From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Bader Subject: Re: [xen-4.1-testing test] 9805: regressions - FAIL Date: Thu, 17 Nov 2011 10:57:48 +0100 Message-ID: <4EC4DA9C.60600@canonical.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Keir Fraser Cc: xen-devel@lists.xensource.com, Ian Jackson , Jan Beulich , Stefano Stabellini List-Id: xen-devel@lists.xenproject.org On 17.11.2011 10:06, Keir Fraser wrote: > On 17/11/2011 08:32, "Jan Beulich" wrote: > >> This is due to a bad backport of c/s 24007:0526644ad2a6: In -unstable, >> evtchn_unmask() must be called with d->event_lock held, while in 4.1 >> the function acquires the lock (and now gets called with the lock already >> held from do_physdev_op()'s case PHYSDEVOP_eoi). The change dates >> back to 23573:584c2e5e03d9, which hardly is a candidate for backporting >> (but maybe the locking change needs to be pulled out of there). > > Interestingly, Ubuntu's 4.1 fix has exactly the same problem. > Hm, yes we should. I am pretty sure I hit that code path often enough, Wonder why I never saw any dead lock there... --Stefan > I'll revert the patch until someone has time to actually implement and test > a working patch against our vanilla 4.1-testing tree. > > -- Keir > > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel