Date: Tue, 5 Dec 2017 20:56:07 +0300
From: Serge Semin <fancer.lancer@gmail.com>
To: Jon Mason <jdmason@kudzu.us>
Cc: Dave Jiang <dave.jiang@intel.com>,
        "Hubbe, Allen" <Allen.Hubbe@emc.com>,
        "S-k, Shyam-sundar" <Shyam-sundar.S-k@amd.com>,
        "Yu, Xiangliang" <Xiangliang.Yu@amd.com>,
        Gary R Hook <gary.hook@amd.com>, Sergey.Semin@t-platforms.ru,
        linux-ntb <linux-ntb@googlegroups.com>,
        linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 03/15] NTB: Fix UB/bug in ntb_mw_get_align()
Message-ID: <20171205175607.GD1701@mobilestation>
References: <20171203191736.3399-1-fancer.lancer@gmail.com>
 <20171203191736.3399-4-fancer.lancer@gmail.com>
 <CAPoiz9yuX+d1Bt1ioy_u9jLXzkWfHwKjoJ+hH5t4BN88YKrRsQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAPoiz9yuX+d1Bt1ioy_u9jLXzkWfHwKjoJ+hH5t4BN88YKrRsQ@mail.gmail.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1968
Lines: 49

On Tue, Dec 05, 2017 at 11:52:32AM -0500, Jon Mason <jdmason@kudzu.us> wrote:
> On Sun, Dec 3, 2017 at 2:17 PM, Serge Semin <fancer.lancer@gmail.com> wrote:
> > Simple (1 << pidx) operation causes undefined behaviour when
> > pidx >= 32. It must be casted to u64 to match the actual return
> > value of ntb_link_is_up() method, so to have all the possible
> > peer indexes covered and to get rid of undefined behaviour.
> > Additionally there are special macros in "linux/bitops.h" to perform
> > the bit-set-shift operations, so it's recommended to have them used
> > for proper bit setting.
> 
> This looks good to me, but also seems like a bug fix.  Please comment
> on if this is not noticed.
> 

The consequences of the bug isn't noticeable at the moment, but potentially
it can be seen on the devices like Switchtec (when multi-portness is finally
added). Anyway it's better to fix it now, than to wait for obvious bug.
Additionally as I said the originator of the code should have used BIT_ULL
instead of the pure bit shifting.

Thanks,
-Sergey

> Thanks,
> Jon
> 
> >
> > Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
> > Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
> > ---
> >  include/linux/ntb.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/include/linux/ntb.h b/include/linux/ntb.h
> > index c1646f2c6344..488e586fb76c 100644
> > --- a/include/linux/ntb.h
> > +++ b/include/linux/ntb.h
> > @@ -764,7 +764,7 @@ static inline int ntb_mw_get_align(struct ntb_dev *ntb, int pidx, int widx,
> >                                    resource_size_t *size_align,
> >                                    resource_size_t *size_max)
> >  {
> > -       if (!(ntb_link_is_up(ntb, NULL, NULL) & (1 << pidx)))
> > +       if (!(ntb_link_is_up(ntb, NULL, NULL) & BIT_ULL(pidx)))
> >                 return -ENOTCONN;
> >
> >         return ntb->ops->mw_get_align(ntb, pidx, widx, addr_align, size_align,
> > --
> > 2.12.0
> >