From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCHSET] kthread_worker: reimplement flush_kthread_work() to allow freeing during execution Date: Mon, 17 Sep 2012 12:40:16 -0700 Message-ID: <20120917194016.GI18677@google.com> References: <20120719211510.GA32763@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, Andrew Morton , Avi Kivity , kvm@vger.kernel.org, Andy Walls , ivtv-devel@ivtvdriver.org, linux-media@vger.kernel.org, Grant Likely , spi-devel-general@lists.sourceforge.net, Linus Torvalds , stable@vger.kernel.org To: Colin Cross Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-spi.vger.kernel.org On Fri, Sep 14, 2012 at 03:50:40PM -0700, Colin Cross wrote: > This patch set fixes a reproducible crash I'm seeing on a 3.4.10 > kernel. flush_kthread_worker (which is different from > flush_kthread_work) is initializing a kthread_work and a completion on > the stack, then queuing it and calling wait_for_completion. Once the > completion is signaled, flush_kthread_worker exits and the stack > region used by the kthread_work may be immediately reused by another > object on the stack, but kthread_worker_fn continues accessing its > work pointer: > work->func(work); <- calls complete, > effectively frees work > smp_wmb(); /* wmb worker-b0 paired with flush-b1 */ > work->done_seq = work->queue_seq; <- overwrites a > new stack object > smp_mb(); /* mb worker-b1 paired with flush-b0 */ > if (atomic_read(&work->flushing)) > wake_up_all(&work->done); <- or crashes here > > These patches fix the problem by not accessing work after work->func > is called, and should be backported to stable. They apply cleanly to > 3.4.10. Upstream commits are 9a2e03d8ed518a61154f18d83d6466628e519f94 > and 46f3d976213452350f9d10b0c2780c2681f7075b. Yeah, you're right. I wonder why this didn't come up before. Greg, can you please pick up these two commits? Thanks. -- tejun