On Wed, Sep 28, 2005 at 09:21:29AM +0900, YAMAMOTO Takashi wrote:
> > I'd like to include VOP_BMAP in there eventually, but we need to
> > figure out how to wean the existing external users from it first.
> > this might just result in a different form of VOP_BMAP, but the
> > current form requires too much knowledge by the caller of the
> > file system's internal locking.
> 
> exactly.
> 
> what's the status of your direct i/o patch?  :-)

according to my notes, it's basically done but not terribly useful yet
without some way to support concurrent writes.  as a step towards that,
I was looking into eliminating the use of recursive VOP_LOCK, but that
turned out to be hard enough that I lost momentum.

-Chuck