From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net X-Spam-Level: X-Spam-ASN: AS53758 23.128.96.0/24 X-Spam-Status: No, score=-3.9 required=3.0 tests=AWL,BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,RCVD_IN_DNSWL_MED, SPF_HELO_PASS,SPF_PASS shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by dcvr.yhbt.net (Postfix) with ESMTP id 18FC71F8C8 for ; Thu, 23 Sep 2021 17:41:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S242601AbhIWRmt (ORCPT ); Thu, 23 Sep 2021 13:42:49 -0400 Received: from cloud.peff.net ([104.130.231.41]:53692 "EHLO cloud.peff.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S242557AbhIWRms (ORCPT ); Thu, 23 Sep 2021 13:42:48 -0400 Received: (qmail 15506 invoked by uid 109); 23 Sep 2021 17:41:15 -0000 Received: from Unknown (HELO peff.net) (10.0.1.2) by cloud.peff.net (qpsmtpd/0.94) with ESMTP; Thu, 23 Sep 2021 17:41:15 +0000 Authentication-Results: cloud.peff.net; auth=none Received: (qmail 29833 invoked by uid 111); 23 Sep 2021 17:41:15 -0000 Received: from coredump.intra.peff.net (HELO sigill.intra.peff.net) (10.0.0.2) by peff.net (qpsmtpd/0.94) with (TLS_AES_256_GCM_SHA384 encrypted) ESMTPS; Thu, 23 Sep 2021 13:41:15 -0400 Authentication-Results: peff.net; auth=none Date: Thu, 23 Sep 2021 13:41:14 -0400 From: Jeff King To: =?utf-8?B?w4Z2YXIgQXJuZmrDtnLDsA==?= Bjarmason Cc: Junio C Hamano , git@vger.kernel.org, Emily Shaffer , SZEDER =?utf-8?B?R8OhYm9y?= , Taylor Blau , Jonathan Nieder Subject: Re: What's cooking in git.git (Sep 2021, #06; Mon, 20) Message-ID: References: <875yut8nns.fsf@evledraar.gmail.com> <87mto343ze.fsf@evledraar.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <87mto343ze.fsf@evledraar.gmail.com> Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org On Thu, Sep 23, 2021 at 06:51:12PM +0200, Ævar Arnfjörð Bjarmason wrote: > > Isn't the direction of dependency the other way? Once fixes go in, > > checks will no longer complain, but until fixes are reviewed and > > applied, checking at CI will break the testing and this would need > > working around by marking various tests as "not ready to be tested". > > The fixes are structured as fixing the code, and then for both > self-documentation & testing turning on: > > TEST_PASSES_SANITIZE_LEAK=true > > In the same commit, I could fix the leaks and have a second series later > for turning on the regression test, or just turning on the test and > having it kick in once it's merged with ab/sanitize-leak-ci, but waiting > until ab/sanitize-leak-ci hit master first seemed less confusing for > everyone. > > But if you'd like to have 'em now with either of those caveats... Yeah, I think it's much easier to demonstrate the leak and that it is fixed when we have the infrastructure from that series. So the leaks themselves can obviously be fixed without depending on that series, but it's much easier to review the patches when we have that extra tooling. -Peff