Re: umount oops

2008-07-25 Thread Lukas Vacek
sorry, I made a typo in the testcase (the second mount)
Basically, it might be enough to mount two different btrfs filesystems
to two different locations, umount one of them and watch
/var/log/kern.log for the oops

dd if=/dev/zero of=mountme bs=4k count=10
dd if=/dev/zero of=mountme2 bs=4k count=10
mkfs.btrfs mountme
mkfs.btrfs mountme2
mkdir loop loop2
mount -o loop mountme loop
mount -o loop mountme2 loop2
umount loop
# wait a moment

On 7/24/08, Lukas Vacek [EMAIL PROTECTED] wrote:
 Hi,

 I tried very promising btrfs to test it a little and I experienced a
 little bug in implementation. I'm not sure where the bug lies however
 this works quite well to reproduce the problem:

 dd if=/dev/zero of=mountme bs=4k count=10
 dd if=/dev/zero of=mountme2 bs=4k count=10
 mkfs.btrfs mountme
 mkfs.btrfs mountme2
 mkdir loop loop2
 mount -o loop mountme loop
 mount -o loop mountme loop2
 umount loop
 # wait a moment

 maybe SMP machine will be necessary to experience the same

 thanks for the (otherwise ;-)) grea work and have a nice day,
 Lukas V.

 the interesting part of log goes next:

 Jul 24 22:44:00 minerva kernel: [ 1478.326985] device fsid
 5442602040543dd1-d32561672012f7a2 devid 1 transid 12 /dev/loop1
 Jul 24 22:44:54 minerva kernel: [ 1532.882212] Unable to handle kernel
 paging request at 7effdc171f47 RIP:
 Jul 24 22:44:54 minerva kernel: [ 1532.882256]  [wq_per_cpu+0x15/0x20]
 wq_per_cpu+0x15/0x20
 Jul 24 22:44:54 minerva kernel: [ 1532.882405] PGD 0
 Jul 24 22:44:54 minerva kernel: [ 1532.882476] Oops:  [1] SMP
 Jul 24 22:44:54 minerva kernel: [ 1532.882572] CPU 1
 Jul 24 22:44:54 minerva kernel: [ 1532.882641] Modules linked in: loop
 binfmt_misc af_packet i915 drm rfcomm l2cap bluetooth ppdev ipv6
 acpi_cpufreq cpufreq_ondemand cpufreq_stats freq_table
 cpufreq_userspace cpufreq_conservative cpufreq_powersave video output
 container sbs sbshc dock battery iptable_filter ip_tables x_tables
 aes_x86_64 dm_crypt dm_mod ac btrfs libcrc32c lp snd_hda_intel
 snd_hwdep snd_pcm_oss snd_pcm snd_page_alloc snd_mixer_oss
 snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event
 snd_seq snd_timer iTCO_wdt iTCO_vendor_support snd_seq_device button
 snd evdev shpchp pci_hotplug intel_agp soundcore parport_pc parport
 pcspkr ext3 jbd mbcache usbhid hid sg sd_mod ata_piix pata_acpi floppy
 ehci_hcd ata_generic libata scsi_mod r8169 uhci_hcd usbcore thermal
 processor fan fbcon tileblit font bitblit softcursor fuse
 Jul 24 22:44:54 minerva kernel: [ 1532.885056] Pid: 3959, comm:
 btrfs-transacti Not tainted 2.6.24-19-generic #1
 Jul 24 22:44:54 minerva kernel: [ 1532.885117] RIP:
 0010:[wq_per_cpu+0x15/0x20]  [wq_per_cpu+0x15/0x20]
 wq_per_cpu+0x15/0x20
 Jul 24 22:44:54 minerva kernel: [ 1532.885224] RSP:
 0018:81003e6a1c08  EFLAGS: 00010246
 Jul 24 22:44:54 minerva kernel: [ 1532.885281] RAX: 7effdc171f3f
 RBX: 81003db0b770 RCX: 
 Jul 24 22:44:54 minerva kernel: [ 1532.885342] RDX: 
 RSI: 0001 RDI: 810023e8e940
 Jul 24 22:44:54 minerva kernel: [ 1532.885404] RBP: 81003db0a000
 R08:  R09: 882c9d80
 Jul 24 22:44:54 minerva kernel: [ 1532.885465] R10: 81003653ade0
 R11:  R12: 81003653ade0
 Jul 24 22:44:54 minerva kernel: [ 1532.885526] R13: 0001
 R14: 810014366680 R15: 
 Jul 24 22:44:54 minerva kernel: [ 1532.885588] FS:
 () GS:81003e401700()
 knlGS:
 Jul 24 22:44:54 minerva kernel: [ 1532.885671] CS:  0010 DS: 0018 ES:
 0018 CR0: 8005003b
 Jul 24 22:44:54 minerva kernel: [ 1532.885728] CR2: 7effdc171f47
 CR3: 225cc000 CR4: 06e0
 Jul 24 22:44:54 minerva kernel: [ 1532.885789] DR0: 
 DR1:  DR2: 
 Jul 24 22:44:54 minerva kernel: [ 1532.885851] DR3: 
 DR6: 0ff0 DR7: 0400
 Jul 24 22:44:54 minerva kernel: [ 1532.885912] Process btrfs-transacti
 (pid: 3959, threadinfo 81003e6a, task 81003c74c7e0)
 Jul 24 22:44:54 minerva kernel: [ 1532.885996] Stack:
 8024f97c 81001e3db480 882ca4ce 882c9d80
 Jul 24 22:44:54 minerva kernel: [ 1532.886193]  01c08fff
 810014366680 81003653ad40 1000
 Jul 24 22:44:54 minerva kernel: [ 1532.886364]  81003ee140e8
 0001 882e19cf 01c09000
 Jul 24 22:44:54 minerva kernel: [ 1532.886495] Call Trace:
 Jul 24 22:44:54 minerva kernel: [ 1532.886587]
 [shpchp:queue_work+0x2c/0x50] queue_work+0x2c/0x50
 Jul 24 22:44:54 minerva kernel: [ 1532.886665]
 [btrfs:btrfs_wq_submit_bio+0xbe/0xf0]
 :btrfs:btrfs_wq_submit_bio+0xbe/0xf0
 Jul 24 22:44:54 minerva kernel: [ 1532.886743]
 [btrfs:__btree_submit_bio_hook+0x0/0x60]
 :btrfs:__btree_submit_bio_hook+0x0/0x60
 Jul 24 22:44:54 minerva kernel: [ 1532.886826]
 [btrfs:submit_one_bio+0xcf/0x120] :btrfs:submit_one_bio+0xcf/0x120

Re: umount oops

2008-07-25 Thread Chris Mason
On Fri, 2008-07-25 at 13:14 +0200, Lukas Vacek wrote:
 sorry, I made a typo in the testcase (the second mount)
 Basically, it might be enough to mount two different btrfs filesystems
 to two different locations, umount one of them and watch
 /var/log/kern.log for the oops
 

Thanks for this bug report, which Btrfs version was in use?

-chris


--
To unsubscribe from this list: send the line unsubscribe linux-btrfs in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html


Re: umount oops

2008-07-25 Thread Chris Mason
On Fri, 2008-07-25 at 15:54 +0200, Lukas Vacek wrote:
 the newest in the mercurial repo
 
 changeset:   558:9da425337329
 tag: tip

This should be fixed by the unstable tree, the transaction work queues
were not properly being torn down.

-chris


--
To unsubscribe from this list: send the line unsubscribe linux-btrfs in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html