NetBSD Problem Report #43263

From www@NetBSD.org  Wed May  5 20:48:44 2010
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by www.NetBSD.org (Postfix) with ESMTP id 43B2263BA59
	for <gnats-bugs@gnats.NetBSD.org>; Wed,  5 May 2010 20:48:44 +0000 (UTC)
Message-Id: <20100505204844.1291063B8FE@www.NetBSD.org>
Date: Wed,  5 May 2010 20:48:44 +0000 (UTC)
From: frankalahan@gmail.com
Reply-To: frankalahan@gmail.com
To: gnats-bugs@NetBSD.org
Subject: Dragonfly and Darwin default configuration can be improved
X-Send-Pr-Version: www-1.0

>Number:         43263
>Category:       pkg
>Synopsis:       Dragonfly and Darwin default configuration can be improved
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    macos-pkg-people
>State:          open
>Class:          doc-bug
>Submitter-Id:   net
>Arrival-Date:   Wed May 05 20:50:00 +0000 2010
>Last-Modified:  Wed Mar 07 09:38:39 +0000 2012
>Originator:     Francisco García Rodríguez
>Release:        pkgsrc for Darwin
>Organization:
>Environment:
Darwin prometeo2001.fritz.box 10.3.0 Darwin Kernel Version 10.3.0: Fri Feb 26 11:58:09 PST 2010; root:xnu-1504.3.12~1/RELEASE_I386 i386

>Description:
I have noticed that PKGSRC default installation does not enforce the use of the native GCC. This will lead to many packages depending on building lang/gcc which is quite annoying for Darwin users because that package does not compile with Mac. Looking at the sources I believe that DragonFly users will have the same problem.


>How-To-Repeat:
Try to compile gnupg2 with Darwin
>Fix:
In my case I solved my problems forcing the native GCC in my configuration and cleaning the pkgsrc tree (bmake clean). The change does not take effect otherwise:

USE_NATIVE_GCC=     yes

Maybe that should be mentioned on pkgsrc documentation:
http://www.netbsd.org/docs/pkgsrc/platforms.html

I have filed this bug as doc-bug, but maybe it could be solved with bootstrap. However I believe the ideal solution might be achieved with the documentation because I have the feeling that I should compile GCC and later remove that option (I have the feeling that will lead to fewer future problems)

>Release-Note:

>Audit-Trail:
From: Joerg Sonnenberger <joerg@britannica.bec.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: pkg/43263: Dragonfly and Darwin default configuration can be
 improved
Date: Thu, 6 May 2010 01:36:53 +0200

 On Wed, May 05, 2010 at 08:50:00PM +0000, frankalahan@gmail.com wrote:
 > >Description:
 > I have noticed that PKGSRC default installation does not enforce the
 > use of the native GCC.

 Yes and no. mk/compiler/gcc.mk tries to figure out what version of GCC
 is present, but that fails in various situations, e.g. if CC is
 specified manually to the wrong command etc. As usual, the correct trick
 is to figure out why it failed. USE_NATIVE_GCC is just a crude
 hackaround.

 Joerg

Responsible-Changed-From-To: pkg-manager->darwin-pkg-people
Responsible-Changed-By: reed@NetBSD.org
Responsible-Changed-When: Sat, 08 May 2010 12:01:32 +0000
Responsible-Changed-Why:
Assigned to darwin-pkg-people.
(I don't know of problem on DragonFly, but can be assigend to
dragonfly-pkg-people later if needed.)


From: Francisco Garcia <frankalahan@gmail.com>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: pkg/43263: Dragonfly and Darwin default configuration can be improved
Date: Thu, 13 May 2010 08:29:05 +0200

 --Apple-Mail-1--927798115
 Content-Transfer-Encoding: quoted-printable
 Content-Type: text/plain;
 	charset=us-ascii

 That is exactly the reason I had problems. I've been looking at  =
 mk/compiler/gcc.mk

 > # _CC is the full path to the compiler named by ${CC} if it can be =
 found.
 > .if !defined(_CC)
 > _CC:=3D   ${CC:C/^/_asdf_/1:M_asdf_*:S/^_asdf_//}

 I think it would be useful to make some checks. At least checking that =
 either $CC or the output of $CC --version has the "gcc" string

 I tried to add the check, but now I don't know how to report the error =
 or stop the process.


 Am 06.05.2010 um 01:40 schrieb Joerg Sonnenberger:

 > The following reply was made to PR pkg/43263; it has been noted by =
 GNATS.
 >=20
 > From: Joerg Sonnenberger <joerg@britannica.bec.de>
 > To: gnats-bugs@NetBSD.org
 > Cc:=20
 > Subject: Re: pkg/43263: Dragonfly and Darwin default configuration can =
 be
 > improved
 > Date: Thu, 6 May 2010 01:36:53 +0200
 >=20
 > On Wed, May 05, 2010 at 08:50:00PM +0000, frankalahan@gmail.com wrote:
 >>> Description:
 >> I have noticed that PKGSRC default installation does not enforce the
 >> use of the native GCC.
 >=20
 > Yes and no. mk/compiler/gcc.mk tries to figure out what version of GCC
 > is present, but that fails in various situations, e.g. if CC is
 > specified manually to the wrong command etc. As usual, the correct =
 trick
 > is to figure out why it failed. USE_NATIVE_GCC is just a crude
 > hackaround.
 >=20
 > Joerg
 >=20


 --Apple-Mail-1--927798115
 Content-Transfer-Encoding: quoted-printable
 Content-Type: text/html;
 	charset=us-ascii

 <html><head></head><body style=3D"word-wrap: break-word; =
 -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">That =
 is exactly the reason I had problems. I've been looking at =
 &nbsp;mk/compiler/gcc.mk<div><br></div><div><blockquote =
 type=3D"cite"><div><font class=3D"Apple-style-span" color=3D"#000000"># =
 _CC is the full path to the compiler named by ${CC} if it can be =
 found.</font></div><div><font class=3D"Apple-style-span" =
 color=3D"#000000">.if !defined(_CC)</font></div></blockquote><blockquote =
 type=3D"cite"><div><font class=3D"Apple-style-span" =
 color=3D"#000000">_CC:=3D &nbsp; =
 ${CC:C/^/_asdf_/1:M_asdf_*:S/^_asdf_//}</font></div></blockquote><div><br>=
 </div><div>I think it would be useful to make some checks. At least =
 checking that either $CC or the output of $CC --version has the "gcc" =
 string</div><div><br></div><div>I tried to add the check, but now I =
 don't know how to report the error or stop the =
 process.</div><div><br></div><div><div><br></div><div>Am 06.05.2010 um =
 01:40 schrieb Joerg Sonnenberger:</div><br =
 class=3D"Apple-interchange-newline"><blockquote type=3D"cite"><div>The =
 following reply was made to PR pkg/43263; it has been noted by =
 GNATS.<br><br>From: Joerg Sonnenberger &lt;<a =
 href=3D"mailto:joerg@britannica.bec.de">joerg@britannica.bec.de</a>&gt;<br=
 >To: <a =
 href=3D"mailto:gnats-bugs@NetBSD.org">gnats-bugs@NetBSD.org</a><br>Cc: =
 <br>Subject: Re: pkg/43263: Dragonfly and Darwin default configuration =
 can be<br> improved<br>Date: Thu, 6 May 2010 01:36:53 +0200<br><br> On =
 Wed, May 05, 2010 at 08:50:00PM +0000, <a =
 href=3D"mailto:frankalahan@gmail.com">frankalahan@gmail.com</a> =
 wrote:<br><blockquote type=3D"cite"><blockquote =
 type=3D"cite">Description:<br></blockquote></blockquote><blockquote =
 type=3D"cite">I have noticed that PKGSRC default installation does not =
 enforce the<br></blockquote><blockquote type=3D"cite">use of the native =
 GCC.<br></blockquote><br> Yes and no. mk/compiler/gcc.mk tries to figure =
 out what version of GCC<br> is present, but that fails in various =
 situations, e.g. if CC is<br> specified manually to the wrong command =
 etc. As usual, the correct trick<br> is to figure out why it failed. =
 USE_NATIVE_GCC is just a crude<br> hackaround.<br><br> =
 Joerg<br><br></div></blockquote></div><br></div></body></html>=

 --Apple-Mail-1--927798115--

Responsible-Changed-From-To: darwin-pkg-people->macos-pkg-people
Responsible-Changed-By: dholland@NetBSD.org
Responsible-Changed-When: Wed, 07 Mar 2012 09:38:39 +0000
Responsible-Changed-Why:
darwin-pkg-people has been changed to macos-pkg-people


>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.