Re: [dm-devel] [PATCH v7 7/9] block: Add bio_clone_bioset(), bio_clone_kmalloc()

[Tejun Heo <tj kernel org>]

Hello,

On Tue, Aug 28, 2012 at 03:05:32PM -0700, Kent Overstreet wrote:
> On Tue, Aug 28, 2012 at 01:44:01PM -0700, Tejun Heo wrote:
> > On Tue, Aug 28, 2012 at 10:37:34AM -0700, Kent Overstreet wrote:
> > > +static inline struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask)
> > > +{
> > > +	return bio_clone_bioset(bio, gfp_mask, fs_bio_set);
> > > +}
> > > +
> > ...
> > > +static inline struct bio *bio_clone_kmalloc(struct bio *bio, gfp_t gfp_mask)
> > > +{
> > > +	return bio_clone_bioset(bio, gfp_mask, NULL);
> > > +
> > > +}
> > 
> > Do we really need these wrappers?  I'd prefer requiring users to
> > explicit choose @bioset when cloning.
> 
> bio_clone() is an existing api, I agree but I'd prefer to convert
> existing users in a separate patch and when I do that I want to spend
> some time actually looking at the existing code instead of doing the
> conversion blindly (at least some of the existing users are incorrect
> and I'll have to add bio_sets for them).

Aren't there like three users in kernel?  If you wanna clean it up
later, that's fine too but I don't think it would make much difference
either way.

Thanks.

-- 
tejun