Re: [PATCH 1/2] libata: get rid of ATA_MAX_QUEUE loop in ata_qc_complete_multiple()

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

On Thu, Oct 23 2008, Tejun Heo wrote:
> Jens Axboe wrote:
> > @@ -4811,16 +4811,19 @@ int ata_qc_complete_multiple(struct ata_port *ap, u32 qc_active)
> >               return -EINVAL;
> >       }
> >
> > -     for (i = 0; i < ATA_MAX_QUEUE; i++) {
> > +     while (done_mask) {
> >               struct ata_queued_cmd *qc;
> > +             unsigned int next = __ffs(done_mask);
> >
> > -             if (!(done_mask & (1 << i)))
> > -                     continue;
> > -
> > -             if ((qc = ata_qc_from_tag(ap, i))) {
> > +             qc = ata_qc_from_tag(ap, i + next);
> > +             if (qc) {
> >                       ata_qc_complete(qc);
> >                       nr_done++;
> >               }
> > +             if (++next >= ATA_MAX_QUEUE)
> > +                     break;
> > +             i += next;
> > +             done_mask >>= next;
> 
> Shouldn't this be...
> 
>         i += next + 1;
>         if (i >= ATA_MAX_QUEUE)
>                 break;
> 
> or better...
> 
> while (done_mask) {
>         struct ata_queued_cmd *qc;
>         unsigned int next = __ffs(done_mask);
> 
>         tag += next;
>         if ((qc = ata_qc_from_tag(ap, tag))) {
>                 ata_qc_complete(qc);
>                 nr_done++;
>         }
>         next++;
>         tag += next;
>         done_mask >>= next;
> }
> 
> done_mask is guaranteed to be zero at when tag reaches ATA_MAX_QUEUE.

That does indeed look a lot cleaner. I think it was rewritten at some
point and kept some of the logic for not passing 0 into __ffs, but it's
clearly pointless now.

I'll send out a revised patch when it's tested.

-- 
Jens Axboe