2009-09-20 13:25:00

by SandeepKsinha

[permalink] [raw]
Subject: kernel BUG at fs/ext4/move_extent.c:1317!

Hi Akira/others,

I hit the following assert kernel BUG at fs/ext4/move_extent.c:1317!

The file I was playing around with was a simple holey file created like:

[/mnt]
[18:35:13 sinhas]$ dd if=/dev/zero of=/mnt/ext4image seek=1000 count=5
5+0 records in
5+0 records out
2560 bytes (2.6 kB) copied, 0.000146457 s, 17.5 MB/s

I am using v2.6.31. I have seen some patchsets a couple of days from
you. Does any of them address any similar problem.?

Are there any prerequisites for the userspace clients of
EXT4_IOC_MOVE_EXT? Apart from having the offsets equal for orig and
donor inode.
The userspace program (part of OHSM) invoking the ioctl could be found at:
http://ohsm.git.sourceforge.net/git/gitweb.cgi?p=ohsm/mkatiyar/.git;a=blob;f=userspace/relocate.c;h=ea0fa184ce4e692258a671cb0b9ace7a77b9fee9;hb=refs/heads/ohsm_next#l367

[22145.440009] ------------[ cut here ]------------
[22145.440019] kernel BUG at fs/ext4/move_extent.c:1317!
[22145.440031] invalid opcode: 0000 [#10] SMP
[22145.440044] last sysfs file:
/sys/devices/pci0000:00/0000:00:1c.1/0000:03:00.0/ieee80211/phy0/rfkill2/state
[22145.440059] Modules linked in: ohsm ext4 jbd2 crc16 af_packet ppdev
rfcomm bridge stp bnep sco l2cap ipv6 pci_slot sbs sbshc container
binfmt_misc autofs4sbp2 lp parport snd_hda_codec_analog snd_hda_intel
snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy
snd_seq_oss snd_seq_midi arc4 snd_rawmidi ecb joydev
snd_seq_midi_event snd_seq iwlagn snd_timer iwlcore snd_seq_device
mac80211 pcmcia snd btusb thinkpad_acpi led_class iTCO_wdt bluetooth
cfg80211 psmouse yenta_socket rsrc_nonstatic pcmcia_core soundcore
snd_page_alloc iTCO_vendor_support rfkill intel_agp agpgart pcspkr
serio_raw wmi evdev nvram video output ac battery button processor
ext3 jbd mbcache sr_mod cdrom sg ata_generic pata_acpi ohci1394
ata_piix ieee1394 ehci_hcd uhci_hcd e1000e usbcore linear thermal fan
fuse
[22145.440314]
[22145.440325] Pid: 7064, comm: ohsm Tainted: G D (2.6.31 #2) 7665AD6
[22145.440336] EIP: 0060:[<f8722829>] EFLAGS: 00210246 CPU: 0
[22145.440402] EIP is at ext4_move_extents+0x409/0x1600 [ext4]
[22145.440412] EAX: fff8307e EBX: 00000000 ECX: ec2a7e30 EDX: ffffffff
[22145.440425] ESI: f3b09ef0 EDI: 0000007d EBP: f3b09e88 ESP: f3b09da8
[22145.440435] DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[22145.440448] Process ohsm (pid: 7064, ti=f3b08000 task=f6083240
task.ti=f3b08000)
[22145.440456] Stack:
[22145.440463] f3b09e2c f8188be3 f3d59400 00000000 00000001 f3b09dd0
ecce4ac0 f3d59400
[22145.440489] <0> 00000000 f3d59402 00000000 f3f8ac68 f3b09e78
f6307400 f5a2a500 00000001
[22145.440515] <0> f61ecc00 f3b09e54 c149c640 c149c780 00000000
0007d9ff 0007d000 f6307400
[22145.440544] Call Trace:
[22145.440576] [<f8188be3>] ? do_get_write_access+0x763/0xab0 [jbd2]
[22145.440640] [<f8711707>] ? __ext4_handle_dirty_metadata+0x47/0x1a0 [ext4]
[22145.440664] [<c0375c83>] ? security_capable+0x33/0x50
[22145.440720] [<f86e1463>] ? ext4_ioctl+0x273/0xfa0 [ext4]
[22145.440779] [<f870b3c6>] ? ext4_fallocate+0x5f6/0x610 [ext4]
[22145.440835] [<f86e11f0>] ? ext4_ioctl+0x0/0xfa0 [ext4]
[22145.440852] [<c02ca2ed>] ? vfs_ioctl+0x3d/0x110
[22145.440866] [<c028660f>] ? page_referenced+0x1cf/0x320
[22145.440883] [<c02cab71>] ? do_vfs_ioctl+0x5f1/0xae0
[22145.440897] [<c02ace42>] ? do_fallocate+0x172/0x210
[22145.440913] [<c02cb106>] ? sys_ioctl+0xa6/0x100
[22145.440927] [<c0104bdc>] ? sysenter_do_call+0x12/0x28
[22145.440943] [<c028660f>] ? page_referenced+0x1cf/0x320
[22145.440951] Code: ba 75 f8 00 0f 0b eb fe 66 90 83 05 78 ba 75 f8
01 83 15 7c ba 75 f8 00 e9 56 fd ff ff 83 05 78 bd 75 f8 01 83 15 7c
bd 75 f8 00 <0f> 0beb fe 8b 5d c8 8b 43 20 8b 5d cc 3b 43 20 74 35 8b
45 c8
[22145.441098] EIP: [<f8722829>] ext4_move_extents+0x409/0x1600 [ext4]
SS:ESP 0068:f3b09da8
[22145.441193] ---[ end trace 67bb7ca457a8d69b ]---
[/mnt]
[18:35:13 sinhas]$


--
Regards,
Sandeep.






?To learn is to change. Education is a process that changes the learner.?


2009-09-21 01:22:32

by Peng Tao

[permalink] [raw]
Subject: Re: kernel BUG at fs/ext4/move_extent.c:1317!

Hi, Sandeep,

On Sun, Sep 20, 2009 at 9:25 PM, SandeepKsinha <[email protected]> wrote:
> Hi Akira/others,
>
> I hit the following assert kernel BUG at fs/ext4/move_extent.c:1317!
>
> The file I was playing around with was a simple holey file created like:
>
> [/mnt]
> [18:35:13 sinhas]$ dd if=/dev/zero of=/mnt/ext4image seek=1000 count=5
> 5+0 records in
> 5+0 records out
> 2560 bytes (2.6 kB) copied, 0.000146457 s, 17.5 MB/s
>
> I am using v2.6.31. I have seen some patchsets a couple of days from
> you. Does any of them address any similar problem.?
Yes. The BUG_ON is unnecessary and is removed in the latest mainline
kernel. Please refer to commit
daea696dbac0e33af3cfe304efbfb8d74e0effe6 for details.

>
> Are there any prerequisites for the userspace clients of
> EXT4_IOC_MOVE_EXT? Apart from having the offsets equal for orig and
> donor inode.
> The userspace program (part of OHSM) invoking the ioctl could be found at:
> http://ohsm.git.sourceforge.net/git/gitweb.cgi?p=ohsm/mkatiyar/.git;a=blob;f=userspace/relocate.c;h=ea0fa184ce4e692258a671cb0b9ace7a77b9fee9;hb=refs/heads/ohsm_next#l367
>
> [22145.440009] ------------[ cut here ]------------
> [22145.440019] kernel BUG at fs/ext4/move_extent.c:1317!
> [22145.440031] invalid opcode: 0000 [#10] SMP
> [22145.440044] last sysfs file:
> /sys/devices/pci0000:00/0000:00:1c.1/0000:03:00.0/ieee80211/phy0/rfkill2/state
> [22145.440059] Modules linked in: ohsm ext4 jbd2 crc16 af_packet ppdev
> rfcomm bridge stp bnep sco l2cap ipv6 pci_slot sbs sbshc container
> binfmt_misc autofs4sbp2 lp parport snd_hda_codec_analog snd_hda_intel
> snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy
> snd_seq_oss snd_seq_midi arc4 snd_rawmidi ecb joydev
> snd_seq_midi_event snd_seq iwlagn snd_timer iwlcore snd_seq_device
> mac80211 pcmcia snd btusb thinkpad_acpi led_class iTCO_wdt bluetooth
> cfg80211 psmouse yenta_socket rsrc_nonstatic pcmcia_core soundcore
> snd_page_alloc iTCO_vendor_support rfkill intel_agp agpgart pcspkr
> serio_raw wmi evdev nvram video output ac battery button processor
> ext3 jbd mbcache sr_mod cdrom sg ata_generic pata_acpi ohci1394
> ata_piix ieee1394 ehci_hcd uhci_hcd e1000e usbcore linear thermal fan
> fuse
> [22145.440314]
> [22145.440325] Pid: 7064, comm: ohsm Tainted: G      D    (2.6.31 #2) 7665AD6
> [22145.440336] EIP: 0060:[<f8722829>] EFLAGS: 00210246 CPU: 0
> [22145.440402] EIP is at ext4_move_extents+0x409/0x1600 [ext4]
> [22145.440412] EAX: fff8307e EBX: 00000000 ECX: ec2a7e30 EDX: ffffffff
> [22145.440425] ESI: f3b09ef0 EDI: 0000007d EBP: f3b09e88 ESP: f3b09da8
> [22145.440435]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
> [22145.440448] Process ohsm (pid: 7064, ti=f3b08000 task=f6083240
> task.ti=f3b08000)
> [22145.440456] Stack:
> [22145.440463]  f3b09e2c f8188be3 f3d59400 00000000 00000001 f3b09dd0
> ecce4ac0 f3d59400
> [22145.440489] <0> 00000000 f3d59402 00000000 f3f8ac68 f3b09e78
> f6307400 f5a2a500 00000001
> [22145.440515] <0> f61ecc00 f3b09e54 c149c640 c149c780 00000000
> 0007d9ff 0007d000 f6307400
> [22145.440544] Call Trace:
> [22145.440576]  [<f8188be3>] ? do_get_write_access+0x763/0xab0 [jbd2]
> [22145.440640]  [<f8711707>] ? __ext4_handle_dirty_metadata+0x47/0x1a0 [ext4]
> [22145.440664]  [<c0375c83>] ? security_capable+0x33/0x50
> [22145.440720]  [<f86e1463>] ? ext4_ioctl+0x273/0xfa0 [ext4]
> [22145.440779]  [<f870b3c6>] ? ext4_fallocate+0x5f6/0x610 [ext4]
> [22145.440835]  [<f86e11f0>] ? ext4_ioctl+0x0/0xfa0 [ext4]
> [22145.440852]  [<c02ca2ed>] ? vfs_ioctl+0x3d/0x110
> [22145.440866]  [<c028660f>] ? page_referenced+0x1cf/0x320
> [22145.440883]  [<c02cab71>] ? do_vfs_ioctl+0x5f1/0xae0
> [22145.440897]  [<c02ace42>] ? do_fallocate+0x172/0x210
> [22145.440913]  [<c02cb106>] ? sys_ioctl+0xa6/0x100
> [22145.440927]  [<c0104bdc>] ? sysenter_do_call+0x12/0x28
> [22145.440943]  [<c028660f>] ? page_referenced+0x1cf/0x320
> [22145.440951] Code: ba 75 f8 00 0f 0b eb fe 66 90 83 05 78 ba 75 f8
> 01 83 15 7c ba 75 f8 00 e9 56 fd ff ff 83 05 78 bd 75 f8 01 83 15 7c
> bd 75 f8 00 <0f> 0beb fe 8b 5d c8 8b 43 20 8b 5d cc 3b 43 20 74 35 8b
> 45 c8
> [22145.441098] EIP: [<f8722829>] ext4_move_extents+0x409/0x1600 [ext4]
> SS:ESP 0068:f3b09da8
> [22145.441193] ---[ end trace 67bb7ca457a8d69b ]---
> [/mnt]
> [18:35:13 sinhas]$
>
>
> --
> Regards,
> Sandeep.
>
>
>
>
>
>
> “To learn is to change. Education is a process that changes the learner.”
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to [email protected]
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>



--
Cheers,
Peng Tao
State Key Laboratory of Networking and Switching Technology
Beijing Univ. of Posts and Telecoms.

2009-09-21 19:06:11

by Greg Freemyer

[permalink] [raw]
Subject: Re: kernel BUG at fs/ext4/move_extent.c:1317!

resend - I inadvertently was not in plain text mode before.

On Sun, Sep 20, 2009 at 9:22 PM, Peng Tao <[email protected]> wrote:
> Hi, Sandeep,
>
> On Sun, Sep 20, 2009 at 9:25 PM, SandeepKsinha <[email protected]> wrote:
>> Hi Akira/others,
>>
>> I hit the following assert kernel BUG at fs/ext4/move_extent.c:1317!
>>
>> The file I was playing around with was a simple holey file created like:
>>
>> [/mnt]
>> [18:35:13 sinhas]$ dd if=/dev/zero of=/mnt/ext4image seek=1000 count=5
>> 5+0 records in
>> 5+0 records out
>> 2560 bytes (2.6 kB) copied, 0.000146457 s, 17.5 MB/s
>>
>> I am using v2.6.31. I have seen some patchsets a couple of days from
>> you. Does any of them address any similar problem.?
> Yes. The BUG_ON is unnecessary and is removed in the latest mainline
> kernel. Please refer to commit
> daea696dbac0e33af3cfe304efbfb8d74e0effe6 for details.
<snip>
> --
> Cheers,
> Peng Tao

Ted,

I think the few small ext4_move_ext() / e4defrag related fixes for
sparse files to work should get pushed to stable. I don't know if
this is the only one or if there are others.

Greg