Further info:
I tested several versions of old kernel starts from v4.7, and they all
failed on two of my physical machines.
But strangely, they all passed in KVM guests using virtio.
Not sure if it's related to the device size (over 50G for each device in
physical machines, while less than 10G for each in VM)
And the profile causing the problem is, unsurprisingly, RAID5 and RAID6.
For other profile it seems to be OK.
Thanks,
Qu
At 12/13/2016 11:15 AM, Qu Wenruo wrote:
Hi
When testing Chris' for-linus-4.10 branch.
I found that even at the branch base, v4.9-rc7, btrfs can't pass
quite a lot of scrub tests, including btrfs/011 and btrfs/069.
The btrfs/069 will fail 100%, with the following back trace:
general protection fault: 0000 [#1] SMP
Modules linked in: btrfs(O) xor zlib_deflate raid6_pq
x86_pkg_temp_thermal ext4 jbd2 mbcache e1000e efivarfs [last unloaded:
btrfs]
CPU: 3 PID: 5300 Comm: kworker/u8:4 Tainted: G O 4.9.0-rc7+
#20
Hardware name: FUJITSU ESPRIMO P720/D3221-A1, BIOS V4.6.5.4 R1.17.0 for
D3221-A1x 03/06/2014
Workqueue: btrfs-endio-raid56 btrfs_endio_raid56_helper [btrfs]
task: ffff88008dbcb740 task.stack: ffffc90001230000
RIP: 0010:[<ffffffff813a2fa8>] [<ffffffff813a2fa8>]
generic_make_request_checks+0x198/0x5a0
RSP: 0018:ffffc90001233b08 EFLAGS: 00010202
RAX: 0000000000000000 RBX: ffff88007f963228 RCX: 0000000000000001
RDX: 0000000080000000 RSI: 0000000000000000 RDI: 6b6b6b6b6b6b6b6b
RBP: ffffc90001233b68 R08: 00000000868a9b14 R09: eab761b200000000
R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000040
R13: 0000000000000004 R14: ffff88008dbc5a88 R15: 0000000000000010
FS: 0000000000000000(0000) GS:ffff880119e00000(0000)
knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000019c7058 CR3: 00000001150b4000 CR4: 00000000001406e0
Stack:
0000000000000002 ffffffff813a4cff ffffffff00000000 0000000000000296
0000000000000292 ffff88008dbcb740 ffffffff00000003 ffff88007f963228
00000000ffffffff 0000000000000004 ffff88008dbc5a88 0000000000000010
Call Trace:
[<ffffffff813a4cff>] ? generic_make_request+0xcf/0x290
[<ffffffff813a4c54>] generic_make_request+0x24/0x290
[<ffffffff813a4cff>] ? generic_make_request+0xcf/0x290
[<ffffffff813a4f2e>] submit_bio+0x6e/0x120
[<ffffffffa087279d>] ? page_in_rbio+0x4d/0x80 [btrfs]
[<ffffffffa08737d0>] ? rbio_orig_end_io+0x80/0x80 [btrfs]
[<ffffffffa0873e31>] finish_rmw+0x401/0x550 [btrfs]
[<ffffffffa0874fc6>] validate_rbio_for_rmw+0x36/0x40 [btrfs]
[<ffffffffa087504d>] raid_rmw_end_io+0x7d/0x90 [btrfs]
[<ffffffff8139c536>] bio_endio+0x56/0x60
[<ffffffffa07e6e5c>] end_workqueue_fn+0x3c/0x40 [btrfs]
[<ffffffffa08285bf>] btrfs_scrubparity_helper+0xef/0x610 [btrfs]
[<ffffffffa0828b9e>] btrfs_endio_raid56_helper+0xe/0x10 [btrfs]
[<ffffffff810ec8df>] process_one_work+0x2af/0x720
[<ffffffff810ec85b>] ? process_one_work+0x22b/0x720
[<ffffffff810ecd9b>] worker_thread+0x4b/0x4f0
[<ffffffff810ecd50>] ? process_one_work+0x720/0x720
[<ffffffff810ecd50>] ? process_one_work+0x720/0x720
[<ffffffff810f39d3>] kthread+0xf3/0x110
[<ffffffff810f38e0>] ? kthread_park+0x60/0x60
[<ffffffff81857647>] ret_from_fork+0x27/0x40
Code: 00 00 0f 1f 44 00 00 65 8b 05 9d 71 c6 7e 89 c0 48 0f a3 05 c3 13
b8 00 0f 92 c3 0f 82 dd 02 00 00 bb 01 00 00 00 e9 8c 00 00 00 <48> 8b
47 08 48 8b 40 50 48 c1 f8 09 48 85 c0 0f 84 99 fe ff ff
RIP [<ffffffff813a2fa8>] generic_make_request_checks+0x198/0x5a0
RSP <ffffc90001233b08>
Is this a known bug or a new one caused by the block layer change?
Thanks,
Qu
--
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
