On Sat, May 16, 2009 at 08:29:54AM +0000, YAMAMOTO Takashi wrote:
> Module Name: src
> Committed By: yamt
> Date: Sat May 16 08:29:54 UTC 2009
>
> Modified Files:
> src/sys/kern: vfs_subr.c
> src/sys/sys: vnode.h
>
> Log Message:
> put a flag bit into v_usecount to prevent vtryget during getcleanvnode.
> this fixes the following deadlock.
>
> a thread doing getcleanvnode:
> pick a vnode
> acqure v_interlock
> v_usecount++
> call vclean
>
> now, another thread doing cache_lookup:
> picks the vnode
> vtryget succeed
> vn_lock succeed
>
> now in vclean:
> set VI_XLOCK (too late to be noticed by the competing thread)
> wait on the vnode lock (this might violate locking order)
>
> the use of a flag bit was suggested by Andrew Doran. PR/41374.
Hi,
do you think it could also help with kern/41417 ?
cache_lookup seems also to be involved in this deadlock ...
--
Manuel Bouyer <bou...@antioche.eu.org>
NetBSD: 26 ans d'experience feront toujours la difference
--
