Re: Bcache upstreaming

[Mike Snitzer <snitzer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>]

On Fri, Feb 01 2013 at 11:18am -0500,
Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:

> Hello, Kent.
> 
> On Fri, Feb 01, 2013 at 08:15:47AM -0800, Kent Overstreet wrote:
> > Eww, not a flag. I meant a completely separate functions, rip out the
> > refcounting entirely and have the refcounting-manipulating versions
> > available as
> 
> No, I mean, internally there needs to be a way whether the currently
> existing linkage is from the old or new interface for exclusive close
> to be able to decide whether it can remove it or not.  Anyways, let's
> wait for Mike for now.

The need for the same holder refcount is like I thought: a DM device's
active and inactive tables can open the same block devices.  I looked at
the prospect of pushing the refcount into DM but I don't think it is as
clean as having the bd_holder_disk struct continue to provide the
refcount.  Pushing it into DM would still require an explicit call to
bd_unlink_disk_holder.

The refcount is really pretty benign; so I'm inclined to leave things as
is.