Re: [PATCH 002 of 5] Replace bio_data with blk_rq_data

[Jens Axboe <jens.axboe@oracle.com>]

On Thu, Aug 16 2007, Neil Brown wrote:
> On Thursday August 16, jens.axboe@oracle.com wrote:
> > On Thu, Aug 16 2007, NeilBrown wrote:
> > > 
> > > Almost every call to bio_data is for the first bio
> > > in a request.  A future patch will add some accounting
> > > information to 'struct request' which will need to be
> > > used to find the start of the request in the bio.
> > > So replace bio_data with blk_rq_data which takes a 'struct request *'
> > > 
> > > The one exception is in dm-emc were using
> > >    page_address(bio->bi_io_vec[0].bv_page);
> > > is appropriate.
> > 
> > This (and 3+4) just look like preparatory patches if we want to merge
> > the full patchset, not bug fixes. I seem to recall you had more bug
> > fixes or cleanups in your patchset, maybe I was mistaken. So nak for now
> > for 2-4, I'd apply 5 but it depends on the previous.
> > 
> 
> I don't remember other bug fixes, but I'll look through and check.

OK, I'm properly remembering incorrectly then.

> 2 and 3 are very simple changes that - I think - make it clearer what
> is happening.

To be honest, I don't see much win in using blk_rq_data() over
bio_data() at all. I'd much much rather just see it go away!

> And I felt 5 was a sufficient improvement to justify it and 4...

5 is nice, I would like to apply that :-)

> I'll go hunting and see what other preliminaries I can find.

Thanks!

-- 
Jens Axboe