Bug 183689 - [lor] lock order reversal when umounting a ZFS disk.
Summary: [lor] lock order reversal when umounting a ZFS disk.
Status: Closed FIXED
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 10.0-BETA3
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-fs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-11-05 17:50 UTC by Edward O'Callaghan
Modified: 2013-11-06 08:00 UTC (History)
0 users

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Edward O'Callaghan 2013-11-05 17:50:00 UTC 
On reboot when unmounting a ZFS root there seems to be a lock order reversal between,
 1st 0xfffff800145fd7c8 zfs (zfs) @ /usr/src/sys/kern/vfs_mount.c:1237
 2nd 0xfffff800142ac240 devfs (devfs) @ /usr/src/sys/kern/vfs_subr.c:210

Since I can't understand the insane locking in vfs_mount.c:1237 I have no idea how to fix this.

----

Waiting (max 60 seconds) for system process `syncer' to stop...Syncing disks, vnodes remaining...0 0 0 0 0 0 0 done
Waiting (max 60 seconds) for system process `bufdaemon' to stop...done
All buffers synced.
lock order reversal:
 1st 0xfffff800145fd7c8 zfs (zfs) @ /usr/src/sys/kern/vfs_mount.c:1237
 2nd 0xfffff800142ac240 devfs (devfs) @ /usr/src/sys/kern/vfs_subr.c:2101
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe084d734460
kdb_backtrace() at kdb_backtrace+0x39/frame 0xfffffe084d734510
witness_checkorder() at witness_checkorder+0xd23/frame 0xfffffe084d7345a0
__lockmgr_args() at __lockmgr_args+0x86c/frame 0xfffffe084d7346d0
vop_stdlock() at vop_stdlock+0x3c/frame 0xfffffe084d7346f0
VOP_LOCK1_APV() at VOP_LOCK1_APV+0xf5/frame 0xfffffe084d734720
_vn_lock() at _vn_lock+0xab/frame 0xfffffe084d734790
vget() at vget+0x70/frame 0xfffffe084d7347e0
devfs_allocv() at devfs_allocv+0xfd/frame 0xfffffe084d734830
devfs_root() at devfs_root+0x43/frame 0xfffffe084d734860
dounmount() at dounmount+0x35a/frame 0xfffffe084d7348e0
vfs_unmountall() at vfs_unmountall+0x61/frame 0xfffffe084d734910
kern_reboot() at kern_reboot+0x548/frame 0xfffffe084d734980
sys_reboot() at sys_reboot+0x58/frame 0xfffffe084d7349a0
amd64_syscall() at amd64_syscall+0x265/frame 0xfffffe084d734ab0
Xfast_syscall() at Xfast_syscall+0xfb/frame 0xfffffe084d734ab0
--- syscall (55, FreeBSD ELF64, sys_reboot), rip = 0x80085f3fc, rsp = 0x7fffffffda38, rbp = 0x7fffffffdba0 ---

---

Fix: 

N/A.
How-To-Repeat: reboot
Comment 1 Mark Linimon freebsd_committer freebsd_triage 2013-11-06 04:19:56 UTC 
Responsible Changed
From-To: freebsd-bugs->freebsd-fs

Over to maintainer(s).
Comment 2 Andriy Gapon freebsd_committer freebsd_triage 2013-11-06 07:57:41 UTC 
This is a harmless LOR and happens with any fs, not just ZFS.

-- 
Andriy Gapon
Comment 3 Andriy Gapon freebsd_committer freebsd_triage 2013-11-06 07:59:09 UTC 
State Changed
From-To: open->closed

Not a problem.