scsi: return correct blkprep status code in case scsi_init_io() fails.
authorJohannes Thumshirn <jthumshirn@suse.de>
Wed, 12 Apr 2017 07:21:19 +0000 (09:21 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 20 Jun 2020 08:24:07 +0000 (10:24 +0200)
commit07e0b16899ebc1f8bf0558561869be1612decd53
treeb898ffb8ff7d0cd186a3b670d09d0608f4ca12a0
parent72787b04621661e9e9a63af4215749a938233e2a
scsi: return correct blkprep status code in case scsi_init_io() fails.

commit e7661a8e5ce10b5321882d0bbaf3f81070903319 upstream.

When instrumenting the SCSI layer to run into the
!blk_rq_nr_phys_segments(rq) case the following warning emitted from the
block layer:

blk_peek_request: bad return=-22

This happens because since commit fd3fc0b4d730 ("scsi: don't BUG_ON()
empty DMA transfers") we return the wrong error value from
scsi_prep_fn() back to the block layer.

[mkp: silenced checkpatch]

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Fixes: fd3fc0b4d730 scsi: don't BUG_ON() empty DMA transfers
Cc: <stable@vger.kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
[iwamatsu: - backport for 4.4.y and 4.9.y
    - Use rq->nr_phys_segments instead of blk_rq_nr_phys_segments]
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/scsi/scsi_lib.c