On Wed, Aug 23, 2017 at 11:23:38AM -0700, Shaohua Li wrote:
> On Wed, Aug 23, 2017 at 07:10:28PM +0200, Christoph Hellwig wrote:
> > The block layer always remaps partitions before calling into the
> > ->make_request methods of drivers. Thus the call to get_start_sect in
> > in_chunk_boundary will always return 0 and can be removed.
> >
> > Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> > ---
> > drivers/md/raid5.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
> > index 0fc2748aaf95..d687aeb1b538 100644
> > --- a/drivers/md/raid5.c
> > +++ b/drivers/md/raid5.c
> > @@ -5092,10 +5092,12 @@ static int raid5_congested(struct mddev *mddev, int bits)
> > static int in_chunk_boundary(struct mddev *mddev, struct bio *bio)
> > {
> > struct r5conf *conf = mddev->private;
> > - sector_t sector = bio->bi_iter.bi_sector + get_start_sect(bio->bi_bdev);
> > + sector_t sector = bio->bi_iter.bi_sector;
> > unsigned int chunk_sectors;
> > unsigned int bio_sectors = bio_sectors(bio);
> >
> > + WARN_ON_ONCE(bio->bi_partno);
> > +
Meh, of course bi_partno is only added a few patches later, so this
breaks bisectability. But given that the patch is already in there's
probably nothing I can do here.
