Re: [PATCH] aoe: end barrier bios with EOPNOTSUPP

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

On Thu, Sep 10 2009, Ed Cashin wrote:
> aoe: end barrier bios with EOPNOTSUPP
> 
> BugLink: http://bugzilla.kernel.org/show_bug.cgi?id=13942
> 
> Bruno Premont noticed that aoe throws a BUG during umount of an XFS in
> 2.6.31:
> 
> [ 5259.349897] aoe: bi_io_vec is NULL
> [ 5259.349940] ------------[ cut here ]------------
> [ 5259.349958] kernel BUG at /usr/src/linux-2.6/drivers/block/aoe/aoeblk.c:177!
> [ 5259.349990] invalid opcode: 0000 [#1]
> 
> The bio in question is a barrier.  Jens Axboe suggested that such bios
> need to be recognized and ended with -EOPNOTSUPP by any driver that
> provides its own ->make_request_fn handler and does not handle
> barriers.
> 
> In testing the changes below eliminate the BUG.
> 
> Signed-off-by: Ed L. Cashin <ecashin@coraid.com>
> ---
>  drivers/block/aoe/aoeblk.c |    3 +++
>  1 files changed, 3 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
> index 2307a27..22efb33 100644
> --- a/drivers/block/aoe/aoeblk.c
> +++ b/drivers/block/aoe/aoeblk.c
> @@ -172,6 +172,9 @@ aoeblk_make_request(struct request_queue *q, struct bio *bio)
>  		BUG();
>  		bio_endio(bio, -ENXIO);
>  		return 0;
> +	} else if (bio_barrier(bio)) {
> +		bio_endio(bio, -EOPNOTSUPP);
> +		return 0;
>  	} else if (bio->bi_io_vec == NULL) {
>  		printk(KERN_ERR "aoe: bi_io_vec is NULL\n");
>  		BUG();
> -- 
> 1.5.6.5

I have applied this with a note that proper barrier support is coming.

-- 
Jens Axboe