From owner-freebsd-gnome@freebsd.org Mon Jan 18 18:39:39 2016 Return-Path: Delivered-To: freebsd-gnome@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 2F55AA8609F for ; Mon, 18 Jan 2016 18:39:39 +0000 (UTC) (envelope-from kubito@gmail.com) Received: from mailman.ysv.freebsd.org (mailman.ysv.freebsd.org [IPv6:2001:1900:2254:206a::50:5]) by mx1.freebsd.org (Postfix) with ESMTP id 0E1A41AD8 for ; Mon, 18 Jan 2016 18:39:39 +0000 (UTC) (envelope-from kubito@gmail.com) Received: by mailman.ysv.freebsd.org (Postfix) id 0B225A8609D; Mon, 18 Jan 2016 18:39:39 +0000 (UTC) Delivered-To: gnome@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 09A2CA8609C; Mon, 18 Jan 2016 18:39:39 +0000 (UTC) (envelope-from kubito@gmail.com) Received: from mail-wm0-x235.google.com (mail-wm0-x235.google.com [IPv6:2a00:1450:400c:c09::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 90B751AD5; Mon, 18 Jan 2016 18:39:38 +0000 (UTC) (envelope-from kubito@gmail.com) Received: by mail-wm0-x235.google.com with SMTP id n5so77601249wmn.0; Mon, 18 Jan 2016 10:39:38 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=sender:from:to:cc:subject:references:date:in-reply-to:message-id :user-agent:mime-version:content-type; bh=1A5VF3VrZtCQeO26plPQBY+7UCAZ6PtpKTO1USseaqw=; b=LbvfNVtVVDN07iRhsIy5+DGb+2hfwxrQuzj/s45tBCVDxwaMnoGPFHeguQLDBFWie6 Rko6+8E7szANrmD7eZQfHeDaqaQKSC74TLow/1uMAbi9MlpW+3XFy8TZwgcgUJ/9OgXC vCyB3JEfXEKWMYnDfYsBvaAjtn7Hmsb01gPbuzjathiiYb38H37tTuFbVPJ/ddpKQp26 JP1ZSVvfSqF/7pqbl5q+tSApg/rKWhDVpojb7ac3UyNz/goqwH5fu5VBDIB9FuCh2M2m t9tFxVuYlsKSKB/Sf6LivGBEignFAGiNPLB9whrA78HZUrND0/+Y9seSC1+nNLN/VOkp sXRA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:sender:from:to:cc:subject:references:date :in-reply-to:message-id:user-agent:mime-version:content-type; bh=1A5VF3VrZtCQeO26plPQBY+7UCAZ6PtpKTO1USseaqw=; b=SROUJpMVI2+Lunnc7GpdiCjrUitsjIe7h/PB9KFq6GV1c8nWxp71FXO9WKxdtIIQ2X mND2Y13DFcUa6P6RheAwwqDXib8hHh1oUvvv6xlUkp070omUR1xlkI8YWhWtfeo7Kmym sAovMStKsB0UIGxTkekteIv3fOUREpf8inVBRbjOiENv8qSwrgVb1xHEZq9k0bpzuYeX S4vD1vI/njZYFXxor/Zi9W3exljqBtvJhKpSAyWbhm6cG9fQXeIW00qPN9ZpzCoHqSju Sha+uZsmRST1klSExS/EGuEQbznJm1NVySFYOWxZwN4OvgZ5DFJLn/Tz2Kll0lYZ3j1c Lvbg== X-Gm-Message-State: ALoCoQnHZDQXTl/vbgipnaAl+cPYjYhx0jwdTId55X9xsYMwE2xI69i1fgoRV/683fgK2hWodaz4D5IBZt/nWCr7DEB/2qFJtw== X-Received: by 10.194.235.137 with SMTP id um9mr29700676wjc.146.1453142377149; Mon, 18 Jan 2016 10:39:37 -0800 (PST) Received: from rodolpho.gmail.com ([46.101.211.41]) by smtp.gmail.com with ESMTPSA id aq3sm25067985wjc.1.2016.01.18.10.39.36 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 18 Jan 2016 10:39:36 -0800 (PST) Sender: Raphael Kubo da Costa From: Raphael Kubo da Costa To: Mathieu Arnold Cc: gnome@FreeBSD.org, perl@FreeBSD.org Subject: Re: libperl stage-qa check and libraries that do not link against libperl.so References: <86d1t0nu6x.fsf@FreeBSD.org> <83298B9CC954C7F5642168D0@atuin.in.mat.cc> <868u3nonxn.fsf@FreeBSD.org> <880733D886684E0C0D43E7D8@atuin.in.mat.cc> <8637tvyzpe.fsf@FreeBSD.org> <1927902E41BBF231A0EF6A54@ogg.in.absolight.net> Date: Mon, 18 Jan 2016 16:39:29 -0200 In-Reply-To: (Mathieu Arnold's message of "Mon, 18 Jan 2016 16:34:56 +0100") Message-ID: <86twmayboe.fsf@FreeBSD.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-BeenThere: freebsd-gnome@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: GNOME for FreeBSD -- porting and maintaining List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Jan 2016 18:39:39 -0000 Mathieu Arnold writes: > tl;dr: > > The qa script work as intented, it caught some software doing something > stupid. OK, thanks for checking this. I've sent https://github.com/libproxy/libproxy/pull/7 to get this changed upstream. > +--On 18 janvier 2016 16:18:40 +0100 Mathieu Arnold wrote: > | > | > | +--On 18 janvier 2016 08:00:29 -0200 Raphael Kubo da Costa > | wrote: > || Mathieu Arnold writes: > || > ||> +--On 17 janvier 2016 23:12:36 +0100 Raphael Kubo da Costa > ||> wrote: > ||> |> For these, if the non-linking is intentional, and not needed, > ||> |> ignore the ***WARNING***. > ||> | > ||> | Right. If linking against libperl.so and not linking against it are > ||> | both fine, does it still make sense to have that stage-qa check? > ||> > ||> There is a stage-qa check because if you link against libperl.so, you > ||> MUST have rpath set to the path of libperl.so, it happens if the port is > ||> doing what Perl should be doing and getting it wrong. Also, sometime, > ||> you don't link with libperl.so, and it's a mistake, and it should be. > ||> > ||> There are a few cases where it's not needed, but when I added the qa > ||> check when I changed libperl.so to be versionned, out of all the ports > ||> that install .so in PERL_ARCH, there were, say, 3 or 4 that had a .so > ||> that did not need to be linked with libperl.so. > || > || How about making the "not linked with libperl.so.x.y" warning non-fatal > || then? Right now I cannot just ignore it as suggested because it makes > || the script exit with a non-zero error code. > | > | I just had a look at the commit, and the qa script is right. The author > | is wrong, the .so contains Perl code and must be linked with libperl.so. > | It must be linked with it so that when the major Perl version is changed > | (like 5.20 -> 5.22 that is happening soon(tm)) the link allows pkg to > | know that the package needs to be reinstalled so that it continues to > | work. > | > || While here: there's a FIXME comment in this part that talks about > || changing the linkage detection logic, but it does not specify what > || should be changed. Do you have more information? I could try to fix both > || issues and send a patch for review. > | > | I forgot about the FIXME. It works as intented right now, I'll have a look > | at it later.