Subject: Re: detachable device: stuck in biowait
To: Bill Studenmund <wrstuden@netbsd.org>
From: Ferry Sutanto <fsutanto@yahoo.com>
List: tech-kern
Date: 01/29/2002 21:28:07
--- Bill Studenmund <wrstuden@netbsd.org> wrote:
> > I notice that although config_detach has finished,
> the
> > kernel still tries to write by calling either
> vn_write
> > or bwrite. I put little hack (a global flag ==
> true if
> > device has gone) which will bail out if the device
> has
> > gone and I called brelse to clean up the buffer.
> FTP
> > exit with EIO error, but the csh (shell) hangs.
> Does
> > anybody have this problem before ?
>
> That sounds like the right kind of premise - if the
> device is gone, in the
> i/o routine, fail.
>
> Not sure why the csh fails. Where was its current
> directory?
The current directory is the mounted directory of the
compact flash. I am suspecting that there is a buffer
which I didn't clean correctly ?
Another question. By printing out the address of the
vnode, I notice that the vnode that returned by
vfinddev() routine is different from the vnode used
for writing all during the write. Is this the
suspected behaviour ?
Thanks!
Ferry
__________________________________________________
Do You Yahoo!?
Great stuff seeking new owners in Yahoo! Auctions!
http://auctions.yahoo.com