Testing a custom kernel (2.6.39) with native reiser4 support

Testing a custom kernel (2.6.39) with native reiser4 support

[Sandro Souza]

Hello my friends.

I made a new custom distro based on debian squeeze, but with 2.6.39
kernel, patched with reiser4
(http://downloads.sourceforge.net/project/reiser4/reiser4-for-linux-2.6/reiser4-for-2.6.39-1.patch.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Freiser4%2Ffiles%2Freiser4-for-linux-2.6%2F&ts=1337035240&use_mirror=ufpr).

Trying to copy a folder from a reiser3 partition to a reiser4
partition, I got error messages.

Testing my custom kernel with "preemption model" in "Preemptible
Kernel", I got these messages:

[  286.945598] ------------[ cut here ]------------
[  286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
[  286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
[  286.945626] last sysfs file:
/sys/devices/pci0000:00/0000:00:0b.0/usb1/1-1/1-1:1.0/host2/target2:0:0/2:0:0:0/block/sdc/sdc1/uevent
[  286.945634] CPU 0
[  286.945641] Modules linked in: mperf cpufreq_powersave
cpufreq_stats cpufreq_userspace cpufreq_conservative ppdev lp
bluetooth rfkill binfmt_misc fuse loop snd_intel8x0 snd_ac97_codec
ac97_bus snd_pcm snd_seq snd_timer snd_seq_device processor tpm_tis
tpm parport_pc parport thermal_sys snd tpm_bios psmouse button
soundcore snd_page_alloc ac i2c_piix4 i2c_core serio_raw pcspkr evdev
joydev usbhid hid ide_cd_mod cdrom sd_mod ata_generic crc_t10dif
pata_acpi ata_piix usb_storage ohci_hcd ide_pci_generic ahci libahci
ehci_hcd libata piix scsi_mod usbcore e1000 ide_core [last unloaded:
scsi_wait_scan]
[  286.945676]
[  286.945683] Pid: 2433, comm: nautilus Not tainted 2.6.39-reiser4 #1
innotek GmbH VirtualBox
[  286.945691] RIP: 0010:[<ffffffff81178fae>]  [<ffffffff81178fae>]
grabbed2flush_reserved_nolock+0x33/0x7b
[  286.945704] RSP: 0018:ffff880014219b38  EFLAGS: 00010297
[  286.945711] RAX: ffff88003b3df900 RBX: ffff88002fe75000 RCX: 0000000000112ef7
[  286.945718] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff88003d3801c0
[  286.945724] RBP: 0000000000000001 R08: 0000000000000000 R09: 000000000000000b
[  286.945730] R10: dead000000100100 R11: ffff880014219c50 R12: ffff88003d385500
[  286.945737] R13: ffff8800119bb000 R14: ffff880014219c38 R15: 0000000000001000
[  286.945744] FS:  00007f126661d700(0000) GS:ffff88003fc00000(0000)
knlGS:0000000000000000
[  286.945750] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[  286.945756] CR2: 00007fba4a3df010 CR3: 000000003be2b000 CR4: 00000000000006f0
[  286.945766] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  286.945772] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[  286.945778] Process nautilus (pid: 2433, threadinfo
ffff880014218000, task ffff88001406e940)
[  286.945783] Stack:
[  286.945788]  0000000000000000 ffff8800119bb000 ffff88003d3801c0
ffffffff8117b484
[  286.945795]  ffff8800119bb050 ffff8800119bb000 0000000000000000
ffffffff8117b529
[  286.945802]  0000000000000000 ffff88003d385608 0000000000000000
ffffffff8118e00d
[  286.945808] Call Trace:
[  286.945817]  [<ffffffff8117b484>] ? jnode_make_dirty_locked+0x112/0x18f
[  286.945824]  [<ffffffff8117b529>] ? znode_make_dirty+0x28/0x87
[  286.945831]  [<ffffffff8118e00d>] ? update_sd+0x344/0x3a4
[  286.945838]  [<ffffffff81387a46>] ? sub_preempt_count+0x83/0x94
[  286.945845]  [<ffffffff8118e096>] ? write_sd_by_inode_common+0x29/0x92
[  286.945852]  [<ffffffff8103cb67>] ? get_parent_ip+0x9/0x1b
[  286.945858]  [<ffffffff8118512b>] ? reiser4_dirty_inode+0x19/0x73
[  286.945865]  [<ffffffff810edecb>] ? T.1132+0x12/0x2e
[  286.945872]  [<ffffffff8104b288>] ? current_fs_time+0x1e/0x24
[  286.945878]  [<ffffffff81112ddb>] ? __mark_inode_dirty+0x28/0x1c8
[  286.945885]  [<ffffffff81108169>] ? file_update_time+0xf7/0x126
[  286.945891]  [<ffffffff811924ee>] ? reiser4_write_end_careful+0x147/0x184
[  286.945897]  [<ffffffff81115083>] ? pipe_to_file+0x152/0x161
[  286.945903]  [<ffffffff81387af5>] ? add_preempt_count+0x9e/0xa0
[  286.945909]  [<ffffffff81114f31>] ? generic_file_splice_write+0x133/0x133
[  286.945914]  [<ffffffff811143cb>] ? splice_from_pipe_feed+0x6d/0xed
[  286.945920]  [<ffffffff81114eb3>] ? generic_file_splice_write+0xb5/0x133
[  286.945933]  [<ffffffff811162a6>] ? sys_splice+0x2f8/0x3d2
[  286.945939]  [<ffffffff8138a7d2>] ? system_call_fastpath+0x16/0x1b
[  286.945944] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 38 05 00
00 48 8b 90 a0 00 00 00 48 8b 9a 88 02 00 00 48 8b 90 b0 00 00 00 48
39 f2 73 04 <0f> 0b eb fe 48 29 f2 48 89 90 b0 00 00 00 8b 07 89 c2 c1
fa 10
[  286.945968] RIP  [<ffffffff81178fae>] grabbed2flush_reserved_nolock+0x33/0x7b
[  286.945974]  RSP <ffff880014219b38>
[  286.945979] ---[ end trace cd39088cc6246a13 ]---

Testing my custom kernel with "preemption model" in "Voluntary  Kernel
Preemption", I got these messages:

[  234.212538] ------------[ cut here ]------------
[  234.212549] kernel BUG at fs/reiser4/block_alloc.c:151!
[  234.212551] invalid opcode: 0000 [#1] SMP
[  234.212553] last sysfs file:
/sys/devices/pci0000:00/0000:00:0b.0/usb1/1-1/1-1:1.0/host2/target2:0:0/2:0:0:0/block/sdc/sdc1/uevent
[  234.212556] CPU 0
[  234.212557] Modules linked in: mperf cpufreq_powersave
cpufreq_stats cpufreq_userspace cpufreq_conservative ppdev lp
bluetooth rfkill binfmt_misc fuse loop snd_intel8x0 i2c_piix4
snd_ac97_codec psmouse joydev processor thermal_sys evdev i2c_core
parport_pc parport tpm_tis pcspkr tpm serio_raw tpm_bios ac ac97_bus
snd_pcm snd_seq snd_timer snd_seq_device snd button soundcore
snd_page_alloc usbhid hid ide_cd_mod cdrom ata_generic sd_mod
pata_acpi ata_piix crc_t10dif usb_storage ide_pci_generic ohci_hcd
ahci libahci ehci_hcd libata piix scsi_mod usbcore e1000 ide_core
[last unloaded: scsi_wait_scan]
[  234.212581]
[  234.212583] Pid: 2530, comm: nautilus Not tainted 2.6.39-reiser4 #1
innotek GmbH VirtualBox
[  234.212585] RIP: 0010:[<ffffffff81173400>]  [<ffffffff81173400>]
grabbed2flush_reserved_nolock+0x33/0x87
[  234.212591] RSP: 0018:ffff88003da41b68  EFLAGS: 00010297
[  234.212592] RAX: ffff88003c939500 RBX: ffff88003bcf9800 RCX: ffff880007e49d80
[  234.212594] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8800215e41c0
[  234.212595] RBP: 0000000000000001 R08: 0000000000000000 R09: 000000000000000b
[  234.212596] R10: ffff88002d3b96b0 R11: ffff88003da41c80 R12: ffff880021693500
[  234.212598] R13: ffff88001056b900 R14: ffff88003da41c68 R15: 0000000000001000
[  234.212605] FS:  00007f94bc7cc700(0000) GS:ffff88003fc00000(0000)
knlGS:0000000000000000
[  234.212607] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[  234.212608] CR2: 0000000000428060 CR3: 000000003c197000 CR4: 00000000000006f0
[  234.212614] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  234.212615] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[  234.212617] Process nautilus (pid: 2530, threadinfo
ffff88003da40000, task ffff880026095b00)
[  234.212618] Stack:
[  234.212619]  0000000000000000 ffff88001056b900 ffff8800215e41c0
ffffffff81175c90
[  234.212621]  0000000000001000 ffff88001056b900 0000000000000000
ffffffff81175d40
[  234.212624]  0000000000000000 ffff880021693608 0000000000000000
ffffffff81188945
[  234.212626] Call Trace:
[  234.212630]  [<ffffffff81175c90>] ? jnode_make_dirty_locked+0x112/0x19a
[  234.212632]  [<ffffffff81175d40>] ? znode_make_dirty+0x28/0xa1
[  234.212634]  [<ffffffff81188945>] ? update_sd+0x368/0x3d9
[  234.212637]  [<ffffffff81171f26>] ? insert_by_coord+0xef/0x101
[  234.212641]  [<ffffffff81199f4d>] ? ctail_insert_unprepped_cluster+0xe3/0xf2
[  234.212643]  [<ffffffff811889df>] ? write_sd_by_inode_common+0x29/0x92
[  234.212645]  [<ffffffff81175a1d>] ? jnode_get_atom+0x34/0xac
[  234.212648]  [<ffffffff8117fab1>] ? reiser4_dirty_inode+0x19/0x73
[  234.212652]  [<ffffffff81049a2c>] ? current_fs_time+0x1e/0x24
[  234.212656]  [<ffffffff810ea179>] ? kfree+0xa8/0xcc
[  234.212659]  [<ffffffff8110d9ee>] ? __mark_inode_dirty+0x22/0x1a4
[  234.212671]  [<ffffffff81103254>] ? file_update_time+0xea/0x115
[  234.212674]  [<ffffffff8118ce02>] ? reiser4_write_end_careful+0x147/0x184
[  234.212677]  [<ffffffff8110fa8c>] ? pipe_to_file+0x12f/0x13e
[  234.212679]  [<ffffffff8110f95d>] ? generic_file_splice_write+0x133/0x133
[  234.212681]  [<ffffffff8110eebb>] ? splice_from_pipe_feed+0x6d/0xed
[  234.212684]  [<ffffffff8110f8df>] ? generic_file_splice_write+0xb5/0x133
[  234.212686]  [<ffffffff811100c7>] ? sys_splice+0x2f8/0x3db
[  234.212689]  [<ffffffff81382412>] ? system_call_fastpath+0x16/0x1b
[  234.212690] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 18 05 00
00 48 8b 90 a0 00 00 00 48 8b 9a 88 02 00 00 48 8b 90 b0 00 00 00 48
39 f2 73 04 <0f> 0b eb fe 48 29 f2 48 89 90 b0 00 00 00 8b 07 89 c2 c1
fa 10
[  234.212706] RIP  [<ffffffff81173400>] grabbed2flush_reserved_nolock+0x33/0x87
[  234.212708]  RSP <ffff88003da41b68>
[  234.212710] ---[ end trace 6e3abf5361ca3bde ]---

Please, I need help to find the right way.

Thanks in advance.

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Edward Shishkin]

On 05/15/2012 12:55 AM, Sandro Souza wrote:
> Hello my friends.

Hello.

>
> I made a new custom distro based on debian squeeze, but with 2.6.39
> kernel, patched with reiser4
> (http://downloads.sourceforge.net/project/reiser4/reiser4-for-linux-2.6/reiser4-for-2.6.39-1.patch.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Freiser4%2Ffiles%2Freiser4-for-linux-2.6%2F&ts=1337035240&use_mirror=ufpr).
>
> Trying to copy a folder from a reiser3 partition to a reiser4
> partition, I got error messages.
>
> Testing my custom kernel with "preemption model" in "Preemptible
> Kernel", I got these messages:
>
> [  286.945598] ------------[ cut here ]------------
> [  286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
> [  286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
> [  286.945626] last sysfs file:
>
[...]

> [  286.945676]
> [  286.945683] Pid: 2433, comm: nautilus Not tainted 2.6.39-reiser4 #1
> innotek GmbH VirtualBox
> [  286.945691] RIP: 0010:[<ffffffff81178fae>]  [<ffffffff81178fae>]
> grabbed2flush_reserved_nolock+0x33/0x7b


It seems we forget to reserve space somewhere in write_end_cryptcompress().

Reiser4 uses technique of "soft ENOSPC": since it is impossible to
estimate precisely amount of disk space that will be spent by any
operation, every thread reserves needed amount of space with store,
so that surplus will be freed at commit time.

It seems we forget to reserve space somewhere in 
write_end_cryptcompress() for some needs, so that all reserved space
is exhausted and update_sd(), which wants 1 reserved block encounters
a zeroed counter of reserved blocks.

I'll try to take a look at this problem at leisure. For now
please don't use loop devices over compressed reiser4 files.

Thanks,
Edward.



> [  286.945704] RSP: 0018:ffff880014219b38  EFLAGS: 00010297
> [  286.945711] RAX: ffff88003b3df900 RBX: ffff88002fe75000 RCX: 0000000000112ef7
> [  286.945718] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff88003d3801c0
> [  286.945724] RBP: 0000000000000001 R08: 0000000000000000 R09: 000000000000000b
> [  286.945730] R10: dead000000100100 R11: ffff880014219c50 R12: ffff88003d385500
> [  286.945737] R13: ffff8800119bb000 R14: ffff880014219c38 R15: 0000000000001000
> [  286.945744] FS:  00007f126661d700(0000) GS:ffff88003fc00000(0000)
> knlGS:0000000000000000
> [  286.945750] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [  286.945756] CR2: 00007fba4a3df010 CR3: 000000003be2b000 CR4: 00000000000006f0
> [  286.945766] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [  286.945772] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> [  286.945778] Process nautilus (pid: 2433, threadinfo
> ffff880014218000, task ffff88001406e940)
> [  286.945783] Stack:
> [  286.945788]  0000000000000000 ffff8800119bb000 ffff88003d3801c0
> ffffffff8117b484
> [  286.945795]  ffff8800119bb050 ffff8800119bb000 0000000000000000
> ffffffff8117b529
> [  286.945802]  0000000000000000 ffff88003d385608 0000000000000000
> ffffffff8118e00d
> [  286.945808] Call Trace:
> [  286.945817]  [<ffffffff8117b484>] ? jnode_make_dirty_locked+0x112/0x18f
> [  286.945824]  [<ffffffff8117b529>] ? znode_make_dirty+0x28/0x87
> [  286.945831]  [<ffffffff8118e00d>] ? update_sd+0x344/0x3a4
> [  286.945838]  [<ffffffff81387a46>] ? sub_preempt_count+0x83/0x94
> [  286.945845]  [<ffffffff8118e096>] ? write_sd_by_inode_common+0x29/0x92
> [  286.945852]  [<ffffffff8103cb67>] ? get_parent_ip+0x9/0x1b
> [  286.945858]  [<ffffffff8118512b>] ? reiser4_dirty_inode+0x19/0x73
> [  286.945865]  [<ffffffff810edecb>] ? T.1132+0x12/0x2e
> [  286.945872]  [<ffffffff8104b288>] ? current_fs_time+0x1e/0x24
> [  286.945878]  [<ffffffff81112ddb>] ? __mark_inode_dirty+0x28/0x1c8
> [  286.945885]  [<ffffffff81108169>] ? file_update_time+0xf7/0x126
> [  286.945891]  [<ffffffff811924ee>] ? reiser4_write_end_careful+0x147/0x184
> [  286.945897]  [<ffffffff81115083>] ? pipe_to_file+0x152/0x161
> [  286.945903]  [<ffffffff81387af5>] ? add_preempt_count+0x9e/0xa0
> [  286.945909]  [<ffffffff81114f31>] ? generic_file_splice_write+0x133/0x133
> [  286.945914]  [<ffffffff811143cb>] ? splice_from_pipe_feed+0x6d/0xed
> [  286.945920]  [<ffffffff81114eb3>] ? generic_file_splice_write+0xb5/0x133
> [  286.945933]  [<ffffffff811162a6>] ? sys_splice+0x2f8/0x3d2
> [  286.945939]  [<ffffffff8138a7d2>] ? system_call_fastpath+0x16/0x1b
> [  286.945944] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 38 05 00
> 00 48 8b 90 a0 00 00 00 48 8b 9a 88 02 00 00 48 8b 90 b0 00 00 00 48
> 39 f2 73 04<0f>  0b eb fe 48 29 f2 48 89 90 b0 00 00 00 8b 07 89 c2 c1
> fa 10
> [  286.945968] RIP  [<ffffffff81178fae>] grabbed2flush_reserved_nolock+0x33/0x7b
> [  286.945974]  RSP<ffff880014219b38>
> [  286.945979] ---[ end trace cd39088cc6246a13 ]---
>
> Testing my custom kernel with "preemption model" in "Voluntary  Kernel
> Preemption", I got these messages:
>
> [  234.212538] ------------[ cut here ]------------
> [  234.212549] kernel BUG at fs/reiser4/block_alloc.c:151!
> [  234.212551] invalid opcode: 0000 [#1] SMP
> [  234.212553] last sysfs file:
> /sys/devices/pci0000:00/0000:00:0b.0/usb1/1-1/1-1:1.0/host2/target2:0:0/2:0:0:0/block/sdc/sdc1/uevent
> [  234.212556] CPU 0
> [  234.212557] Modules linked in: mperf cpufreq_powersave
> cpufreq_stats cpufreq_userspace cpufreq_conservative ppdev lp
> bluetooth rfkill binfmt_misc fuse loop snd_intel8x0 i2c_piix4
> snd_ac97_codec psmouse joydev processor thermal_sys evdev i2c_core
> parport_pc parport tpm_tis pcspkr tpm serio_raw tpm_bios ac ac97_bus
> snd_pcm snd_seq snd_timer snd_seq_device snd button soundcore
> snd_page_alloc usbhid hid ide_cd_mod cdrom ata_generic sd_mod
> pata_acpi ata_piix crc_t10dif usb_storage ide_pci_generic ohci_hcd
> ahci libahci ehci_hcd libata piix scsi_mod usbcore e1000 ide_core
> [last unloaded: scsi_wait_scan]
> [  234.212581]
> [  234.212583] Pid: 2530, comm: nautilus Not tainted 2.6.39-reiser4 #1
> innotek GmbH VirtualBox
> [  234.212585] RIP: 0010:[<ffffffff81173400>]  [<ffffffff81173400>]
> grabbed2flush_reserved_nolock+0x33/0x87
> [  234.212591] RSP: 0018:ffff88003da41b68  EFLAGS: 00010297
> [  234.212592] RAX: ffff88003c939500 RBX: ffff88003bcf9800 RCX: ffff880007e49d80
> [  234.212594] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8800215e41c0
> [  234.212595] RBP: 0000000000000001 R08: 0000000000000000 R09: 000000000000000b
> [  234.212596] R10: ffff88002d3b96b0 R11: ffff88003da41c80 R12: ffff880021693500
> [  234.212598] R13: ffff88001056b900 R14: ffff88003da41c68 R15: 0000000000001000
> [  234.212605] FS:  00007f94bc7cc700(0000) GS:ffff88003fc00000(0000)
> knlGS:0000000000000000
> [  234.212607] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [  234.212608] CR2: 0000000000428060 CR3: 000000003c197000 CR4: 00000000000006f0
> [  234.212614] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [  234.212615] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> [  234.212617] Process nautilus (pid: 2530, threadinfo
> ffff88003da40000, task ffff880026095b00)
> [  234.212618] Stack:
> [  234.212619]  0000000000000000 ffff88001056b900 ffff8800215e41c0
> ffffffff81175c90
> [  234.212621]  0000000000001000 ffff88001056b900 0000000000000000
> ffffffff81175d40
> [  234.212624]  0000000000000000 ffff880021693608 0000000000000000
> ffffffff81188945
> [  234.212626] Call Trace:
> [  234.212630]  [<ffffffff81175c90>] ? jnode_make_dirty_locked+0x112/0x19a
> [  234.212632]  [<ffffffff81175d40>] ? znode_make_dirty+0x28/0xa1
> [  234.212634]  [<ffffffff81188945>] ? update_sd+0x368/0x3d9
> [  234.212637]  [<ffffffff81171f26>] ? insert_by_coord+0xef/0x101
> [  234.212641]  [<ffffffff81199f4d>] ? ctail_insert_unprepped_cluster+0xe3/0xf2
> [  234.212643]  [<ffffffff811889df>] ? write_sd_by_inode_common+0x29/0x92
> [  234.212645]  [<ffffffff81175a1d>] ? jnode_get_atom+0x34/0xac
> [  234.212648]  [<ffffffff8117fab1>] ? reiser4_dirty_inode+0x19/0x73
> [  234.212652]  [<ffffffff81049a2c>] ? current_fs_time+0x1e/0x24
> [  234.212656]  [<ffffffff810ea179>] ? kfree+0xa8/0xcc
> [  234.212659]  [<ffffffff8110d9ee>] ? __mark_inode_dirty+0x22/0x1a4
> [  234.212671]  [<ffffffff81103254>] ? file_update_time+0xea/0x115
> [  234.212674]  [<ffffffff8118ce02>] ? reiser4_write_end_careful+0x147/0x184
> [  234.212677]  [<ffffffff8110fa8c>] ? pipe_to_file+0x12f/0x13e
> [  234.212679]  [<ffffffff8110f95d>] ? generic_file_splice_write+0x133/0x133
> [  234.212681]  [<ffffffff8110eebb>] ? splice_from_pipe_feed+0x6d/0xed
> [  234.212684]  [<ffffffff8110f8df>] ? generic_file_splice_write+0xb5/0x133
> [  234.212686]  [<ffffffff811100c7>] ? sys_splice+0x2f8/0x3db
> [  234.212689]  [<ffffffff81382412>] ? system_call_fastpath+0x16/0x1b
> [  234.212690] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 18 05 00
> 00 48 8b 90 a0 00 00 00 48 8b 9a 88 02 00 00 48 8b 90 b0 00 00 00 48
> 39 f2 73 04<0f>  0b eb fe 48 29 f2 48 89 90 b0 00 00 00 8b 07 89 c2 c1
> fa 10
> [  234.212706] RIP  [<ffffffff81173400>] grabbed2flush_reserved_nolock+0x33/0x87
> [  234.212708]  RSP<ffff88003da41b68>
> [  234.212710] ---[ end trace 6e3abf5361ca3bde ]---
>
> Please, I need help to find the right way.
>
> Thanks in advance.
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Edward Shishkin]

On 05/15/2012 12:12 PM, Edward Shishkin wrote:
> On 05/15/2012 12:55 AM, Sandro Souza wrote:
>> Hello my friends.
>
> Hello.
>
>>
>> I made a new custom distro based on debian squeeze, but with 2.6.39
>> kernel, patched with reiser4
>> (http://downloads.sourceforge.net/project/reiser4/reiser4-for-linux-2.6/reiser4-for-2.6.39-1.patch.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Freiser4%2Ffiles%2Freiser4-for-linux-2.6%2F&ts=1337035240&use_mirror=ufpr).
>>
>>
>> Trying to copy a folder from a reiser3 partition to a reiser4
>> partition, I got error messages.
>>
>> Testing my custom kernel with "preemption model" in "Preemptible
>> Kernel", I got these messages:
>>
>> [ 286.945598] ------------[ cut here ]------------
>> [ 286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
>> [ 286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
>> [ 286.945626] last sysfs file:
>>
> [...]
>
>> [ 286.945676]
>> [ 286.945683] Pid: 2433, comm: nautilus Not tainted 2.6.39-reiser4 #1
>> innotek GmbH VirtualBox
>> [ 286.945691] RIP: 0010:[<ffffffff81178fae>] [<ffffffff81178fae>]
>> grabbed2flush_reserved_nolock+0x33/0x7b
>
>
> Reiser4 uses technique of "soft ENOSPC": since it is impossible to
> estimate precisely amount of disk space that will be spent by any
> operation, every thread reserves needed amount of space with store,
> so that surplus will be freed at commit time.
>
> It seems we forget to reserve space somewhere in
> write_end_cryptcompress()


^write_end^write_begin

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Edward Shishkin]

On 05/15/2012 12:12 PM, Edward Shishkin wrote:
> On 05/15/2012 12:55 AM, Sandro Souza wrote:
>> Hello my friends.
>
> Hello.
>
>>
>> I made a new custom distro based on debian squeeze, but with 2.6.39
>> kernel, patched with reiser4
>> (http://downloads.sourceforge.net/project/reiser4/reiser4-for-linux-2.6/reiser4-for-2.6.39-1.patch.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Freiser4%2Ffiles%2Freiser4-for-linux-2.6%2F&ts=1337035240&use_mirror=ufpr).
>>
>>
>> Trying to copy a folder from a reiser3 partition to a reiser4
>> partition, I got error messages.
>>
>> Testing my custom kernel with "preemption model" in "Preemptible
>> Kernel", I got these messages:
>>
>> [ 286.945598] ------------[ cut here ]------------
>> [ 286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
>> [ 286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
>> [ 286.945626] last sysfs file:
>>
> [...]
>
>> [ 286.945676]
>> [ 286.945683] Pid: 2433, comm: nautilus Not tainted 2.6.39-reiser4 #1
>> innotek GmbH VirtualBox
>> [ 286.945691] RIP: 0010:[<ffffffff81178fae>] [<ffffffff81178fae>]
>> grabbed2flush_reserved_nolock+0x33/0x7b
>
>
> It seems we forget to reserve space somewhere in write_end_cryptcompress().
>
> Reiser4 uses technique of "soft ENOSPC": since it is impossible to
> estimate precisely amount of disk space that will be spent by any
> operation, every thread reserves needed amount of space with store,
> so that surplus will be freed at commit time.
>
> It seems we forget to reserve space somewhere in
> write_end_cryptcompress() for some needs, so that all reserved space
> is exhausted and update_sd(), which wants 1 reserved block encounters
> a zeroed counter of reserved blocks.


BTW we don't reserve space exactly for update_sd() (one block should be
reserved). Anybody care to fix this?

Thanks,
Edward.

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Edward Shishkin]

[-- Attachment #1: Type: text/plain, Size: 2334 bytes --]

Hello.

Please, try the attached patch: it should help...
I have also fixed a deadlock because of reiser4
lock ordering violation:

do_lo_send_aops() keeps page locked and calls file_update_time()
which calls reiser4_update_sd, which tries to acquire a longterm
lock (bad).

Thanks,
Edward.

On 05/15/2012 12:55 AM, Sandro Souza wrote:
> Hello my friends.
>
> I made a new custom distro based on debian squeeze, but with 2.6.39
> kernel, patched with reiser4

[...]

> Trying to copy a folder from a reiser3 partition to a reiser4
> partition, I got error messages.
>
> Testing my custom kernel with "preemption model" in "Preemptible
> Kernel", I got these messages:
>
> [  286.945598] ------------[ cut here ]------------
> [  286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
> [  286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
> [  286.945626] last sysfs file:

[...]

> ffffffff8118e00d
> [  286.945808] Call Trace:
> [  286.945817]  [<ffffffff8117b484>] ? jnode_make_dirty_locked+0x112/0x18f
> [  286.945824]  [<ffffffff8117b529>] ? znode_make_dirty+0x28/0x87
> [  286.945831]  [<ffffffff8118e00d>] ? update_sd+0x344/0x3a4
> [  286.945838]  [<ffffffff81387a46>] ? sub_preempt_count+0x83/0x94
> [  286.945845]  [<ffffffff8118e096>] ? write_sd_by_inode_common+0x29/0x92
> [  286.945852]  [<ffffffff8103cb67>] ? get_parent_ip+0x9/0x1b
> [  286.945858]  [<ffffffff8118512b>] ? reiser4_dirty_inode+0x19/0x73
> [  286.945865]  [<ffffffff810edecb>] ? T.1132+0x12/0x2e
> [  286.945872]  [<ffffffff8104b288>] ? current_fs_time+0x1e/0x24
> [  286.945878]  [<ffffffff81112ddb>] ? __mark_inode_dirty+0x28/0x1c8
> [  286.945885]  [<ffffffff81108169>] ? file_update_time+0xf7/0x126
> [  286.945891]  [<ffffffff811924ee>] ? reiser4_write_end_careful+0x147/0x184
> [  286.945897]  [<ffffffff81115083>] ? pipe_to_file+0x152/0x161
> [  286.945903]  [<ffffffff81387af5>] ? add_preempt_count+0x9e/0xa0
> [  286.945909]  [<ffffffff81114f31>] ? generic_file_splice_write+0x133/0x133
> [  286.945914]  [<ffffffff811143cb>] ? splice_from_pipe_feed+0x6d/0xed
> [  286.945920]  [<ffffffff81114eb3>] ? generic_file_splice_write+0xb5/0x133
> [  286.945933]  [<ffffffff811162a6>] ? sys_splice+0x2f8/0x3d2
> [  286.945939]  [<ffffffff8138a7d2>] ? system_call_fastpath+0x16/0x1b
> [  286.945944] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 38 05 00

[-- Attachment #2: reiser4-fix-reservation-and-deadlock.patch --]
[-- Type: text/plain, Size: 7859 bytes --]

Reserve space in reiser4_write_begin() properly:
1) for update_sd() when changing file szie;
2) for update_sd() when updating mtime/ctime

Fix deadlock (reiser4 lock ordering violation):
release page lock in reiser4_dirty_inode()
before aquiring a longterm lock.

Signed-off-by: Edward Shishkin <edward.shishkin@gmail.com>
---
 fs/reiser4/context.h                     |    4 +++
 fs/reiser4/plugin/file/cryptcompress.c   |   29 +++++++++++++++++++--
 fs/reiser4/plugin/file/file.c            |   41 +++++++++++++++++++++----------
 fs/reiser4/plugin/file/file_conversion.c |   11 +++-----
 fs/reiser4/super_ops.c                   |   17 +++++++++---
 5 files changed, 76 insertions(+), 26 deletions(-)

Index: linux-2.6.39/fs/reiser4/plugin/file/cryptcompress.c
===================================================================
--- linux-2.6.39.orig/fs/reiser4/plugin/file/cryptcompress.c
+++ linux-2.6.39/fs/reiser4/plugin/file/cryptcompress.c
@@ -1521,7 +1521,9 @@ static int update_sd_cryptcompress(struc
 					  BA_CAN_COMMIT);
 	if (result)
 		return result;
-	inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+	if (!IS_NOCMTIME(inode))
+		inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+
 	result = reiser4_update_sd(inode);
 
 	if (unlikely(result != 0))
@@ -3755,7 +3757,9 @@ int write_begin_cryptcompress(struct fil
 	struct reiser4_slide *win;
 	struct cluster_handle *clust;
 	struct cryptcompress_info *info;
+	reiser4_context *ctx;
 
+	ctx = get_current_context();
 	inode = page->mapping->host;
 	info = cryptcompress_inode_data(inode);
 
@@ -3789,9 +3793,13 @@ int write_begin_cryptcompress(struct fil
 		ClearPageUptodate(page);
 		goto err0;
 	}
-	/* Success. All resources (including checkin_mutex)
-	   will be released in ->write_end() */
+	/*
+	 * Success. All resources (including checkin_mutex)
+	 * will be released in ->write_end()
+	 */
+	ctx->locked_page = page;
 	*fsdata = (void *)buf;
+
 	return 0;
  err0:
 	put_cluster_handle(clust);
@@ -3812,15 +3820,18 @@ int write_end_cryptcompress(struct file 
 	struct inode *inode;
 	struct cluster_handle *clust;
 	struct cryptcompress_info *info;
+	reiser4_context *ctx;
 
 	assert("edward-1566",
 	       lock_stack_isclean(get_current_lock_stack()));
+	ctx = get_current_context();
 	inode = page->mapping->host;
 	info = cryptcompress_inode_data(inode);
 	clust = (struct cluster_handle *)fsdata;
 	hint = clust->hint;
 
 	unlock_page(page);
+	ctx->locked_page = NULL;
 	set_cluster_pages_dirty(clust, inode);
 	ret = checkin_logical_cluster(clust, inode);
 	if (ret) {
@@ -3831,6 +3842,18 @@ int write_end_cryptcompress(struct file 
 	mutex_unlock(&info->checkin_mutex);
 
 	put_cluster_handle(clust);
+
+	if (pos + copied > inode->i_size) {
+		/*
+		 * i_size has been updated in
+		 * checkin_logical_cluster
+		 */
+		ret = reiser4_update_sd(inode);
+		if (unlikely(ret != 0))
+			warning("edward-1603",
+				"Can not update stat-data: %i. FSCK?",
+				ret);
+	}
 	kfree(fsdata);
 	return ret;
 }
Index: linux-2.6.39/fs/reiser4/plugin/file/file.c
===================================================================
--- linux-2.6.39.orig/fs/reiser4/plugin/file/file.c
+++ linux-2.6.39/fs/reiser4/plugin/file/file.c
@@ -2128,6 +2128,7 @@ ssize_t write_unix_file(struct file *fil
 		new_size = *pos + count;
 
 	while (left) {
+		int update_sd = 0;
 		if (left < to_write)
 			to_write = left;
 
@@ -2239,18 +2240,27 @@ ssize_t write_unix_file(struct file *fil
 		assert("edward-1555",
 		       ergo(uf_info->container == UF_CONTAINER_TAILS,
 			    write_op == reiser4_write_tail));
-		if (*pos + written > inode->i_size)
+		if (*pos + written > inode->i_size) {
 			INODE_SET_FIELD(inode, i_size, *pos + written);
-		file_update_time(file);
-		/* space for update_sd was reserved in write_op */
-		result = reiser4_update_sd(inode);
-		if (result) {
-			warning("edward-1574",
-				"Can not update stat-data: %i. FSCK?",
-				result);
-			drop_access(uf_info);
-			context_set_commit_async(ctx);
-			break;
+			update_sd = 1;
+		}
+		if (!IS_NOCMTIME(inode)) {
+			inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+			update_sd = 1;
+		}
+		if (update_sd) {
+			/*
+			 * space for update_sd was reserved in write_op
+			 */
+			result = reiser4_update_sd(inode);
+			if (result) {
+				warning("edward-1574",
+					"Can not update stat-data: %i. FSCK?",
+					result);
+				drop_access(uf_info);
+				context_set_commit_async(ctx);
+				break;
+			}
 		}
 		drop_access(uf_info);
 		ea = NEITHER_OBTAINED;
@@ -2768,14 +2778,19 @@ int write_end_unix_file(struct file *fil
 		SetPageError(page);
 		goto exit;
 	}
-	if (pos + copied > inode->i_size)
+	if (pos + copied > inode->i_size) {
 		INODE_SET_FIELD(inode, i_size, pos + copied);
+		ret = reiser4_update_sd(inode);
+		if (unlikely(ret != 0))
+			warning("edward-1604",
+				"Can not update stat-data: %i. FSCK?",
+				ret);
+	}
  exit:
 	drop_exclusive_access(info);
 	return ret;
 }
 
-
 /*
  * Local variables:
  * c-indentation-style: "K&R"
Index: linux-2.6.39/fs/reiser4/plugin/file/file_conversion.c
===================================================================
--- linux-2.6.39.orig/fs/reiser4/plugin/file/file_conversion.c
+++ linux-2.6.39/fs/reiser4/plugin/file/file_conversion.c
@@ -671,9 +671,11 @@ int reiser4_write_begin_careful(struct f
 		ret = PTR_ERR(ctx);
 		goto err2;
 	}
-	ret = reiser4_grab_space_force(/* one for stat data update */
-					  estimate_update_common(inode),
-					  BA_CAN_COMMIT);
+	ret = reiser4_grab_space_force(/* for update_sd:
+					* one when updating file size and
+					* one when updating mtime/ctime */
+				       2 * estimate_update_common(inode),
+				       BA_CAN_COMMIT);
 	if (ret)
 		goto err1;
 	ret = PROT_PASSIVE(int, write_begin, (file, page, pos, len, fsdata));
@@ -713,9 +715,6 @@ int reiser4_write_end_careful(struct fil
 	ret = PROT_PASSIVE(int, write_end, (file, page, pos, copied, fsdata));
 	page_cache_release(page);
 
-	file_update_time(file);
-	/* space for update_sd was reserved in reiser4_write_begin */
-	ret = reiser4_update_sd(inode);
 	/* don't commit transaction under inode semaphore */
 	context_set_commit_async(ctx);
 	reiser4_exit_context(ctx);
Index: linux-2.6.39/fs/reiser4/context.h
===================================================================
--- linux-2.6.39.orig/fs/reiser4/context.h
+++ linux-2.6.39/fs/reiser4/context.h
@@ -87,6 +87,10 @@ struct reiser4_context {
 	 * flushed */
 	int nr_captured;
 	int nr_children;	/* number of child contexts */
+	struct page *locked_page; /* page that should be unlocked in
+				   * reiser4_dirty_inode() before taking
+				   * a longterm lock (to not violate
+				   * reiser4 lock ordering) */
 #if REISER4_DEBUG
 	/* debugging information about reiser4 locks held by the current
 	 * thread */
Index: linux-2.6.39/fs/reiser4/super_ops.c
===================================================================
--- linux-2.6.39.orig/fs/reiser4/super_ops.c
+++ linux-2.6.39/fs/reiser4/super_ops.c
@@ -166,16 +166,25 @@ static void reiser4_destroy_inode(struct
 static void reiser4_dirty_inode(struct inode *inode)
 {
 	int result;
+	reiser4_context *ctx;
 
 	if (!is_in_reiser4_context())
 		return;
-	assert("", !IS_RDONLY(inode));
-	assert("", (inode_file_plugin(inode)->estimate.update(inode) <=
-		    get_current_context()->grabbed_blocks));
+	assert("edward-1606", !IS_RDONLY(inode));
+	assert("edward-1607",
+	       (inode_file_plugin(inode)->estimate.update(inode) <=
+		get_current_context()->grabbed_blocks));
+
+	ctx = get_current_context();
+	if (ctx->locked_page)
+		unlock_page(ctx->locked_page);
 
 	result = reiser4_update_sd(inode);
+
+	if (ctx->locked_page)
+		lock_page(ctx->locked_page);
 	if (result)
-		warning("", "failed to dirty inode for %llu: %d",
+		warning("edward-1605", "failed to dirty inode for %llu: %d",
 			get_inode_oid(inode), result);
 }
 

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Sandro Souza]

Hello my friends.

Edward, you ARE the man! :D

Your patch works wonderfully.

Inside my virtual machine, I made a new custom distro (based on debian
squeeze, but with 2.6.39 kernel, patched with reiser4 native support),
and I was able to install it in another virtual machine (/boot as
reiser3, / as reiser4 and /home as reiser4).

Everything are working fine now. :D

Now, I'll burn this new custom distro in a DVD and I'll install it on
my notebook (no more VM) to test it.

Please, may you tell me if the reiser4 has some auto-defragmentation feature?

The compression feature (gzip1) is amazing. If we have
auto-defragmentation feature also.....man!!! No words. :D

Thank you and the entire reiser4 team. You all are making the
difference. REALLY!

2012/5/20 Edward Shishkin <edward.shishkin@gmail.com>:
> Hello.
>
> Please, try the attached patch: it should help...
> I have also fixed a deadlock because of reiser4
> lock ordering violation:
>
> do_lo_send_aops() keeps page locked and calls file_update_time()
> which calls reiser4_update_sd, which tries to acquire a longterm
> lock (bad).
>
> Thanks,
> Edward.
>
>
> On 05/15/2012 12:55 AM, Sandro Souza wrote:
>>
>> Hello my friends.
>>
>> I made a new custom distro based on debian squeeze, but with 2.6.39
>> kernel, patched with reiser4
>
>
> [...]
>
>
>> Trying to copy a folder from a reiser3 partition to a reiser4
>> partition, I got error messages.
>>
>> Testing my custom kernel with "preemption model" in "Preemptible
>> Kernel", I got these messages:
>>
>> [  286.945598] ------------[ cut here ]------------
>> [  286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
>> [  286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
>> [  286.945626] last sysfs file:
>
>
> [...]
>
>
>> ffffffff8118e00d
>> [  286.945808] Call Trace:
>> [  286.945817]  [<ffffffff8117b484>] ? jnode_make_dirty_locked+0x112/0x18f
>> [  286.945824]  [<ffffffff8117b529>] ? znode_make_dirty+0x28/0x87
>> [  286.945831]  [<ffffffff8118e00d>] ? update_sd+0x344/0x3a4
>> [  286.945838]  [<ffffffff81387a46>] ? sub_preempt_count+0x83/0x94
>> [  286.945845]  [<ffffffff8118e096>] ? write_sd_by_inode_common+0x29/0x92
>> [  286.945852]  [<ffffffff8103cb67>] ? get_parent_ip+0x9/0x1b
>> [  286.945858]  [<ffffffff8118512b>] ? reiser4_dirty_inode+0x19/0x73
>> [  286.945865]  [<ffffffff810edecb>] ? T.1132+0x12/0x2e
>> [  286.945872]  [<ffffffff8104b288>] ? current_fs_time+0x1e/0x24
>> [  286.945878]  [<ffffffff81112ddb>] ? __mark_inode_dirty+0x28/0x1c8
>> [  286.945885]  [<ffffffff81108169>] ? file_update_time+0xf7/0x126
>> [  286.945891]  [<ffffffff811924ee>] ?
>> reiser4_write_end_careful+0x147/0x184
>> [  286.945897]  [<ffffffff81115083>] ? pipe_to_file+0x152/0x161
>> [  286.945903]  [<ffffffff81387af5>] ? add_preempt_count+0x9e/0xa0
>> [  286.945909]  [<ffffffff81114f31>] ?
>> generic_file_splice_write+0x133/0x133
>> [  286.945914]  [<ffffffff811143cb>] ? splice_from_pipe_feed+0x6d/0xed
>> [  286.945920]  [<ffffffff81114eb3>] ?
>> generic_file_splice_write+0xb5/0x133
>> [  286.945933]  [<ffffffff811162a6>] ? sys_splice+0x2f8/0x3d2
>> [  286.945939]  [<ffffffff8138a7d2>] ? system_call_fastpath+0x16/0x1b
>> [  286.945944] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 38 05 00
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Testing a custom kernel (2.6.39) with native reiser4 support

[Edward Shishkin]

On 05/24/2012 07:53 PM, Sandro Souza wrote:
> Hello my friends.
>
> Edward, you ARE the man! :D
>
> Your patch works wonderfully.
>
> Inside my virtual machine, I made a new custom distro (based on debian
> squeeze, but with 2.6.39 kernel, patched with reiser4 native support),
> and I was able to install it in another virtual machine (/boot as
> reiser3, / as reiser4 and /home as reiser4).
>
> Everything are working fine now. :D

Thanks.

Actually not everything is so iridescent:
there are a couple of bugs which can be really annoying..
We'll continue to narrow down them at leisure. For now make
a friendship with fsck..


>
> Now, I'll burn this new custom distro in a DVD and I'll install it on
> my notebook (no more VM) to test it.
>
> Please, may you tell me if the reiser4 has some auto-defragmentation feature?


Under construction.
I think that repacker and next features will be paid options
(money will go for Reiser's children).

Thanks,
Edward.


>
> The compression feature (gzip1) is amazing. If we have
> auto-defragmentation feature also.....man!!! No words. :D
>
> Thank you and the entire reiser4 team. You all are making the
> difference. REALLY!
>
> 2012/5/20 Edward Shishkin<edward.shishkin@gmail.com>:
>> Hello.
>>
>> Please, try the attached patch: it should help...
>> I have also fixed a deadlock because of reiser4
>> lock ordering violation:
>>
>> do_lo_send_aops() keeps page locked and calls file_update_time()
>> which calls reiser4_update_sd, which tries to acquire a longterm
>> lock (bad).
>>
>> Thanks,
>> Edward.
>>
>>
>> On 05/15/2012 12:55 AM, Sandro Souza wrote:
>>>
>>> Hello my friends.
>>>
>>> I made a new custom distro based on debian squeeze, but with 2.6.39
>>> kernel, patched with reiser4
>>
>>
>> [...]
>>
>>
>>> Trying to copy a folder from a reiser3 partition to a reiser4
>>> partition, I got error messages.
>>>
>>> Testing my custom kernel with "preemption model" in "Preemptible
>>> Kernel", I got these messages:
>>>
>>> [  286.945598] ------------[ cut here ]------------
>>> [  286.945609] kernel BUG at fs/reiser4/block_alloc.c:151!
>>> [  286.945617] invalid opcode: 0000 [#1] PREEMPT SMP
>>> [  286.945626] last sysfs file:
>>
>>
>> [...]
>>
>>
>>> ffffffff8118e00d
>>> [  286.945808] Call Trace:
>>> [  286.945817]  [<ffffffff8117b484>] ? jnode_make_dirty_locked+0x112/0x18f
>>> [  286.945824]  [<ffffffff8117b529>] ? znode_make_dirty+0x28/0x87
>>> [  286.945831]  [<ffffffff8118e00d>] ? update_sd+0x344/0x3a4
>>> [  286.945838]  [<ffffffff81387a46>] ? sub_preempt_count+0x83/0x94
>>> [  286.945845]  [<ffffffff8118e096>] ? write_sd_by_inode_common+0x29/0x92
>>> [  286.945852]  [<ffffffff8103cb67>] ? get_parent_ip+0x9/0x1b
>>> [  286.945858]  [<ffffffff8118512b>] ? reiser4_dirty_inode+0x19/0x73
>>> [  286.945865]  [<ffffffff810edecb>] ? T.1132+0x12/0x2e
>>> [  286.945872]  [<ffffffff8104b288>] ? current_fs_time+0x1e/0x24
>>> [  286.945878]  [<ffffffff81112ddb>] ? __mark_inode_dirty+0x28/0x1c8
>>> [  286.945885]  [<ffffffff81108169>] ? file_update_time+0xf7/0x126
>>> [  286.945891]  [<ffffffff811924ee>] ?
>>> reiser4_write_end_careful+0x147/0x184
>>> [  286.945897]  [<ffffffff81115083>] ? pipe_to_file+0x152/0x161
>>> [  286.945903]  [<ffffffff81387af5>] ? add_preempt_count+0x9e/0xa0
>>> [  286.945909]  [<ffffffff81114f31>] ?
>>> generic_file_splice_write+0x133/0x133
>>> [  286.945914]  [<ffffffff811143cb>] ? splice_from_pipe_feed+0x6d/0xed
>>> [  286.945920]  [<ffffffff81114eb3>] ?
>>> generic_file_splice_write+0xb5/0x133
>>> [  286.945933]  [<ffffffff811162a6>] ? sys_splice+0x2f8/0x3d2
>>> [  286.945939]  [<ffffffff8138a7d2>] ? system_call_fastpath+0x16/0x1b
>>> [  286.945944] Code: 25 80 cc 00 00 53 48 83 ec 08 48 8b 80 38 05 00