> ok, I busted that when I did the UBC stuff then.
> it was not my intent to protect anything with the global lock.
> it would be good for us to avoid making this situation worse by
> adding more usages that are protected only by the global lock.
> 
> -Chuck

then i think it's appropriate to introduce a small lock in ufs inode.
protecting it by VOP_LOCK is difficult because of
the locking order constraint you mentioned.

YAMAMOTO Takashi