* 2.6.0-test9: access beyond end of device
@ 2003-10-29 10:12 Alexander V. Lukyanov
2003-10-29 20:40 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Alexander V. Lukyanov @ 2003-10-29 10:12 UTC (permalink / raw)
To: linux-kernel
I have tried to run 2.6.0-test9 and got this error very quickly.
Details:
heavily loaded squid server with two ext3 filesystems for cache on
two IC35L040AVVN07-0 40GiB hard disks (ibm), TCQ enabled.
Oct 29 10:52:28 mars kernel: attempt to access beyond end of device
Oct 29 10:52:28 mars kernel: hda4: rw=0, want=4241606720, limit=77256585
Oct 29 10:52:28 mars kernel: attempt to access beyond end of device
Oct 29 10:52:28 mars kernel: hda4: rw=0, want=4241606720, limit=77256585
Oct 29 10:52:33 mars kernel: EXT3-fs error (device hda4): ext3_free_blocks: Freeing blocks not in datazone - block = 1067071751, count = 1
Oct 29 10:52:33 mars kernel: Aborting journal on device hda4.
--
Alexander.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.6.0-test9: access beyond end of device
2003-10-29 10:12 2.6.0-test9: access beyond end of device Alexander V. Lukyanov
@ 2003-10-29 20:40 ` Andrew Morton
2003-10-30 9:22 ` Alexander V. Lukyanov
0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2003-10-29 20:40 UTC (permalink / raw)
To: Alexander V. Lukyanov; +Cc: linux-kernel
"Alexander V. Lukyanov" <lav@netis.ru> wrote:
>
> I have tried to run 2.6.0-test9 and got this error very quickly.
>
> Details:
> heavily loaded squid server with two ext3 filesystems for cache on
> two IC35L040AVVN07-0 40GiB hard disks (ibm), TCQ enabled.
Please force an fsck against those partitions, then see if it is repeatable
with TCQ disabled.
> Oct 29 10:52:28 mars kernel: attempt to access beyond end of device
> Oct 29 10:52:28 mars kernel: hda4: rw=0, want=4241606720, limit=77256585
> Oct 29 10:52:28 mars kernel: attempt to access beyond end of device
> Oct 29 10:52:28 mars kernel: hda4: rw=0, want=4241606720, limit=77256585
> Oct 29 10:52:33 mars kernel: EXT3-fs error (device hda4): ext3_free_blocks: Freeing blocks not in datazone - block = 1067071751, count = 1
> Oct 29 10:52:33 mars kernel: Aborting journal on device hda4.
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.6.0-test9: access beyond end of device
2003-10-29 20:40 ` Andrew Morton
@ 2003-10-30 9:22 ` Alexander V. Lukyanov
2003-10-30 9:39 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Alexander V. Lukyanov @ 2003-10-30 9:22 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel
[-- Attachment #1: Type: text/plain, Size: 647 bytes --]
On Wed, Oct 29, 2003 at 12:40:03PM -0800, Andrew Morton wrote:
> > heavily loaded squid server with two ext3 filesystems for cache on
> > two IC35L040AVVN07-0 40GiB hard disks (ibm), TCQ enabled.
>
> Please force an fsck against those partitions, then see if it is repeatable
> with TCQ disabled.
Ok, without TCQ it worked a little longer, but produced another error.
Now it can be related to network interface configuration: it has two
100Mb ethernets, one of which has mtu 576. kernel 2.4.22 used to produce
warning "sending pkt_too_big to myself" from time to time.
--
Alexander. | http://www.yars.free.net/~lav/
[-- Attachment #2: oops1 --]
[-- Type: text/plain, Size: 4333 bytes --]
Oct 30 11:41:52 mars kernel: BUG: dst underflow 0: c01d15f1
Oct 30 11:41:52 mars kernel: recvmsg bug: copied F7C6B8D0 seq F7C6BE78
Oct 30 11:41:52 mars last message repeated 31 times
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!sk->sk_forward_alloc)
Oct 30 11:41:52 mars kernel: recvmsg bug: copied F7C6B8D0 seq F7C6BE78
Oct 30 11:41:52 mars last message repeated 4 times
Oct 30 11:41:52 mars Oct 30 11:41:52
Oct 30 11:41:52 mars P╨йVЕ
Oct 30 11:41:52 mars ^I|Ц{ ^H
Oct 30 11:41:52 mars E
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!atomic_read(&sk->sk_rmem_alloc)) failed at net/ipv4/af_inet.c (154)
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!sk->sk_forward_alloc) failed at net/ipv4/af_inet.c (157)
Oct 30 11:42:00 mars kernel: process `named' is using obsolete setsockopt SO_BSDCOMPAT
Oct 30 11:42:21 mars last message repeated 4 times
Oct 30 11:42:23 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:42:23 mars kernel: recvmsg bug: copied 38B12037 seq 38B1205F
Oct 30 11:42:39 mars kernel: process `named' is using obsolete setsockopt SO_BSDCOMPAT
Oct 30 11:42:41 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:42:41 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:43:18 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:43:18 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:43:18 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:43:18 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
...lots and lots of the same KERNEL and recvmsg messages.
Oct 30 11:47:45 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:47:45 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:47:46 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:47:46 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:48:06 mars squid[579]: Squid Parent: child process 582 exited due to signal 11
Oct 30 11:48:06 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:48:06 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:48:06 mars kernel: Unable to handle kernel paging request at virtual address 005e1a00
Oct 30 11:48:06 mars kernel: printing eip:
Oct 30 11:48:06 mars kernel: c01d14a0
Oct 30 11:48:06 mars kernel: *pde = 00000000
Oct 30 11:48:06 mars kernel: Oops: 0000 [#1]
Oct 30 11:48:06 mars kernel: CPU: 0
Oct 30 11:48:06 mars kernel: EIP: 0060:[<c01d14a0>] Not tainted
Oct 30 11:48:06 mars kernel: EFLAGS: 00010207
Oct 30 11:48:06 mars kernel: EIP is at skb_release_data+0x37/0x9f
Oct 30 11:48:06 mars kernel: eax: c4e82480 ebx: d022f080 ecx: c4e82480 edx: 005e1a00
Oct 30 11:48:06 mars kernel: esi: 00000000 edi: 00000001 ebp: 00000000 esp: e1163eb0
Oct 30 11:48:06 mars kernel: ds: 007b es: 007b ss: 0068
Oct 30 11:48:06 mars kernel: Process squid (pid: 582, threadinfo=e1162000 task=e0ef26a0)
Oct 30 11:48:06 mars kernel: Stack: 00000000 d022f080 dc10cb80 c01d151b d022f080 00000000 00000000 c01d15b6
Oct 30 11:48:06 mars kernel: d022f080 00000000 00000001 dc10cbc4 c01f45f1 d022f080 00000000 266e2ecf
Oct 30 11:48:06 mars kernel: 00000000 00000000 e0ef26a0 c01f5434 d9fb0100 00000000 00000000 c39889a4
Oct 30 11:48:06 mars kernel: Call Trace:
Oct 30 11:48:06 mars kernel: [<c01d151b>] kfree_skbmem+0x13/0x2c
Oct 30 11:48:06 mars kernel: [<c01d15b6>] __kfree_skb+0x82/0xfa
Oct 30 11:48:06 mars kernel: [<c01f45f1>] tcp_close+0x98/0x6eb
Oct 30 11:48:06 mars kernel: [<c01f5434>] tcp_setsockopt+0xfe/0x68a
Oct 30 11:48:06 mars kernel: [<c0212232>] inet_release+0x53/0x61
Oct 30 11:48:06 mars kernel: [<c01cdefd>] sock_release+0x5d/0x66
Oct 30 11:48:06 mars kernel: [<c01ce74b>] sock_close+0x36/0x4e
Oct 30 11:48:06 mars kernel: [<c0141d50>] __fput+0xb3/0xc5
Oct 30 11:48:06 mars kernel: [<c014065c>] filp_close+0x59/0x86
Oct 30 11:48:06 mars kernel: [<c01406d9>] sys_close+0x50/0x5f
Oct 30 11:48:06 mars kernel: [<c0108e41>] sysenter_past_esp+0x52/0x71
Oct 30 11:48:06 mars kernel:
Oct 30 11:48:06 mars kernel: Code: 8b 02 a9 00 08 00 00 75 17 8b 42 04 85 c0 74 48 ff 4a 04 0f
[-- Attachment #3: oops-decoded --]
[-- Type: text/plain, Size: 3125 bytes --]
ksymoops 2.4.4 on i686 2.4.22. Options used
-V (default)
-k /var/log/ksyms.1 (specified)
-l /proc/modules (default)
-o /lib/modules/2.6.0-test9 (specified)
-m /1/linux-2.6.0-test9/System.map (specified)
Warning (read_ksyms): no kernel symbols in ksyms, is /var/log/ksyms.1 a valid ksyms file?
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Oct 30 11:48:06 mars kernel: Unable to handle kernel paging request at virtual address 005e1a00
Oct 30 11:48:06 mars kernel: c01d14a0
Oct 30 11:48:06 mars kernel: *pde = 00000000
Oct 30 11:48:06 mars kernel: Oops: 0000 [#1]
Oct 30 11:48:06 mars kernel: CPU: 0
Oct 30 11:48:06 mars kernel: EIP: 0060:[<c01d14a0>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
Oct 30 11:48:06 mars kernel: EFLAGS: 00010207
Oct 30 11:48:06 mars kernel: eax: c4e82480 ebx: d022f080 ecx: c4e82480 edx: 005e1a00
Oct 30 11:48:06 mars kernel: esi: 00000000 edi: 00000001 ebp: 00000000 esp: e1163eb0
Oct 30 11:48:06 mars kernel: ds: 007b es: 007b ss: 0068
Oct 30 11:48:06 mars kernel: Stack: 00000000 d022f080 dc10cb80 c01d151b d022f080 00000000 00000000 c01d15b6
Oct 30 11:48:06 mars kernel: d022f080 00000000 00000001 dc10cbc4 c01f45f1 d022f080 00000000 266e2ecf
Oct 30 11:48:06 mars kernel: 00000000 00000000 e0ef26a0 c01f5434 d9fb0100 00000000 00000000 c39889a4
Oct 30 11:48:06 mars kernel: [<c01d151b>] kfree_skbmem+0x13/0x2c
Oct 30 11:48:06 mars kernel: [<c01d15b6>] __kfree_skb+0x82/0xfa
Oct 30 11:48:06 mars kernel: [<c01f45f1>] tcp_close+0x98/0x6eb
Oct 30 11:48:06 mars kernel: [<c01f5434>] tcp_setsockopt+0xfe/0x68a
Oct 30 11:48:06 mars kernel: [<c0212232>] inet_release+0x53/0x61
Oct 30 11:48:06 mars kernel: [<c01cdefd>] sock_release+0x5d/0x66
Oct 30 11:48:06 mars kernel: [<c01ce74b>] sock_close+0x36/0x4e
Oct 30 11:48:06 mars kernel: [<c0141d50>] __fput+0xb3/0xc5
Oct 30 11:48:06 mars kernel: [<c014065c>] filp_close+0x59/0x86
Oct 30 11:48:06 mars kernel: [<c01406d9>] sys_close+0x50/0x5f
Oct 30 11:48:06 mars kernel: [<c0108e41>] sysenter_past_esp+0x52/0x71
Oct 30 11:48:06 mars kernel: Code: 8b 02 a9 00 08 00 00 75 17 8b 42 04 85 c0 74 48 ff 4a 04 0f
>>EIP; c01d14a0 <skb_release_data+37/9f> <=====
Code; c01d14a0 <skb_release_data+37/9f>
00000000 <_EIP>:
Code; c01d14a0 <skb_release_data+37/9f> <=====
0: 8b 02 mov (%edx),%eax <=====
Code; c01d14a2 <skb_release_data+39/9f>
2: a9 00 08 00 00 test $0x800,%eax
Code; c01d14a7 <skb_release_data+3e/9f>
7: 75 17 jne 20 <_EIP+0x20>
Code; c01d14a9 <skb_release_data+40/9f>
9: 8b 42 04 mov 0x4(%edx),%eax
Code; c01d14ac <skb_release_data+43/9f>
c: 85 c0 test %eax,%eax
Code; c01d14ae <skb_release_data+45/9f>
e: 74 48 je 58 <_EIP+0x58>
Code; c01d14b0 <skb_release_data+47/9f>
10: ff 4a 04 decl 0x4(%edx)
Code; c01d14b3 <skb_release_data+4a/9f>
13: 0f 00 00 sldtl (%eax)
1 warning issued. Results may not be reliable.
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: 2.6.0-test9: access beyond end of device
2003-10-30 9:22 ` Alexander V. Lukyanov
@ 2003-10-30 9:39 ` Andrew Morton
2003-10-30 11:02 ` Alexander V. Lukyanov
0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2003-10-30 9:39 UTC (permalink / raw)
To: Alexander V. Lukyanov; +Cc: linux-kernel, netdev
"Alexander V. Lukyanov" <lav@netis.ru> wrote:
>
> On Wed, Oct 29, 2003 at 12:40:03PM -0800, Andrew Morton wrote:
> > > heavily loaded squid server with two ext3 filesystems for cache on
> > > two IC35L040AVVN07-0 40GiB hard disks (ibm), TCQ enabled.
> >
> > Please force an fsck against those partitions, then see if it is repeatable
> > with TCQ disabled.
>
> Ok, without TCQ it worked a little longer, but produced another error.
> Now it can be related to network interface configuration: it has two
> 100Mb ethernets, one of which has mtu 576. kernel 2.4.22 used to produce
> warning "sending pkt_too_big to myself" from time to time.
>
Congratulations, you broke the 2.6 networking code!
Please send a full report, including your .config and a description of how
you have set everything up to netdev@oss.sgi.com.
Is it repeatable?
Oct 30 11:41:52 mars kernel: BUG: dst underflow 0: c01d15f1
Oct 30 11:41:52 mars kernel: recvmsg bug: copied F7C6B8D0 seq F7C6BE78
Oct 30 11:41:52 mars last message repeated 31 times
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!sk->sk_forward_alloc)
Oct 30 11:41:52 mars kernel: recvmsg bug: copied F7C6B8D0 seq F7C6BE78
Oct 30 11:41:52 mars last message repeated 4 times
Oct 30 11:41:52 mars Oct 30 11:41:52
Oct 30 11:41:52 mars E
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!atomic_read(&sk->sk_rmem_alloc)) failed at net/ipv4/af_inet.c (154)
Oct 30 11:41:52 mars kernel: KERNEL: assertion (!sk->sk_forward_alloc) failed at net/ipv4/af_inet.c (157)
Oct 30 11:42:00 mars kernel: process `named' is using obsolete setsockopt SO_BSDCOMPAT
Oct 30 11:42:21 mars last message repeated 4 times
Oct 30 11:42:23 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:42:23 mars kernel: recvmsg bug: copied 38B12037 seq 38B1205F
Oct 30 11:42:39 mars kernel: process `named' is using obsolete setsockopt SO_BSDCOMPAT
Oct 30 11:42:41 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:42:41 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:43:18 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:43:18 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:43:18 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:43:18 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
...lots and lots of the same KERNEL and recvmsg messages.
Oct 30 11:47:45 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:47:45 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:47:46 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:47:46 mars kernel: recvmsg bug: copied DEEFC9B5 seq DEEFCA39
Oct 30 11:48:06 mars squid[579]: Squid Parent: child process 582 exited due to signal 11
Oct 30 11:48:06 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:48:06 mars kernel: KERNEL: assertion (flags & MSG_PEEK) failed at net/ipv4/tcp.c (1565)
Oct 30 11:48:06 mars kernel: Unable to handle kernel paging request at virtual address 005e1a00
Oct 30 11:48:06 mars kernel: printing eip:
Oct 30 11:48:06 mars kernel: c01d14a0
Oct 30 11:48:06 mars kernel: *pde = 00000000
Oct 30 11:48:06 mars kernel: Oops: 0000 [#1]
Oct 30 11:48:06 mars kernel: CPU: 0
Oct 30 11:48:06 mars kernel: EIP: 0060:[<c01d14a0>] Not tainted
Oct 30 11:48:06 mars kernel: EFLAGS: 00010207
Oct 30 11:48:06 mars kernel: EIP is at skb_release_data+0x37/0x9f
Oct 30 11:48:06 mars kernel: eax: c4e82480 ebx: d022f080 ecx: c4e82480 edx: 005e1a00
Oct 30 11:48:06 mars kernel: esi: 00000000 edi: 00000001 ebp: 00000000 esp: e1163eb0
Oct 30 11:48:06 mars kernel: ds: 007b es: 007b ss: 0068
Oct 30 11:48:06 mars kernel: Process squid (pid: 582, threadinfo=e1162000 task=e0ef26a0)
Oct 30 11:48:06 mars kernel: Stack: 00000000 d022f080 dc10cb80 c01d151b d022f080 00000000 00000000 c01d15b6
Oct 30 11:48:06 mars kernel: d022f080 00000000 00000001 dc10cbc4 c01f45f1 d022f080 00000000 266e2ecf
Oct 30 11:48:06 mars kernel: 00000000 00000000 e0ef26a0 c01f5434 d9fb0100 00000000 00000000 c39889a4
Oct 30 11:48:06 mars kernel: Call Trace:
Oct 30 11:48:06 mars kernel: [<c01d151b>] kfree_skbmem+0x13/0x2c
Oct 30 11:48:06 mars kernel: [<c01d15b6>] __kfree_skb+0x82/0xfa
Oct 30 11:48:06 mars kernel: [<c01f45f1>] tcp_close+0x98/0x6eb
Oct 30 11:48:06 mars kernel: [<c01f5434>] tcp_setsockopt+0xfe/0x68a
Oct 30 11:48:06 mars kernel: [<c0212232>] inet_release+0x53/0x61
Oct 30 11:48:06 mars kernel: [<c01cdefd>] sock_release+0x5d/0x66
Oct 30 11:48:06 mars kernel: [<c01ce74b>] sock_close+0x36/0x4e
Oct 30 11:48:06 mars kernel: [<c0141d50>] __fput+0xb3/0xc5
Oct 30 11:48:06 mars kernel: [<c014065c>] filp_close+0x59/0x86
Oct 30 11:48:06 mars kernel: [<c01406d9>] sys_close+0x50/0x5f
Oct 30 11:48:06 mars kernel: [<c0108e41>] sysenter_past_esp+0x52/0x71
Oct 30 11:48:06 mars kernel:
Oct 30 11:48:06 mars kernel: Code: 8b 02 a9 00 08 00 00 75 17 8b 42 04 85 c0 74 48 ff 4a 04 0f
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.6.0-test9: access beyond end of device
2003-10-30 9:39 ` Andrew Morton
@ 2003-10-30 11:02 ` Alexander V. Lukyanov
0 siblings, 0 replies; 5+ messages in thread
From: Alexander V. Lukyanov @ 2003-10-30 11:02 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel, netdev
On Thu, Oct 30, 2003 at 01:39:04AM -0800, Andrew Morton wrote:
> Congratulations, you broke the 2.6 networking code!
Probably not. I have found a problem in ip_wccp module which I have loaded.
It used ip_rcv while netif_rx is the proper function to use.
After I have fixed that, linux-2.6.0-test9 works well. Sorry for confusion.
I'll report if I find other problems. Thanks!
--
Alexander.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2003-10-30 11:03 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-10-29 10:12 2.6.0-test9: access beyond end of device Alexander V. Lukyanov
2003-10-29 20:40 ` Andrew Morton
2003-10-30 9:22 ` Alexander V. Lukyanov
2003-10-30 9:39 ` Andrew Morton
2003-10-30 11:02 ` Alexander V. Lukyanov
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).