From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49895) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bF1S2-0003M7-VX for qemu-devel@nongnu.org; Mon, 20 Jun 2016 11:49:28 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bF1Ry-00042E-UO for qemu-devel@nongnu.org; Mon, 20 Jun 2016 11:49:26 -0400 Received: from mail-lf0-x242.google.com ([2a00:1450:4010:c07::242]:33761) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bF1Ry-00042A-Mc for qemu-devel@nongnu.org; Mon, 20 Jun 2016 11:49:22 -0400 Received: by mail-lf0-x242.google.com with SMTP id l188so5874365lfe.0 for ; Mon, 20 Jun 2016 08:49:22 -0700 (PDT) References: <1466181227-14934-1-git-send-email-alex.bennee@linaro.org> From: Sergey Fedorov Message-ID: <57681080.20200@gmail.com> Date: Mon, 20 Jun 2016 18:49:20 +0300 MIME-Version: 1.0 In-Reply-To: <1466181227-14934-1-git-send-email-alex.bennee@linaro.org> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [RFC 0/7] Safe watch and breakpoint manipulation List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?UTF-8?Q?Alex_Benn=c3=a9e?= , mttcg@listserver.greensocs.com, qemu-devel@nongnu.org, fred.konrad@greensocs.com, a.rigo@virtualopensystems.com, cota@braap.org, bobby.prani@gmail.com Cc: mark.burton@greensocs.com, pbonzini@redhat.com, jan.kiszka@siemens.com, rth@twiddle.net, peter.maydell@linaro.org, claudio.fontana@huawei.com On 17/06/16 19:33, Alex Bennée wrote: > Last time I went through the MTTCG code the access to the > break/watchpoint code was annotated with "RCU?". The code currently > gets away with avoiding locks for the gdbstub as the guest execution > state is usually halted. However when used for modelling architectural > debug registers there is no such protection. I'm not so sure if there's any architecture which permits changing breakpoins/watchpoints of one core from another. > The patch series changes things in stages. > > First we move the break/watchpoints into an array which is more > amenable to RCU control that the QLIST. We then control the life time > of references to break/watchpoint data by removing long held > references in the target code and getting information when needed from > the core. Then we stop dynamically allocation the watch/breakpoint > data and store it directly in the array which makes iteration across > the list a bit more cache friendly than referenced pointers. Finally > addition and removal of elements of the array is put under RCU > control. This ensures there is always a safe array of data to check > in the run-loop. I a little bit unsure if we really want to complicate things with RCU. Why don't we simply protect the lists with a mutex given that there's no contention expected? BTW, as it comes to debugging, I suppose we don't expect great performance anyway. Kind regards, Sergey