Re: SG does not ignore dxferp (direct io + mmap)

From: "Ewan D. Milne" <emilne@redhat.com>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	Johannes Thumshirn <jthumshirn@suse.de>,
	Laurence Oberman <loberman@redhat.com>,
	Eyal Ben David <bdeyal@gmail.com>,
	dgilbert@interlog.com, linux-scsi@vger.kernel.org
Subject: Re: SG does not ignore dxferp (direct io + mmap)
Date: Fri, 02 Dec 2016 15:37:07 -0500	[thread overview]
Message-ID: <1480711027.28416.200.camel@localhost.localdomain> (raw)
In-Reply-To: <yq18trzhink.fsf@sermon.lab.mkp.net>

On Thu, 2016-12-01 at 08:40 -0500, Martin K. Petersen wrote:
> >>>>> "Ewan" == Ewan D Milne <emilne@redhat.com> writes:
...
> Specifically, the problem appears to be caused by the removal of
> the setting of bio->bi_bdev, which would previously be set to NULL.
> If I add:
> 
> diff --git a/block/bio.c b/block/bio.c
> index 0723d4c..ecac37b 100644
> --- a/block/bio.c
> +++ b/block/bio.c
> @@ -1351,6 +1351,7 @@ struct bio *bio_map_user_iov(struct request_queue
> *q,
>         if (iter->type & WRITE)
>                 bio->bi_rw |= REQ_WRITE;
>  
> +       bio->bi_bdev = NULL;
>         bio->bi_flags |= (1 << BIO_USER_MAPPED);
>  
>         /*
> 
> Ewan> The test passes (no zero byte corruption).
> 

Adding this seemed to work on top of the commit commit id mentioned
above (37f19e57a0de3c4a3417aa13ff4d04f1e0dee4b3)
but does not help on 4.7, because the test case now takes the
bio_copy_user_iov() code path rather than bio_map_user_iov().

-Ewan