On Thu, 08 Sep 2005 10:22:50 +0900
YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp> wrote:

> > YAMAMOTO please review it again (reload the URL).
> 
> - no one seems to allocate tbufq.

Fixed.
 
> - don't assume that splbio() is a right way to protect bufq.
>   how to protect bufq is up to each drivers.
> 
> - copying bufq_state is evil.
>   (although i know that there is a driver which does it..)
>   it's the reason why i suggested to refine bufq interface.

I'm copying bufq_state for the moment, we could fix that in
the future.
 
> - don't assume BUFQ_SORT_RAWBLOCK.

I'm working on it plus passing them by-name rather than by-bits as you
suggested.

> - why disk_ prefix rather than bufq_ ?

<sys/bufq.h> is only for the kernel, not userland. I had to move the
BUFQ_ definitions into disk.h for dkctl(8), and all functions in there
are prefixed with disk_ (for consistency).