Re: [PATCH 05/13] block: Only clone bio vecs that are in use

[Kent Overstreet <koverstreet@google.com>]

On Fri, May 18, 2012 at 09:13:34AM -0700, Tejun Heo wrote:
> On Thu, May 17, 2012 at 10:59:52PM -0400, koverstreet@google.com wrote:
> > diff --git a/fs/bio.c b/fs/bio.c
> > index e2c0970..de0733e 100644
> > --- a/fs/bio.c
> > +++ b/fs/bio.c
> > @@ -435,8 +435,9 @@ EXPORT_SYMBOL(bio_phys_segments);
> >   */
> >  void __bio_clone(struct bio *bio, struct bio *bio_src)
> >  {
> > -	memcpy(bio->bi_io_vec, bio_src->bi_io_vec,
> > -		bio_src->bi_max_vecs * sizeof(struct bio_vec));
> > +	memcpy(bio->bi_io_vec,
> > +	       bio_iovec(bio_src),
> 
> Unnecessary line break.
> 
> > +	       bio_segments(bio_src) * sizeof(struct bio_vec));
> >  
> >  	/*
> >  	 * most users will be overriding ->bi_bdev with a new target,
> > @@ -445,10 +446,10 @@ void __bio_clone(struct bio *bio, struct bio *bio_src)
> >  	bio->bi_sector = bio_src->bi_sector;
> >  	bio->bi_bdev = bio_src->bi_bdev;
> >  	bio->bi_flags |= 1 << BIO_CLONED;
> > +	bio->bi_flags &= ~(1 << BIO_SEG_VALID);
> 
> Can probably be conditionalized on bi_idx?

I've never been that clear on the semantics of BIO_SEG_VALID. I'll defer
to you on that.