Re: [PATCH 1/2] Validate the DataOut length in WRITE_SAME_1*

FUJITA Tomonori <[email protected]> Wed, 05 Oct 2016 06:49:17 +0900 (JST)
Newsgroups org.kernel.vger.stgt
Message-ID <[email protected]>
On Sun,  2 Oct 2016 09:04:01 -0700
Ronnie Sahlberg <[email protected]> wrote:

> WRITE_SAME_1* takes either no data at all or exactly one block of data
> as DataOut. Check for this and fail the cdb if the DataOut size differs.
> 
> (Some initiators use DataOut size == 0 (primarily together with unmap)
> to mean "use a block all filled with 0" so that is why I think we
> should be liberal and allow size==0 too here.)
> 
> Signed-off-by: Ronnie Sahlberg <[email protected]>
> ---
>  usr/sbc.c | 8 ++++++++
>  1 file changed, 8 insertions(+)

Applied both, thanks!