LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
To: Dave Hansen <haveblue@us.ibm.com>
Cc: akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
Christoph Hellwig <hch@lst.de>
Subject: Re: [2.6.25-rc5-mm1] BUG() at mnt_want_write().
Date: Wed, 12 Mar 2008 15:35:38 +0900 [thread overview]
Message-ID: <200803120635.m2C6Zc03091406@www262.sakura.ne.jp> (raw)
In-Reply-To: <1205291010.9828.60.camel@nimitz.home.sr71.net>
Hello.
I modified /usr/share/initrd-tools/init and tested.
----- start of patch -----
diff -u /usr/share/initrd-tools/init.org /usr/share/initrd-tools/init
--- /usr/share/initrd-tools/init.org 2005-05-27 08:44:00.000000000 +0900
+++ /usr/share/initrd-tools/init 2008-03-12 15:14:31.000000000 +0900
@@ -359,7 +359,10 @@
read root < bin/root
umount -n bin
+echo "About to write real-root-dev"
+echo "hello" > proc/version
echo $root > proc/sys/kernel/real-root-dev
+echo "Wrote real-root-dev"
get_cmdline
@@ -415,7 +418,9 @@
cd /
mount -nt proc proc proc
+echo "About to read real-root-dev"
rootdev=$(cat proc/sys/kernel/real-root-dev)
+echo "Read real-root-dev"
cmdline=$(cat /proc/cmdline)
umount -n proc
if [ $rootdev != 256 ]; then
----- end of patch -----
The result was
-----------------------------------------------------------------------------
initrd-tools: 0.1.81.1
do_sys_open: bin/root
do_sys_open: /etc/ld.so.preload
do_sys_open: /etc/ld.so.cache
do_sys_open: /lib/tls/i686/mmx/cmov/libblkid.so.1
do_sys_open: /lib/tls/i686/mmx/libblkid.so.1
do_sys_open: /lib/tls/i686/cmov/libblkid.so.1
do_sys_open: /lib/tls/i686/libblkid.so.1
do_sys_open: /lib/tls/mmx/cmov/libblkid.so.1
do_sys_open: /lib/tls/mmx/libblkid.so.1
do_sys_open: /lib/tls/cmov/libblkid.so.1
do_sys_open: /lib/tls/libblkid.so.1
do_sys_open: /lib/i686/mmx/cmov/libblkid.so.1
do_sys_open: /lib/i686/mmx/libblkid.so.1
do_sys_open: /lib/i686/cmov/libblkid.so.1
do_sys_open: /lib/i686/libblkid.so.1
do_sys_open: /lib/mmx/cmov/libblkid.so.1
do_sys_open: /lib/mmx/libblkid.so.1
do_sys_open: /lib/cmov/libblkid.so.1
do_sys_open: /lib/libblkid.so.1
do_sys_open: /lib/tls/libuuid.so.1
do_sys_open: /lib/libuuid.so.1
do_sys_open: /lib/tls/libc.so.6
do_sys_open: /etc/mtab
About to write real-root-dev
do_sys_open: proc/version
BUG: unable to handle kernel paging request at df366fe8
IP: [<c017e0d0>] mnt_want_write+0x41/0x5f
*pde = 1f884163 *pte = 1f366160
Oops: 0002 [#1] SMP DEBUG_PAGEALLOC
last sysfs file:
Modules linked in:
Pid: 1, comm: init Not tainted (2.6.25-rc5-mm1 #2)
EIP: 0060:[<c017e0d0>] EFLAGS: 00010286 CPU: 0
EIP is at mnt_want_write+0x41/0x5f
EAX: 00000000 EBX: c1453460 ECX: 00000001 EDX: df366f78
ESI: df363f78 EDI: 00000000 EBP: df8e1ef0 ESP: df8e1ee4
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process init (pid: 1, ti=df8e1000 task=df8f8020 task.ti=df8e1000)
Stack: ffffffeb 00000200 df8e1f20 df8e1f7c c0173481 ffffa87b 00000000 ffffff9c
c053481a 00008242 00000000 00000002 00008241 df363f78 df56cf50 df363f78
df56cf50 57b32ee9 00000007 df911005 00000300 00000000 00000000 c016a9ca
Call Trace:
[<c0173481>] ? do_filp_open+0x233/0x52f
[<c016a9ca>] ? get_unused_fd_flags+0xb3/0xbe
[<c016aabd>] ? do_sys_open+0x50/0xd0
[<c016ab53>] ? sys_open+0x16/0x18
[<c0103966>] ? syscall_call+0x7/0xb
=======================
Code: 8d 1c 02 89 d8 e8 d3 b4 1c 00 89 f0 e8 7e ff ff ff 85 c0 74 07 bf e2 ff ff ff eb 1f 8b 53 2c 39 f2 74 15 85 d2 74 0e 8b 43 28 90 <01> 42 70 c7 43 28
00 00 00 00 89 73 2c ff 43 28 89 d8 e8 cf b5
EIP: [<c017e0d0>] mnt_want_write+0x41/0x5f SS:ESP 0068:df8e1ee4
---[ end trace e13043ee7a72cd3f ]---
Kernel panic - not syncing: Attempted to kill init!
-----------------------------------------------------------------------------
"proc/version" caused the same problem with "proc/sys/kernel/real-root-dev".
And if I specify "bin/foo" instead of "proc/version", the result was
-----------------------------------------------------------------------------
About to write real-root-dev
do_sys_open: bin/foo
/sbin/init: 363: cannot create bin/foo: Read-only file system
do_sys_open: proc/sys/kernel/real-root-dev
BUG: unable to handle kernel paging request at df379fe8
-----------------------------------------------------------------------------
So, no problem with accessing "bin/", but has problem with "proc/".
Regards.
next prev parent reply other threads:[~2008-03-12 6:42 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-12 1:37 penguin-kernel
2008-03-12 3:03 ` Dave Hansen
2008-03-12 3:22 ` Andrew Morton
2008-03-12 9:25 ` Peter Zijlstra
2008-03-12 16:52 ` Matthew Wilcox
2008-03-13 0:21 ` Tetsuo Handa
2008-03-13 20:41 ` fix for boot-time mnt_want_write() bug Dave Hansen
2008-03-14 0:15 ` Tetsuo Handa
2008-03-12 5:46 ` [2.6.25-rc5-mm1] BUG() at mnt_want_write() Tetsuo Handa
2008-03-12 6:35 ` Tetsuo Handa [this message]
2008-03-12 3:43 ` Dave Hansen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200803120635.m2C6Zc03091406@www262.sakura.ne.jp \
--to=penguin-kernel@i-love.sakura.ne.jp \
--cc=akpm@linux-foundation.org \
--cc=haveblue@us.ibm.com \
--cc=hch@lst.de \
--cc=linux-kernel@vger.kernel.org \
--subject='Re: [2.6.25-rc5-mm1] BUG() at mnt_want_write().' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).