Re: [PATCH] io_uring/rsrc: send exact nr_segs for fixed buffer

[Keith Busch <kbusch@kernel.org>]

On Thu, Apr 17, 2025 at 01:27:55AM +0530, Nitesh Shetty wrote:
> > +       /*
> > +        * Offset trimmed front segments too, if any, now trim the tail.
> > +        * For is_kbuf we'll iterate them as they may be different sizes,
> > +        * otherwise we can just do straight up math.
> > +        */
> > +       if (len + offset < imu->len) {
> > +               bvec = iter->bvec;
> > +               if (imu->is_kbuf) {
> > +                       while (len > bvec->bv_len) {
> > +                               len -= bvec->bv_len;
> > +                               bvec++;
> > +                       }
> > +                       iter->nr_segs = bvec - iter->bvec;
> > +               } else {
> > +                       size_t vec_len;
> > +
> > +                       vec_len = bvec->bv_offset + iter->iov_offset +
> > +                                       iter->count + ((1UL << folio_shift) - 1);
> > +                       iter->nr_segs = vec_len >> folio_shift;
> > +               }
> > +       }
> >         return 0;
> >  }
> This might not be needed for is_kbuf , as they already update nr_seg
> inside iov_iter_advance.

iov_iter_advance only handles the front segs. We still need something
for any trailing segments that are not part of this mapping request.