.h
-rw-r--r-- 1 root wheel 5522 May 19 04:48 /sys/vm/pmap.h
-rw-r--r-- 1 root wheel 40414 May 22 22:55 /sys/vm/swap_pager.c
-rw-r--r-- 1 root wheel 3507 May 22 22:55 /sys/vm/swap_pager.h
-rw-r--r-- 1 root wheel 2831 Dec 9 23:34 /sys/vm/vm.h
-rw-r--r-- 1 root wheel 4793 May 2 10:54 /sys/vm/vm_extern.h
-rw-r--r-- 1 root wheel 28302 May 30 22:53 /sys/vm/vm_fault.c
-rw-r--r-- 1 root wheel 15164 May 30 22:53 /sys/vm/vm_glue.c
-rw-r--r-- 1 root wheel 3565 Jul 13 1995 /sys/vm/vm_inherit.h
-rw-r--r-- 1 root wheel 4289 Dec 11 05:02 /sys/vm/vm_init.c
-rw-r--r-- 1 root wheel 13416 May 22 22:55 /sys/vm/vm_kern.c
-rw-r--r-- 1 root wheel 3621 May 18 04:51 /sys/vm/vm_kern.h
-rw-r--r-- 1 root wheel 61663 May 30 22:53 /sys/vm/vm_map.c
-rw-r--r-- 1 root wheel 9603 May 19 04:48 /sys/vm/vm_map.h
-rw-r--r-- 1 root wheel 6494 May 18 04:51 /sys/vm/vm_meter.c
-rw-r--r-- 1 root wheel 22071 May 30 22:53 /sys/vm/vm_mmap.c
-rw-r--r-- 1 root wheel 37868 May 30 22:53 /sys/vm/vm_object.c
-rw-r--r-- 1 root wheel 7289 May 19 04:48 /sys/vm/vm_object.h
-rw-r--r-- 1 root wheel 27036 May 30 22:53 /sys/vm/vm_page.c
-rw-r--r-- 1 root wheel 10391 May 18 04:51 /sys/vm/vm_page.h
-rw-r--r-- 1 root wheel 26630 May 30 22:53 /sys/vm/vm_pageout.c
-rw-r--r-- 1 root wheel 4299 Nov 21 1995 /sys/vm/vm_pageout.h
-rw-r--r-- 1 root wheel 9095 May 18 04:51 /sys/vm/vm_pager.c
-rw-r--r-- 1 root wheel 4161 Dec 11 05:02 /sys/vm/vm_pager.h
-rw-r--r-- 1 root wheel 5619 Dec 9 23:34 /sys/vm/vm_param.h
-rw-r--r-- 1 root wheel 3821 Dec 7 10:56 /sys/vm/vm_prot.h
-rw-r--r-- 1 root wheel 7984 Mar 28 10:50 /sys/vm/vm_swap.c
-rw-r--r-- 1 root wheel 3622 Jan 18 23:04 /sys/vm/vm_unix.c
-rw-r--r-- 1 root wheel 22236 May 3 22:47 /sys/vm/vnode_pager.c
-rw-r--r-- 1 root wheel 2415 Dec 14 04:58 /sys/vm/vnode_pager.h
More changes to come?
Thanks,
Dan O'Brien
---
> From dob Thu May 30 08:34:23 1996
> To: dyson@freebsd.org
> Cc: cbsky!ejc
> Subject: Vm fixes NG
>
> John,
>
> I supped latest vm fixes last night (as of 9:30 PM EDT), built kernel,
> and retested make world (using -pipe!). After about an hour, it panic'ed.
>
> Trace:
> _Debugger
> _panic
> _vm_page_free
> _pmap_release
> _vm_space_free
> _cpu_wait
> _wait1
> _wait4
> _syscall
> _Xsyscall
>
> I was using the modem on another virtual console at the time.
>
> I was using the standard splvm setting and not splvm == splhigh.
>
>
> System is 100Mhz Pentium, 8Mb memory, 0Kb L2 cache.
>
> Oh, you wanna buy this system as a test system? It seems to find all the
> holes? I would sell the case & mobo with 8Mb memory & 1Mb S3 video and
> built-in IDE controller, plus floppy. No harddisk, keyboard or monitor.
>
>
> Thanks,
>
> Dan O'Brien
>