[PATCH 3/3] nvme-cli: fix nvme-connect-all using hostnqn

[J Freyensee]

On Fri, 2016-10-21 at 06:22 -0700, Christoph Hellwig wrote:
> On Wed, Oct 19, 2016 at 12:43:07PM -0700, Jay Freyensee wrote:
> > 
> > The example in the man pages:
> > 
> > nvme connect-all --transport=rdma --traddr=192.168.1.3 \
> > --hostnqn=host1-rogue-nqn
> > 
> > fails because nvme-cli fails to actually use hostnqn upon
> > connect.  This patch fixes that.
> > 
> > Signed-off-by: Jay Freyensee <james_p_freyensee at linux.intel.com>
> > ---
> >  fabrics.c | 6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/fabrics.c b/fabrics.c
> > index 51e424e..801fe4f 100644
> > --- a/fabrics.c
> > +++ b/fabrics.c
> > @@ -540,6 +540,12 @@ static int connect_ctrl(struct
> > nvmf_disc_rsp_page_entry *e)
> >  		return -EINVAL;
> >  	p += len;
> >  
> > +	if (cfg.hostnqn)
> > +	len = sprintf(p, ",hostnqn=%s", cfg.hostnqn);
> > +	if (len < 0)
> > +		return -EINVAL;
> > +	p += len;
> 
> The indentation here looks odd, I'd expect the line conditional
> on the if to be indented.  

Yes, I didn't do that indentation right, I'll adjust the code block per
comments.

Jay

> Also while the code should work fine as-is
> I'd move the len check and increment into the conditional, as they
> are
> pointless without doing the sprintf.

> 
> _______________________________________________
> Linux-nvme mailing list
> Linux-nvme at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-nvme