On 04/28/2011 11:25 AM, Ruediger Meier wrote:> Hi,
>
> About 3 weeks ago I was testing btrfs with enabled compression and got
this:
>
>
> [1376756.337203] ------------[ cut here ]------------
> [1376756.337211] kernel BUG at
/usr/src/packages/BUILD/kernel-desktop-2.6.37.1/linux-2.6.37/fs/btrfs/inode.c:6403!
> [1376756.337217] invalid opcode: 0000 [#1] PREEMPT SMP
> [1376756.337221] last sysfs file:
/sys/devices/system/cpu/cpu3/cache/index2/shared_cpu_map
> [1376756.337226] CPU 2
> [1376756.337228] Modules linked in: btrfs zlib_deflate ext4 jbd2 crc16
minix hfs vfat fat loop w83627ehf hwmon_vid
> coretemp snd_pcm_oss snd_mixer_oss snd_seq snd_seq_device edd nfs lockd
fscache nfs_acl auth_rpcgss sunrpc microcode
> cpufreq_conservative cpufreq_userspace cpufreq_powersave acpi_cpufreq mperf
xfs exportfs dm_mod sg firewire_ohci
> sr_mod button snd_hda_codec_realtek cdrom firewire_core e1000e
snd_hda_intel snd_hda_codec iTCO_wdt
> iTCO_vendor_support snd_hwdep snd_pcm snd_timer snd soundcore
snd_page_alloc crc_itu_t serio_raw i2c_i801 reiserfs fan
> processor ata_generic pata_jmicron thermal thermal_sys [last unloaded:
nvidia]
> [1376756.337277]
> [1376756.337279] Pid: 10477, comm: tee Tainted: P
2.6.37.1-1.2-desktop #1 /DP43TF
> [1376756.337287] RIP: 0010:[<ffffffffa03124da>]
[<ffffffffa03124da>] btrfs_truncate+0x21a/0x220 [btrfs]
> [1376756.337307] RSP: 0018:ffff88000baf5c28 EFLAGS: 00010286
> [1376756.337310] RAX: 00000000ffffffe4 RBX: ffff88002c11f8a0 RCX:
0000000000000000
> [1376756.337314] RDX: 0000000000000008 RSI: ffff880031f847f0 RDI:
ffff8801c172ab00
> [1376756.337318] RBP: ffff88012f515800 R08: 0000000000000005 R09:
ffff88000baf5a38
> [1376756.337321] R10: 00000000ffffffe4 R11: 0000000000000001 R12:
ffff88018665ed18
> [1376756.337325] R13: ffff88018665eb80 R14: 0000000000000000 R15:
0000000000000008
> [1376756.337330] FS: 00007f5630c4f700(0000) GS:ffff8800bf900000(0000)
knlGS:0000000000000000
> [1376756.337334] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [1376756.337337] CR2: 00007f56307bcce0 CR3: 000000011d895000 CR4:
00000000000406e0
> [1376756.337341] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
> [1376756.337345] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
> [1376756.337349] Process tee (pid: 10477, threadinfo ffff88000baf4000, task
ffff8800320ac880)
> [1376756.337353] Stack:
> [1376756.337355] ffff88018665ed18 0000000000000000 ffff88002c11f8a0
ffff88000baf5d38
> [1376756.337362] 0000000000000000 ffffffff81107938 ffff88012f515800
ffff880000000000
> [1376756.337367] ffff88000baf5d38 ffff88018665ed18 ffff88012f515800
ffffffffa0312847
> [1376756.337373] Call Trace:
> [1376756.337421] [<ffffffff81107938>] vmtruncate+0x58/0x70
> [1376756.337433] [<ffffffffa0312847>] btrfs_setattr_size+0xa7/0x240
[btrfs]
> [1376756.337469] [<ffffffffa0312a92>] btrfs_setattr+0xb2/0xc0
[btrfs]
> [1376756.337500] [<ffffffff8116b33b>] notify_change+0x19b/0x370
> [1376756.337508] [<ffffffff81150a2a>] do_truncate+0x5a/0x90
> [1376756.337513] [<ffffffff81160310>] finish_open+0x160/0x1d0
> [1376756.337519] [<ffffffff81160b48>] do_filp_open+0x1a8/0x620
> [1376756.337524] [<ffffffff811518f6>] do_sys_open+0x66/0x110
> [1376756.337531] [<ffffffff81002f8b>] system_call_fastpath+0x16/0x1b
> [1376756.337537] [<00007f56307bccf0>] 0x7f56307bccf0
> [1376756.337539] Code: df e8 eb c7 ff ff 85 c0 74 b9 0f 0b 4c 89 e2 48 89
ee 48 89 df e8 f7 4e 01 00 e9 c9 fe ff ff 0f
> 0b 0f 0b 0f 0b 0f 0b 0f 0b 0f 0b<0f> 0b 0f 1f 40 00 41 57 48 8d 87
68 fe ff ff 41 56 41 55 41 54
> [1376756.337568] RIP [<ffffffffa03124da>] btrfs_truncate+0x21a/0x220
[btrfs]
> [1376756.337578] RSP<ffff88000baf5c28>
> [1376756.337847] ---[ end trace 32985f059586e488 ]---
>
>
>
> Now a tee process and btrfs-transacti are hanging in state D. (That is not
the
> same tee pid as the one from the trace! It was started about 30 seconds
after
> that bug trace happened.)
>
> Changing directories on that mount still works. Every read or write results
in
> state D too.
> The machine is still up so I could get some more information if wanted.
Note
> that btrfs resides on lvm2 volume here.
>
>
> gonzo:~ # lsb-release -d
> Description: openSUSE 11.4 (x86_64)
>
> gonzo:~ # uname -a
> Linux gonzo 2.6.37.1-1.2-desktop #1 SMP PREEMPT 2011-02-21 10:34:10 +0100
x86_64 x86_64 x86_64 GNU/Linux
>
> gonzo:~ # grep btrfs /proc/mounts
> /dev/mapper/system-scratch /var/scratch btrfs rw,relatime,compress 0 0
>
> gonzo:~ # ps aux |grep "tee\|btrfs"
> freundt 4055 0.0 0.0 3960 280 ? D Apr07 0:00 tee
orig/12546.xml
> root 4557 0.0 0.0 0 0 ? S Apr07 2:15
[btrfs-endio-1]
> root 6638 0.0 0.0 0 0 ? S Apr07 0:56
[btrfs-endio-wri]
> root 6639 0.0 0.0 0 0 ? S Apr07 1:00
[btrfs-worker-1]
> root 15192 0.0 0.0 0 0 ? S Apr07 0:07
[flush-btrfs-12]
> root 29027 0.0 0.0 0 0 ? S Apr06 0:00
[btrfs-genwork-0]
> root 29028 0.0 0.0 0 0 ? S Apr06 0:20
[btrfs-submit-0]
> root 29030 0.0 0.0 0 0 ? S Apr06 0:00
[btrfs-fixup-0]
> root 29032 0.0 0.0 0 0 ? S Apr06 0:01
[btrfs-endio-met]
> root 29035 0.0 0.0 0 0 ? S Apr06 0:00
[btrfs-freespace]
> root 29036 0.0 0.0 0 0 ? S Apr06 0:00
[btrfs-cleaner]
> root 29037 0.0 0.0 0 0 ? D Apr06 0:16
[btrfs-transacti]
> root 29346 0.0 0.0 0 0 ? S Apr06 23:40
[btrfs-delalloc-]
> root 32731 0.0 0.0 0 0 ? S Apr06 0:05
[btrfs-endio-met]
>
>
The truncate stuff has changed drastically recently, can you try again
on a newer kernel? Thanks,
Josef
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html