Content-Type: multipart/mixed; boundary="===============8394864254569129589==" MIME-Version: 1.0 From: Dan Carpenter Subject: Re: [linux-stable-rc:linux-3.16.y 6719/6738] net/sched/sch_netem.c:581 netem_enqueue() error: we previously assumed 'skb' could be null (see line 505) Date: Thu, 10 Oct 2019 14:00:16 +0300 Message-ID: <20191010110016.GM13286@kadam> In-Reply-To: List-Id: To: kbuild@lists.01.org --===============8394864254569129589== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Wed, Oct 09, 2019 at 04:05:44PM +0100, Ben Hutchings wrote: > On Wed, 2019-10-09 at 11:12 +0300, Dan Carpenter wrote: > > tree: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-st= able-rc.git linux-3.16.y > > head: c508a924ddd1119a8037c1ec23f560eade0444c6 > > commit: 61763432af929bd3d091a18469c130719f00c6de [6719/6738] net: netem= : fix backlog accounting for corrupted GSO frames > > = > > If you fix the issue, kindly add following tag > > Reported-by: kbuild test robot > > Reported-by: Dan Carpenter > > = > > smatch warnings: > > net/sched/sch_netem.c:581 netem_enqueue() error: we previously assumed = 'skb' could be null (see line 505) > > net/sched/sch_netem.c:581 netem_enqueue() error: dereferencing freed me= mory 'skb' > = > It seems like the same (or similar) bugs are introduced by the upstream > version. Huh. Smatch ignores those because unnecessary checks for NULL are common inside macros, and unlikely() is a macro so it doesn't print an inconsistent checking warning. I'll write a special case for that and see what turns up. regards, dan carpenter --===============8394864254569129589==--