Re: [RFC PATCH 1/3] block: add rq->complete_io hook for request stacking

[Hannes Reinecke <hare@suse.de>]

Hi Kiyoshi,

Kiyoshi Ueda wrote:
> Hi Hannes,
> 
> On Mon, 03 Mar 2008 17:24:50 +0100, Hannes Reinecke wrote:
>> Hi Kiyoshi,
>>
>> Kiyoshi Ueda wrote:
>>> This patch adds ->complete_io() hook for request stacking.
>>> Request stacking drivers (such as request-based dm) can set
>>> a callback for completion.
>>> (The hook is not called in blk_end_io(), since request-based dm uses
>>>  it for clone completion in the following appendix patches.)
>>>
>> [ .. ]
>> I would rather have rq->complete_io() to be pointing to blk_end_io in the
>> default case, this way rq->complete_io() would always be valid and we
>> would be saving us the if() clause.
> 
> Thank you for the comment.
> 
> I'm thinking about an idea come from Mike and Jens after the session
> at LSF'08:
>   o stacking driver clones both bio and request and uses
>     bio->bi_end_io and rq->end_io
>   o stacking driver uses blk_complete_request() in rq->end_io
>     so that stacking driver can work without queue lock
> With this idea, we don't need to add the new hook to request,
> and we can use request stacking on drivers using __blk_end_request().
> 
Huh? I can't really imagine how this should work. If you start using
bio->bi_end_io() for stacking usage you'll end up with doing weird thing
with the bios ... Can you elaborate a bit more here?

> Currently, I'm trying to convert request-based dm to use the idea above.
> I'll post the new version as a RFC in the near future.
> 
Oh, cool. Looking forward to it.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		      zSeries & Storage
hare@suse.de			      +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Markus Rex, HRB 16746 (AG Nürnberg)