On Mon, 13 Aug 2012 10:04:54 +0800 Shaohua Li <shli@xxxxxxxxxx> wrote:
> On Mon, Aug 13, 2012 at 11:50:51AM +1000, NeilBrown wrote:
> > On Fri, 10 Aug 2012 10:51:19 +0800 Shaohua Li <shli@xxxxxxxxxxxx> wrote:
> >
> > > @@ -4094,6 +4159,19 @@ static void make_request(struct mddev *m
> > > bi->bi_next = NULL;
> > > bi->bi_phys_segments = 1; /* over-loaded to count active stripes */
> > >
> > > + /* block layer doesn't correctly do alignment even we set correct alignment */
> > > + if (unlikely(bi->bi_rw & REQ_DISCARD)) {
> > > + int stripe_sectors = conf->chunk_sectors *
> > > + (conf->raid_disks - conf->max_degraded);
> >
> > This isn't right when an array is being reshaped.
> > I suspect that during a reshape we should only attempt DISCARD on the part of
> > the array which has already been reshaped. On the other section we can
> > either fail the discard (is that a good idea?) or write zeros.
>
> I had a check in below for-loop for reshape, is it enough? If not, I'd like
> just ignore discard request for reshape. We force discard_zero_data to be 0, so
> should be ok.
Yes, you are right - that is sufficient. I hadn't read it properly.
>
> I'll fix other two issues. Will repost the raid5 discard patches later.
thanks,
NeilBrown
Attachment:
signature.asc
Description: PGP signature
