Content-Type: multipart/mixed; boundary="===============4858131174269503144==" MIME-Version: 1.0 From: Paul E. McKenney To: lkp@lists.01.org Subject: Re: [percpu] BUG: unable to handle kernel NULL pointer dereference at (null) Date: Tue, 02 Sep 2014 09:21:10 -0700 Message-ID: <20140902162110.GB5001@linux.vnet.ibm.com> In-Reply-To: List-Id: --===============4858131174269503144== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Tue, Sep 02, 2014 at 11:00:12AM -0500, Christoph Lameter wrote: > On Tue, 2 Sep 2014, Paul E. McKenney wrote: > = > > Before this commit, raw_cpu_add_return() didn't build. The commit > > didn't affect anything else. > > > > So I don't understand how anything could work before this commit and > > be broken after it. Enlightenment? > = > Where is that raw_cpu_add_return statement? > = > grep does not match anything like that in kernel/rcu/* > and this_cpu_add_return is used extensively in the network subsystem for > statistics. Heh! Good point! Added by ac1bea85781e (sched,rcu: Make cond_resched() report RCU quiescent states), removed by 4a81e8328d379 (rcu: Reduce overhead of cond_resched() checks for RCU). So, as you say, no effect on contemporary kernels. Thanx, Paul --===============4858131174269503144==--