On Tue, Sep 05, 2006 at 12:55:02PM +0000, YAMAMOTO Takashi wrote:
> The following reply was made to PR kern/34293; it has been noted by GNATS.
> 
> From: yamt@mwd.biglobe.ne.jp (YAMAMOTO Takashi)
> To: mlelstv@serpens.de
> Cc: gnats-bugs@NetBSD.org, kern-bug-people@NetBSD.org,
> 	gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org
> Subject: Re: PR/34293 CVS commit: src/sys/dev
> Date: Tue,  5 Sep 2006 21:53:09 +0900 (JST)
> 
>  > > while vndthread is working on behalf of the pagedaemon here,
>  > > it is not a pagedaemon.  so it can't get special treatment for pagedaemon
>  > > from buffer cache etc and can deadlock.
>  > 
>  > In my case it is definitely not working on behalf of the pagedaemon.
>  
>  did you have plenty of free memory?  it wasn't clear from the PR.

I can't say. At that time the machine was hardly responsive as
all disk I/O was stalled.

>  you said "separate pool for nested I/O buffers" worked.

It worked some times. But most times the processes were stuck
waiting on getnewblk.

-- 
                                Michael van Elst
Internet: mlelstv@serpens.de
                                "A potential Snark may lurk in every tree."