From mboxrd@z Thu Jan 1 00:00:00 1970 From: Piotr =?utf-8?B?RGHFgmVr?= Subject: Re: Static analysis Date: Wed, 23 Nov 2016 08:25:24 +0100 Message-ID: <20161123072524.GE26997@predictor> References: <20161123070637.GF26846@rskikr.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: Received: from predictor.org.pl ([185.5.97.54]:39168 "EHLO predictor.org.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751644AbcKWHcG (ORCPT ); Wed, 23 Nov 2016 02:32:06 -0500 Content-Disposition: inline In-Reply-To: <20161123070637.GF26846@rskikr.localdomain> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Brad Hubbard Cc: ceph-devel@vger.kernel.org On Wed, Nov 23, 2016 at 05:06:37PM +1000, Brad Hubbard wrote: > I did a run of the Clang Static Analyzer and it turned up multiple issues which > Kefu gobbled up :) The reports are quite detailed and walk you through how the > issue can present. > > See https://github.com/ceph/ceph/pull/12145 and > https://github.com/ceph/ceph/pull/12148 > > Given how easy it is to run [1] (although slow) would it be a good idea to be > running this regulary, perhaps in addition to coverity scans? > > [1] mkdir build && cd build && scan-build cmake .. && scan-build make -jX +1. Certainly this will help finding out obscure bugs that show up later in unexpected places. -- Piotr Dałek branch@predictor.org.pl http://blog.predictor.org.pl