kern/166851: Copying directory from the mounted UFS disk image hangs in DL+ vnread state (source and destination both on the same ZFS)

[Yuri]

>Number:         166851
>Category:       kern
>Synopsis:       Copying directory from the mounted UFS disk image hangs in DL+ vnread state (source and destination both on the same ZFS)
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Wed Apr 11 18:00:27 UTC 2012
>Closed-Date:
>Last-Modified:
>Originator:     Yuri
>Release:        9.0-STABLE amd64
>Organization:
n/a
>Environment:
>Description:
I have 82GB UFS image file (ufs-snapshot.img) mounted on some directory ufs-snapshot.mount. (mount /dev/`mdconfig -a -t vnode -f ufs-snapshot.img` ufs-snapshot.mount) Both original file ufs-snapshot.img and ufs-snapshot.mount are on the same ZFS disk.

Command 'cp -R ufs-snapshot.mount/usr other-dir/' hanged in the middle with DL+ status:
$ ps ax | grep cp
73635  10  DL+     0:12.19 cp -R ufs-snapshot.mount/usr other-dir/
'top' shows it in vnread state:
73635 root            1  20    0 10084K  2672K vnread  1   0:12  0.00% cp

When I ran 'ls' in the same mounted directory it hanged too with D+ status:
$ ps ax | grep ls
75882   2  D+      0:00.00 ls ufs-snapshot.mount/

Subsequent 'ls' of this ZFS disk's top directory hangs in the same way. And after few hours system freezes.

It appears that cp -R creates the gridlock condition when it copies from UFS contained in the image file within some ZFS disk into the same ZFS disk.

>How-To-Repeat:

>Fix:


>Release-Note:
>Audit-Trail:
>Unformatted: