From: Ville Herva <vherva@mail.niksula.cs.hut.fi>
To: Alan Cox <laughing@shared-source.org>
Cc: linux-kernel@vger.kernel.org
Subject: [OOPS] Re: Linux 2.4.5-ac1
Date: Sun, 27 May 2001 19:26:50 +0300 [thread overview]
Message-ID: <20010527192650.H11981@niksula.cs.hut.fi> (raw)
In-Reply-To: <20010526225825.A31713@lightning.swansea.linux.org.uk>
In-Reply-To: <20010526225825.A31713@lightning.swansea.linux.org.uk>; from laughing@shared-source.org on Sat, May 26, 2001 at 10:58:25PM +0100
On Sat, May 26, 2001 at 10:58:25PM +0100, you [Alan Cox] claimed:
>
> o Free the initial ramdisk correctly
Who made this fix, or who can I contact?
I have a reproducible oops on 2.4.4ac17 (see
http://marc.theaimsgroup.com/?l=linux-kernel&m=99079948404775&w=2 for
details) that seems to be related:
(1)Unable to handle kernel NULL pointer
dereference at virtual address 00000013
>>EIP; c02997f6 <__devices_1045+5a/a8> <=====
Trace; c0136447 <ioctl_by_bdev+77/90>
Trace; c017fec3 <batch_entropy_process+b3/c0>
Trace; c0118769 <__run_task_queue+49/60>
Trace; c011b0f6 <update_wall_time+16/50>
Trace; c011b304 <timer_bh+24/250>
Trace; c011868c <bh_action+1c/60>
Trace; c0118596 <tasklet_hi_action+36/60>
Trace; c011849b <do_softirq+5b/80>
Trace; c010832f <do_IRQ+9f/b0>
Trace; c0106f14 <ret_from_intr+0/20>
Trace; c0199756 <rd_make_request+d6/100>
Trace; c01330d6 <try_to_free_buffers+f6/140>
Trace; c0131447 <getblk+e7/100>
Trace; c0122cf5 <truncate_list_pages+145/170>
Trace; c01416db <destroy_inode+1b/20>
Trace; c0142b81 <iput+121/130>
Trace; c0136641 <blkdev_put+71/a0>
Trace; c013484a <kill_super+da/100>
Trace; c0134ab0 <do_umount+e0/f0>
Trace; c0105000 <do_linuxrc+0/e0>
Trace; c01177e6 <sys_waitpid+16/20>
Trace; c0105000 <do_linuxrc+0/e0>
Trace; c01051da <prepare_namespace+fa/120>
Trace; c010520e <init+e/140>
Trace; c0105000 <do_linuxrc+0/e0>
Trace; c01056a6 <kernel_thread+26/30>
Trace; c0105200 <init+0/140>
Code; c02997f6 <__devices_1045+5a/a8>
00000000 <_EIP>:
Code; c02997f6 <__devices_1045+5a/a8> <=====
0: 8b 40 10 mov 0x10(%eax),%eax <=====
Code; c02997f9 <__devices_1045+5d/a8>
3: 83 f8 02 cmp $0x2,%eax
Code; c02997fc <__devices_1045+60/a8>
6: 7e 62 jle 6a <_EIP+0x6a> c0299860
<__devices_104a
+c/20>
Code; c02997fe <__devices_1045+62/a8>
8: b8 f0 ff ff ff mov $0xfffffff0,%eax
Code; c0299803 <__devices_1045+67/a8>
d: eb 74 jmp 83 <_EIP+0x83> c0299879
<__devices_104b
+5/18>
Code; c0299805 <__devices_1045+69/a8>
f: 85 c9 test %ecx,%ecx
Code; c0299807 <__devices_1045+6b/a8>
11: b8 ea ff 00 00 mov $0xffea,%eax
Judging from the stack trace, I suspect that after kill_super is called (and
hence, root device invalidated and freed), a function that that assumes a
valid rootdev (ioctl_by_bdev) is called. Rdev is NULL, and hence the null
ptr deref. Is this anywhere near the truth?
-- v --
v@iki.fi
PS: A this is a 600Mhz Xeon w/ 256MB RAM. Right after boot I did
'diff -nauR linux-2.4.5 linux-2.4.4ac17' on redhat 2.4.2-2, and it never
finished:
root@machine:/poista>diff -nauR linux-2.4.5 linux-2.4.4ac17
zsh: terminated diff -nauR linux-2.4.5 linux-2.4.4ac17
root@machine:/poista>free
total used free shared buffers cached
Mem: 255572 253140 2432 0 8332 215108
-/+ buffers/cache: 29700 225872
Swap: 0 0 0
root@machine:/poista>uname -a
Linux machine 2.4.2-2 #1 Sun Apr 8 20:41:30 EDT 2001 i686 unknown
root@machine:/poista>dmesg | tail -2
Out of Memory: Killed process 804 (xfs).
Out of Memory: Killed process 15857 (diff).
So diff (which used very little RAM) filled the _cache_ and OOM rambo kicked
in. Pretty embarrassing to say at least...
next prev parent reply other threads:[~2001-05-27 16:27 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-05-26 21:58 Alan Cox
2001-05-27 16:26 ` Ville Herva [this message]
2001-05-27 21:12 ` initrd oops [was Re: Linux 2.4.5-ac1] Ville Herva
2001-05-28 7:25 ` initrd oops; still happens with 2.4.5ac2 Ville Herva
2001-05-28 8:51 ` Ville Herva
2001-05-28 9:02 ` Masaru Kawashima
2001-05-28 10:05 ` initrd oops with 2.4.5ac2: FIXED by Kawashima (the other oops may remain) Ville Herva
2001-05-28 10:28 ` initrd oops with 2.4.5ac2: Tthe other oops remains (one fixed) Ville Herva
2001-05-28 10:48 ` initrd oops with 2.4.5ac2: megaraid OOPSes Ville Herva
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=20010527192650.H11981@niksula.cs.hut.fi \
--to=vherva@mail.niksula.cs.hut.fi \
--cc=laughing@shared-source.org \
--cc=linux-kernel@vger.kernel.org \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®