I'm trying to track down a memory corruption bug in the atl1 network
device driver that is exposed only when operating with 4GB or more
memory. (NB: The driver uses a 32-bit DMA mask.) The bug is hit under
certain conditions whenever the network interface is commanded down.
The information provided here is against recent -git, but this problem
also afflicts the atl1 driver in kernels 2.6.2[345].y. Dmesg and
config attached.
I can reproduce the bug at will by simply using scp to copy a few
hundred megabytes from a remote host, then by executing 'ifconfig eth0
down'.
Here is the relevant console output when the bug is hit. I note that
the apparent corrupting data beginning at address 0xffff81010fcff402 is
actually a received ethernet frame. (The value 00:17:31:4e:9d:41 is
the MAC address of the local host, corresponding to the destination
address of a received frame.)
Can someone with more experience than me please take a look and give me
some advice or explain what might be happening here? (What may be
obvious to you is probably not obvious to me.)
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
BUG kmalloc-2048: Poison overwritten
---------------------------------------------------------------------------=
--
INFO: 0xffff81010fcff402-0xffff81010fcff9f9. First byte 0x0 instead of 0x6b
INFO: Allocated in dev_alloc_skb+0x16/0x2c age=3D9459 cpu=3D0 pid=3D2754
INFO: Freed in skb_release_data+0xa8/0xad age=3D9434 cpu=3D0 pid=3D2754
INFO: Slab 0xffffe20005d6f540 objects=3D15 used=3D0 fp=3D0xffff81010fcfb1b0=
flags=3D0x8000000000002082
INFO: Object 0xffff81010fcff3f0 @offset=3D29680 fp=3D0xffff81010fcfd2d0
Bytes b4 0xffff81010fcff3e0: 3b bd 00 00 01 00 00 00 5a 5a 5a 5a 5a 5a 5a =
5a ;=EF=BF=BD......ZZZZZZZZ
Object 0xffff81010fcff3f0: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b =
6b kkkkkkkkkkkkkkkk
Object 0xffff81010fcff400: 6b 6b 00 17 31 4e 9d 41 00 1b fc 95 53 da 08 =
00 kk..1N.A..=EF=BF=BD.S=EF=BF=BD..
Object 0xffff81010fcff410: 45 08 02 2c 83 1f 40 00 40 06 31 e1 c0 a8 01 =
03 E..,..@.@.1=EF=BF=BD=EF=BF=BD=EF=BF=BD..
Object 0xffff81010fcff420: c0 a8 01 70 58 b9 c0 18 93 5f c2 41 51 d2 a7 =
5d =EF=BF=BD=EF=BF=BD.pX=EF=BF=BD=EF=BF=BD.._=EF=BF=BDAQ=D2=A7]
Object 0xffff81010fcff430: 80 18 00 e8 ee ef 00 00 01 01 08 0a 00 e2 58 =
16 ...=EF=BF=BD=EF=BF=BD=EF=BF=BD.......=EF=BF=BDX.
Object 0xffff81010fcff440: 00 00 bd 38 4b 93 e4 7f 3e 8d 8c 2b 41 dc 9b =
36 ..=EF=BF=BD8K.=EF=BF=BD.>..+A=EF=BF=BD.6
Object 0xffff81010fcff450: 4d 9f b7 cf 2a 2c 07 06 d8 2f 23 de 5a 34 90 =
cb M.=EF=BF=BD=EF=BF=BD*,..=EF=BF=BD/#=EF=BF=BDZ4.
Object 0xffff81010fcff460: 6d d7 36 5b 2c 04 19 06 74 95 3f c5 3c c8 a5 =
9a m=EF=BF=BD6[,...t.?=EF=BF=BD<=C8=A5.
Redzone 0xffff81010fcffbf0: bb bb bb bb bb bb bb bb =
=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=
=20
Padding 0xffff81010fcffc30: 5a 5a 5a 5a 5a 5a 5a 5a =
ZZZZZZZZ =20
Pid: 2757, comm: ifconfig Not tainted 2.6.25 #1
Call Trace:
[<ffffffff8108cf02>] print_trailer+0x123/0x12c
[<ffffffff8108cfaf>] check_bytes_and_report+0xa4/0xcb
[<ffffffff8108d2de>] check_object+0xca/0x212
[<ffffffff8108d66d>] __free_slab+0x85/0xfd
[<ffffffff811e5cf2>] ? skb_release_data+0xa8/0xad
[<ffffffff8108d71d>] discard_slab+0x38/0x3a
[<ffffffff8108e0f2>] __slab_free+0xdb/0x2ac
[<ffffffff8108e3fa>] kfree+0xbc/0xcc
[<ffffffff811e5cf2>] ? skb_release_data+0xa8/0xad
[<ffffffff811e5cf2>] skb_release_data+0xa8/0xad
[<ffffffff811e63b3>] skb_release_all+0xc9/0xce
[<ffffffff811e5b4d>] __kfree_skb+0x11/0x78
[<ffffffff811e5bdb>] kfree_skb+0x27/0x29
[<ffffffffa00e63aa>] :atl1:atl1_clean_rx_ring+0x7e/0xe2
[<ffffffffa00e64d7>] :atl1:atl1_down+0xc9/0xce
[<ffffffffa00e8dcd>] :atl1:atl1_close+0x18/0x27
[<ffffffff811ebd43>] dev_close+0x57/0x72
[<ffffffff811eba47>] dev_change_flags+0xa8/0x164
[<ffffffff8122f380>] devinet_ioctl+0x26a/0x5f6
[<ffffffff8122fbad>] inet_ioctl+0x92/0xaa
[<ffffffff811df5f4>] sock_ioctl+0x1da/0x202
[<ffffffff8109f186>] vfs_ioctl+0x2a/0x77
[<ffffffff8109f435>] do_vfs_ioctl+0x262/0x27f
[<ffffffff8109f4a9>] sys_ioctl+0x57/0x7a
[<ffffffff8100bff7>] tracesys+0xd5/0xda
FIX kmalloc-2048: Restoring 0xffff81010fcff402-0xffff81010fcff9f9=3D0x6b
Thanks,
Jay
| Alan Cox | [PATCH 00/76] Queued TTY Patches |
| Linus Torvalds | Linux 2.6.27 |
| Eric W. Biederman | [PATCH] nfsd/nfs4state: Remove unnecessary daemonize call. |
| Artem Bityutskiy | [PATCH 10/44 take 2] [UBI] debug unit implementation |
git: | |
| Daniel Barkalow | Re: I don't want the .git directory next to my code. |
| Johannes Schindelin | Re: [PATCH] RFC: git lazy clone proof-of-concept |
| Johannes Schindelin | Re: [ANNOUNCE] GIT 1.5.4 |
| Johannes Schindelin | Re: git-diff on touched files: bug or feature? |
| Richard Stallman | Real men don't attack straw men |
| Juan Miscaro | When will OpenBSD support UTF8? |
| Stefan Beke | mail dovecot: pipe() failed: Too many open files |
| L. V. Lammert | Re: About Xen: maybe a reiterative question but .. |
| Michael Buesch | Re: Mark IPW2100 as BROKEN: Fatal interrupt. Scheduling firmware restart. |
| Johannes Berg | Re: mac80211 truesize bugs |
| Vitaliy Gusev | [TCP]: TCP_DEFER_ACCEPT causes leak sockets |
| Alexey Dobriyan | [PATCH 10/33] netns ct: per-netns /proc/net/nf_conntrack_expect |
| Shared swap partition | 35 minutes ago | Linux general |
| high memory | 1 day ago | Linux kernel |
| semaphore access speed | 1 day ago | Applications and Utilities |
| the kernel how to power off the machine | 2 days ago | Linux kernel |
| Easter Eggs in windows XP | 2 days ago | Windows |
| Root password | 2 days ago | Linux general |
| Where/when DNOTIFY is used? | 2 days ago | Linux kernel |
| How to convert Linux Kernel built-in module into a loadable module | 2 days ago | Linux kernel |
| Linux 2.6.24 and I/O schedulers | 2 days ago | Linux kernel |
| USB Driver -- Interrupt Polling -- A Little Help Please | 2 days ago | Linux general |
