Re: [RFC] [PATCH] A clean approach to writeout throttling

[Andrew Morton <akpm@linux-foundation.org>]

On Thu, 6 Dec 2007 16:04:41 -0800
Daniel Phillips <phillips@phunq.net> wrote:

> The runner up key idea is that we will gain a notion of "block device 
> stack" (or block stack for short, so that we may implement block 
> stackers) which for the time being will simply be Device Mapper's 
> notion of device stack, however many warts that may have.  It's there 
> now and we use it for ddsnap.

Perhaps all we need to track is the outermost point?

submit_bio(...)
{
	bool remove_the_rq = false;

	...
	if (current->the_rq == NULL) {
		current->the_rq = rq;
		remove_the_rq = true;
	}
	...
	if (remove_the_rq)
		current->the_rq = NULL;
}

?