From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 315F8C4167D for ; Fri, 3 Nov 2023 12:21:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229744AbjKCMVP (ORCPT ); Fri, 3 Nov 2023 08:21:15 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48744 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233785AbjKCMVO (ORCPT ); Fri, 3 Nov 2023 08:21:14 -0400 Received: from ganesha.gnumonks.org (ganesha.gnumonks.org [IPv6:2001:780:45:1d:225:90ff:fe52:c662]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2A676184 for ; Fri, 3 Nov 2023 05:21:07 -0700 (PDT) Received: from [78.30.35.151] (port=42066 helo=gnumonks.org) by ganesha.gnumonks.org with esmtpsa (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qytAa-00EpIr-Tg; Fri, 03 Nov 2023 13:21:05 +0100 Date: Fri, 3 Nov 2023 13:21:00 +0100 From: Pablo Neira Ayuso To: Thomas Haller Cc: NetFilter , fw@strlen.de Subject: Re: [PATCH nft v2 4/7] build: no recursive make for "files/**/Makefile.am" Message-ID: References: <20231019130057.2719096-1-thaller@redhat.com> <20231019130057.2719096-5-thaller@redhat.com> <174b4dbc0df7fec4d0fdbe2c9cb96d4fca5ecd5b.camel@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <174b4dbc0df7fec4d0fdbe2c9cb96d4fca5ecd5b.camel@redhat.com> Precedence: bulk List-ID: X-Mailing-List: netfilter-devel@vger.kernel.org On Thu, Nov 02, 2023 at 05:53:43PM +0100, Thomas Haller wrote: > On Thu, 2023-11-02 at 17:05 +0100, Pablo Neira Ayuso wrote: [...] > > But this is completely inconsistent with what we have in other > > existing Netfilter trees. > > That would also be fixable, by adjusting those trees (I'd volunteer).  > > The question is what's better, and not what the projects copy-pasted > since 1995 do. I don't think it is worth the update, maybe some simplification to remove silly things such as Makefile.am with one singleton line, but there are better things to look at IMO. [...] > > Please do not couple tests with make process. > > On the branch, those tests work and it's convenient to run them and > reasonably fast! `make -j distcheck` takes 59 seconds on my machine. CI is what is missing, a single run is proving not giving much in return these days after your improvements. The recent bugs that were uncovered have been spotted by running this is a loop, and also exercising standalone 30s-stress from Florian for many hours. A few minutes does not harm (I can check how long it takes on my AMD Epyc box that I use for testing), but CI might provide more reliable information on what is going on. Thanks.