NetBSD Problem Report #40663

From www@NetBSD.org  Mon Feb 16 13:03:25 2009
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by narn.NetBSD.org (Postfix) with ESMTP id 17AC763BB35
	for <gnats-bugs@gnats.netbsd.org>; Mon, 16 Feb 2009 13:03:25 +0000 (UTC)
Message-Id: <20090216130324.D9E0D63B8C3@narn.NetBSD.org>
Date: Mon, 16 Feb 2009 13:03:24 +0000 (UTC)
From: harenberg@physik.uni-wuppertal.de
Reply-To: harenberg@physik.uni-wuppertal.de
To: gnats-bugs@NetBSD.org
Subject: math/py-numpy seems to miss the f2c requirement (at least when building under Linux)
X-Send-Pr-Version: www-1.0

>Number:         40663
>Category:       pkg
>Synopsis:       math/py-numpy seems to miss the f2c requirement (at least when building under Linux)
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    pkg-manager
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Feb 16 13:05:00 +0000 2009
>Closed-Date:    Tue Dec 13 19:19:00 +0000 2016
>Last-Modified:  Tue Dec 13 19:19:00 +0000 2016
>Originator:     Torsten Harenberg
>Release:        Linux 2.6.9
>Organization:
Bergische Universität Wuppertal
>Environment:
Linux grid-ui.physik.uni-wuppertal.de 2.6.9-67.0.4.EL_SFS2.3_0smp #1 SMP Mon Jul 21 13:57:38 CEST 2008 i686 i686 i386 GNU/Linux
>Description:
When trying to build py-numpy, it throws an exeption about missing f2c. When building lang/f2c, the package builds okay.

=> Returning to build of py25-numpy-1.1.0
===> Overriding tools for py25-numpy-1.1.0
===> Extracting for py25-numpy-1.1.0
===> Patching for py25-numpy-1.1.0
=> Applying pkgsrc patches for py25-numpy-1.1.0
===> Creating toolchain wrappers for py25-numpy-1.1.0
ERROR: f2c>=20001205nb3 is not installed; can't buildlink files.
*** Error code 1

Stop.
bmake: stopped in /usr/pkgsrc/math/py-numpy
*** Error code 1


>How-To-Repeat:
Linux:

cd PGKSRC_DIR/math/py-numpy
PKG_DIR/bin/bmake install
>Fix:
only a requirement missing?

>Release-Note:

>Audit-Trail:
From: "OBATA Akio" <obache@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: pkg/40663: math/py-numpy seems to miss the f2c requirement (at
 least when building under Linux)
Date: Tue, 17 Feb 2009 21:08:33 +0900

 On Mon, 16 Feb 2009 22:05:00 +0900, <harenberg@physik.uni-wuppertal.de> wrote:

 > => Applying pkgsrc patches for py25-numpy-1.1.0
 > ===> Creating toolchain wrappers for py25-numpy-1.1.0
 > ERROR: f2c>=20001205nb3 is not installed; can't buildlink files.
 > *** Error code 1

 It's known problem, same as PR#37536.

 Workaround:
 insert .include "../../mk/compiler.mk" to py-numpy/Makefile (like devel/kdevelop-base).

 Index: math/py-numpy/Makefile
 ===================================================================
 RCS file: /cvsroot/pkgsrc/math/py-numpy/Makefile,v
 retrieving revision 1.2
 diff -u -r1.2 Makefile
 --- math/py-numpy/Makefile	19 Dec 2008 22:09:55 -0000	1.2
 +++ math/py-numpy/Makefile	17 Feb 2009 11:50:56 -0000
 @@ -18,6 +18,8 @@
  PLIST_SUBST+=	PYVERSSUFFIX=${PYVERSSUFFIX}
  MAKE_ENV+=	ATLAS=None

 +.include "../../mk/compler.mk"
 +
  .include "../../lang/python/extension.mk"
  .include "../../math/blas/buildlink3.mk"
  .include "../../math/lapack/buildlink3.mk"


 Background:
 math/py-numpy/Makefile added "fortran" to "USE_LANGUAGES".
 then include
   lang/python/extension.mk
 => lang/python25/buildlink3.mk
 ==> mk/bdb.buildlink3.mk
 ===> databases/db4/buildlink3.mk
 ====> mk/compiler.mk
 =====> comipler/f2c.mk
 ======> lang/f2c/buildlink3.mk

 Then f2c/buildlink3.mk recognizes that it is included by the other buildlink3 file,
 not depends directly, no need to add itself to DEPENDS.

 Following patch should avoid the problem without any harmful effect.

 Index: mk/compiler/f2c.mk
 ===================================================================
 RCS file: /cvsroot/pkgsrc/mk/compiler/f2c.mk,v
 retrieving revision 1.11
 diff -u -r1.11 f2c.mk
 --- mk/compiler/f2c.mk	24 May 2008 07:27:11 -0000	1.11
 +++ mk/compiler/f2c.mk	17 Feb 2009 11:31:33 -0000
 @@ -99,6 +99,9 @@

  # Add the dependency on f2c.
  .  include "../../lang/f2c/buildlink3.mk"
 +.  if !defined(BUILDLINK_DEPENDS) || empty(BUILDLINK_DEPENDS:Mf2c)
 +BUILDLINK_DEPENDS+=	f2c
 +.  endif

  .  if defined(F2C_DIR) && !empty(F2C_DIR)
  PKGSRC_MAKE_ENV+=	F2C_DIR=${F2C_DIR:Q}

 -- 
 "Of course I love NetBSD":-)
 OBATA Akio / obache@NetBSD.org

State-Changed-From-To: open->feedback
State-Changed-By: obache@NetBSD.org
State-Changed-When: Sun, 13 Dec 2009 08:56:33 +0000
State-Changed-Why:
Should be fixed by mk/compiler/f2c.mk rev1.12.
Please confirm it on your environment.


State-Changed-From-To: feedback->open
State-Changed-By: obache@NetBSD.org
State-Changed-When: Sun, 13 Dec 2009 09:05:24 +0000
State-Changed-Why:
Not fixed, just changed from "no f2c" to "no libf2c".


State-Changed-From-To: open->feedback
State-Changed-By: cheusov@NetBSD.org
State-Changed-When: Sun, 08 May 2011 14:40:14 +0000
State-Changed-Why:
I cannot reproduce the problem on CentOS-5.5 and Debian-5.
Do you still experience the problem?


From: Torsten Harenberg <harenberg@physik.uni-wuppertal.de>
To: gnats-bugs@NetBSD.org
Cc: pkg-manager@netbsd.org, pkgsrc-bugs@netbsd.org,
	gnats-admin@netbsd.org, cheusov@NetBSD.org
Subject: Re: pkg/40663 (math/py-numpy seems to miss the f2c requirement (at least when building under Linux))
Date: Thu, 2 Jun 2011 13:39:25 +0200

 Am 08.05.2011 um 16:40 schrieb cheusov@NetBSD.org:

 > Synopsis: math/py-numpy seems to miss the f2c requirement (at least =
 when building under Linux)
 >=20
 > State-Changed-From-To: open->feedback
 > State-Changed-By: cheusov@NetBSD.org
 > State-Changed-When: Sun, 08 May 2011 14:40:14 +0000
 > State-Changed-Why:
 > I cannot reproduce the problem on CentOS-5.5 and Debian-5.
 > Do you still experience the problem?



 No.. seems not to exist anymore on RHEL 5 based OSes. I don't have any =
 single RHEL 4 based in production, so I cannot verify if this has also =
 gone there, but I would suggest to closed this PR, as it became quite =
 irrelevant.

 BR,

    Torsten


 --
 <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
 <>                                                              <>
 <> Dr. Torsten Harenberg     harenberg@physik.uni-wuppertal.de  <>
 <> Bergische Universitaet                                       <>
 <> FB C - Physik             Tel.: +49 (0)202 439-3521          <>
 <> Gaussstr. 20              Fax : +49 (0)202 439-2811          <>
 <> 42097 Wuppertal                                              <>
 <>                                                              <>
 <><><><><><><>< Of course it runs NetBSD http://www.netbsd.org ><>=20


State-Changed-From-To: feedback->open
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Sun, 12 Jun 2011 05:15:51 +0000
State-Changed-Why:
Feedback received.


From: David Holland <dholland-pbugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: OBATA Akio <obache@netbsd.org>
Subject: Re: pkg/40663: math/py-numpy seems to miss the f2c requirement (at
 least when building under Linux)
Date: Sun, 12 Jun 2011 05:20:44 +0000

 On Tue, Feb 17, 2009 at 12:10:05PM +0000, OBATA Akio wrote:
  >  > ===> Creating toolchain wrappers for py25-numpy-1.1.0
  >  > ERROR: f2c>=20001205nb3 is not installed; can't buildlink files.
  >  > *** Error code 1
  >  
  >  It's known problem, same as PR#37536.
 [...]

  >  Background:
  >  math/py-numpy/Makefile added "fortran" to "USE_LANGUAGES".
  >  then include
  >    lang/python/extension.mk
  >  => lang/python25/buildlink3.mk
  >  ==> mk/bdb.buildlink3.mk
  >  ===> databases/db4/buildlink3.mk
  >  ====> mk/compiler.mk
  >  =====> comipler/f2c.mk
  >  ======> lang/f2c/buildlink3.mk
  >  
  >  Then f2c/buildlink3.mk recognizes that it is included by the other
  >  buildlink3 file, not depends directly, no need to add itself to
  >  DEPENDS.

 So, at this point there's only one buildlink3.mk file that includes
 compiler.mk; it's databases/db46/buildlink3.mk, which includes it so
 it can test PKGSRC_COMPILER.

 Is there any reason the inclusion of compiler.mk can't be moved
 outside of the BUILDLINK_TREE block? That would solve the problem
 properly.

 ...although I don't see why this failure mechanism would be dependent
 on OS type or version. I guess it's probably because of db being
 builtin.

 -- 
 David A. Holland
 dholland@netbsd.org

State-Changed-From-To: open->closed
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Tue, 13 Dec 2016 19:19:00 +0000
State-Changed-Why:
The offending bl3.mk was removed along with db46 some time back.


>Unformatted:

Home	PR Database Search