Debian Bug report logs - #381341
replace amd64 hack with something better

version graph

Package: wine; Maintainer for wine is Debian Wine Party <pkg-wine-party@lists.alioth.debian.org>; Source for wine is src:wine (PTS, buildd, popcon).

Reported by: Robert Millan <rmh@aybabtu.com>

Date: Thu, 3 Aug 2006 19:03:18 UTC

Severity: important

Merged with 462542

Found in versions wine/0.9.34-1, wine/0.9.53-1, wine/0.9.57-1

Fixed in version wine/0.9.61-1

Done: Ove Kaaven <ovek@arcticnet.no>

Bug is archived. No further changes may be made.

Toggle useless messages

View this report as an mbox folder, status mbox, maintainer mbox

Report forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Robert Millan <rmh@aybabtu.com>:
New Bug report received and forwarded. Copy sent to Ove Kaaven <ovek@arcticnet.no>. (full text, mbox, link).

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Debian Bug Tracking System <submit@bugs.debian.org>
Subject: (partial) amd64 biarch support
Date: Thu, 03 Aug 2006 20:57:57 +0200
[Message part 1 (text/plain, inline)]
Package: wine
Severity: normal
Tags: patch

I've been looking at some of the things needed to get wine properly working
on amd64 using biarch.

 - A pair of binutils patches:

   #379596: Problems linking 32bit libraries on AMD64

   #381306: ld exits succesfuly when attempting horrid linking of mixed 32/64 bit objects

 - A patch for wine debian/control (attached)

 - Sorting out this blurb in debian/rules.  It hardcodes library paths that
   are not necessarily of the right width.  Why aren't the upstream autoconf
   checks good enough?

        ICUUC_LIB="` \
        (dpkg -s libicu34-dev >/dev/null && \
         echo "/usr/lib/libsicuuc.a") || \
        (dpkg -s libicu28-dev >/dev/null && \
         echo "/usr/lib/libsicuuc.a") || \
        (dpkg -s libicu21-dev >/dev/null && \
         echo "/usr/lib/libicuuc.a") || \
        echo ""`" \
        ICUDATA_LIB="` \
        (dpkg -s libicu34-dev >/dev/null && \
         echo "/usr/lib/libsicudata.a") || \
        (dpkg -s libicu28-dev >/dev/null && \
         echo "/usr/lib/libsicudata.a") || \
        (dpkg -s libicu21-dev >/dev/null && \
         echo "/usr/lib/libicudata.a") || \
        echo ""`" \

  - Adding 32bit versions of a few X11 libraries... oh well.

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.17-1-amd64-k8
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8)

[biarch.diff (text/plain, attachment)]

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Andreas Jochens <aj@andaco.de>:
Extra info received and forwarded to list. Copy sent to Ove Kaaven <ovek@arcticnet.no>. (full text, mbox, link).

Message #10 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Andreas Jochens <aj@andaco.de>
To: 381341@bugs.debian.org, control@bugs.debian.org
Subject: wine: FTBFS (amd64): Please add support to build wine on amd64
Date: Fri, 20 Oct 2006 12:08:44 +0200
retitle 381341 wine: FTBFS (amd64): Please add support to build wine on amd64
thanks

Hello,

the attached patch will make 'wine' build on amd64 provided that some
extra lib*.so symlinks are being added to the ia32-libs package (a
separate report with a patch has been filed to the BTS for ia32-libs).

The attached patch does the following:

- change the architecture lines in debian/control to include amd64

- change debian/rules to add ia32-libs to the Depends of libwine on amd64

- change debian/rules so that it does not try to move glut32* files on amd64

- add '-fatal-warnings' to the ICU library check in configure{,.ac}
  to make that test fail on amd64 because there is no 32-bit version 
  of the library available on amd64

Regards
Andreas Jochens

diff -urN ../tmp-orig/wine-0.9.20/configure ./configure
--- ../tmp-orig/wine-0.9.20/configure	2006-10-20 08:00:51.000000000 +0000
+++ ./configure	2006-10-20 07:54:08.000000000 +0000
@@ -11810,7 +11810,7 @@
         TEST_ICUDATA_LIB="${ICUDATA_LIB-${i}data.a}"
         { echo "$as_me:$LINENO: checking whether can link with ICU libraries $TEST_ICUUC_LIB and $TEST_ICUDATA_LIB" >&5
 echo $ECHO_N "checking whether can link with ICU libraries $TEST_ICUUC_LIB and $TEST_ICUDATA_LIB... $ECHO_C" >&6; }
-        LIBS="$saved_libs $TEST_ICUUC_LIB $TEST_ICUDATA_LIB -lstdc++ -lgcc_s"
+        LIBS="-fatal-warnings $saved_libs $TEST_ICUUC_LIB $TEST_ICUDATA_LIB -lstdc++ -lgcc_s"
     	cat >conftest.$ac_ext <<_ACEOF
 /* confdefs.h.  */
 _ACEOF
diff -urN ../tmp-orig/wine-0.9.20/configure.ac ./configure.ac
--- ../tmp-orig/wine-0.9.20/configure.ac	2006-10-20 08:00:51.000000000 +0000
+++ ./configure.ac	2006-10-20 07:54:08.000000000 +0000
@@ -608,7 +608,7 @@
         TEST_ICUUC_LIB="${ICUUC_LIB-${i}uc.a}"
         TEST_ICUDATA_LIB="${ICUDATA_LIB-${i}data.a}"
         AC_MSG_CHECKING(whether can link with ICU libraries $TEST_ICUUC_LIB and $TEST_ICUDATA_LIB)
-        LIBS="$saved_libs $TEST_ICUUC_LIB $TEST_ICUDATA_LIB -lstdc++ -lgcc_s"
+        LIBS="-fatal-warnings $saved_libs $TEST_ICUUC_LIB $TEST_ICUDATA_LIB -lstdc++ -lgcc_s"
     	AC_TRY_LINK([#include <unicode/ubidi.h>],[ubidi_open()],
                     [AC_DEFINE(HAVE_ICU,1,[Define to 1 if the ICU libraries are installed])
             	     AC_SUBST(ICULIBS,"$TEST_ICUUC_LIB $TEST_ICUDATA_LIB -lstdc++ -lgcc_s")
diff -urN ../tmp-orig/wine-0.9.20/debian/control ./debian/control
--- ../tmp-orig/wine-0.9.20/debian/control	2006-10-20 08:00:51.000000000 +0000
+++ ./debian/control	2006-10-20 07:59:56.000000000 +0000
@@ -19,11 +19,12 @@
  libicu36-dev | libicu34-dev (>= 3.4-4) | libicu28-dev | libicu21-dev,
  libfontconfig1-dev, libssl-dev, libcapi20-dev (>= 1:3.3.0.20041024-2) [!kfreebsd-i386 !kfreebsd-amd64 !hurd-i386],
  libhal-dev, libdbus-1-dev, libgphoto2-2-dev, liblcms1-dev, libldap2-dev,
- libxml2-dev, libxslt1-dev, fontforge, prelink
+ libxml2-dev, libxslt1-dev, fontforge, prelink,
+ libc6-dev-i386 [amd64], ia32-libs [amd64], lib32z1-dev [amd64], lib32asound2-dev [amd64], lib32ncurses5-dev [amd64]
 Standards-Version: 3.6.0
 
 Package: wine
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: ${debconf-depends}, libwine (= ${Source-Version}), xbase-clients (>= 4.0) | xcontrib
 Recommends: wine-utils, msttcorefonts
 Suggests: wine-doc, binfmt-support
@@ -39,7 +40,7 @@
  Wine is often updated.
 
 Package: libwine-dev
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Section: libdevel
 Depends: libwine (= ${Source-Version}), libc6-dev
 Replaces: libwine (<< 0.0.20010216)
@@ -55,8 +56,8 @@
 
 Package: libwine
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
-Depends: ${debconf-depends}, ${shlibs:Depends}${freetype}
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
+Depends: ${debconf-depends}, ${shlibs:Depends}${freetype}${ia32-libs}
 Replaces: libwine0.0.971116, wine (<< 0.9.11-1), libwine-twain (<< 0.9.13)
 Conflicts: libwine0.0.971116
 Recommends: libwine-print, libwine-gl${icu}
@@ -71,7 +72,7 @@
 
 Package: libwine-alsa
 Section: libs
-Architecture: i386 powerpc sparc
+Architecture: amd64 i386 powerpc sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (ALSA Sound Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -83,7 +84,7 @@
 
 Package: libwine-arts
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Replaces: libwine (<< 0.0.20020710)
 Description: Windows API Implementation (aRts Sound Module)
@@ -96,7 +97,7 @@
 
 Package: libwine-capi
 Section: libs
-Architecture: i386 powerpc sparc
+Architecture: amd64 i386 powerpc sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (ISDN Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -109,7 +110,7 @@
 
 Package: libwine-cms
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Replaces: libwine (<< 0.0.20050830)
 Description: Windows API Implementation (Color Management Module)
@@ -123,7 +124,7 @@
 
 Package: libwine-esd
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (EsounD Sound Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -135,7 +136,7 @@
 
 Package: libwine-gl
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (OpenGL Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -148,7 +149,7 @@
 
 Package: libwine-gphoto2
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (Camera Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -161,7 +162,7 @@
 
 Package: libwine-jack
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}${jack}
 Description: Windows API Implementation (JACK Sound Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -173,7 +174,7 @@
 
 Package: libwine-ldap
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (LDAP Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -186,7 +187,7 @@
 
 Package: libwine-nas
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Replaces: libwine (<< 0.0.20020710)
 Description: Windows API Implementation (NAS Sound Module)
@@ -199,7 +200,7 @@
 
 Package: libwine-print
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: amd64 i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}${freetype}${cupsys}
 Replaces: libwine (<< 0.0.20020710)
 Description: Windows API Implementation (Printing Module)
@@ -213,7 +214,7 @@
 
 Package: libwine-sane
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), ${shlibs:Depends}
 Description: Windows API Implementation (Scanner Module)
  This is a BETA release of Wine, the free MS-Windows API implementation.
@@ -226,7 +227,7 @@
 
 Package: libwine-twain
 Section: libs
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine-sane
 Suggests: libwine-gphoto2
 Description: Windows API Implementation (empty transitional package)
@@ -234,7 +235,7 @@
  and libwine-gphoto2 packages. It can safely be removed.
 
 Package: wine-utils
-Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
+Architecture: any-amd64 any-i386 any-powerpc any-sparc
 Depends: libwine (= ${Source-Version}), wine
 Replaces: libwine-dev (<< 0.0.20010216)
 Description: Windows API Implementation (Utilities)
diff -urN ../tmp-orig/wine-0.9.20/debian/rules ./debian/rules
--- ../tmp-orig/wine-0.9.20/debian/rules	2006-10-20 08:00:51.000000000 +0000
+++ ./debian/rules	2006-10-20 07:59:37.000000000 +0000
@@ -182,7 +182,9 @@
 	mv debian/libwine/usr/lib/wine/d3d9*     debian/libwine-gl/usr/lib/wine
 	mv debian/libwine/usr/lib/wine/d3dx8*    debian/libwine-gl/usr/lib/wine
 	mv debian/libwine/usr/lib/wine/glu32*    debian/libwine-gl/usr/lib/wine
+ifneq ($(DEB_HOST_ARCH_CPU), amd64)
 	mv debian/libwine/usr/lib/wine/glut32*   debian/libwine-gl/usr/lib/wine
+endif
 	mv debian/libwine/usr/lib/wine/opengl32* debian/libwine-gl/usr/lib/wine
 	mv debian/libwine/usr/lib/wine/wined3d*  debian/libwine-gl/usr/lib/wine
 	mv debian/libwine/usr/lib/wine/winejack* debian/libwine-jack/usr/lib/wine
@@ -247,6 +249,11 @@
 	dh_installdeb -s
 	dh_shlibdeps -s -ldlls:libs:
 
+	# if the distro we're compiling for has ia32-libs, depend on it
+	(dpkg -s ia32-libs >/dev/null && \
+	 echo "ia32-libs=, ia32-libs" >> debian/libwine.substvars) || \
+	true
+
 	# if the distro we're compiling for has freetype, depend on it
 	(dpkg -s libfreetype6-dev >/dev/null && \
 	 echo "freetype=, libfreetype6" >> debian/libwine.substvars && \

Changed Bug title. Request was from Andreas Jochens <aj@andaco.de> to control@bugs.debian.org. (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #15 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Subject: patch update (please apply!)
Date: Mon, 15 Jan 2007 23:32:56 +0100
Hi,

Andreas patch needed an update to sync with latest version of the package.  I'm
attaching it.  It works fine, provided that ia32-libs is patched (see #394230
which I just reopened).

Please, can you include this patch in next upload?

etch will most likely be our last release before the 64-bit migration window
finishes in late 2008 [1].  If our 64-bit port has to go through this process
without win32 emulation I'm sure we'll regret it.  Let's make sure it gets into
etch!

[1] see http://catb.org/~esr/writings/world-domination/world-domination-201.html
  for an in-depth analisys based on Moore's law

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #18 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Cc: Andreas Jochens <aj@andaco.de>
Subject: patch update (please apply!)
Date: Mon, 15 Jan 2007 23:33:46 +0100
[Message part 1 (text/plain, inline)]
Hi,

Andreas patch needed an update to sync with latest version of the package.  I'm
attaching it.  It works fine, provided that ia32-libs is patched (see #394230
which I just reopened).

Please, can you include this patch in next upload?

etch will most likely be our last release before the 64-bit migration window
finishes in late 2008 [1].  If our 64-bit port has to go through this process
without win32 emulation I'm sure we'll regret it.  Let's make sure it gets into
etch!

[1] see http://catb.org/~esr/writings/world-domination/world-domination-201.html
  for an in-depth analisys based on Moore's law

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

[amd64.diff (text/x-diff, attachment)]

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #23 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: ***SPAM*** Re: Bug#381341: patch update (please apply!)
Date: Tue, 16 Jan 2007 09:35:33 -0500
Robert Millan wrote:
> Hi,
> 
> Andreas patch needed an update to sync with latest version of the package.  I'm
> attaching it.  It works fine, provided that ia32-libs is patched (see #394230
> which I just reopened).
> 
> Please, can you include this patch in next upload?

I don't like the patch.

When you build Wine on amd64, we know that Wine is built as a 32-bit 
binary, of course. Therefore, we want Wine to build against 32-bit 
libraries, not 64-bit libraries. And Wine has a *lot* of build 
dependencies, not all of which seem to have 32-bit equivalents. When you 
add a build dependency on ia32-libs, you get a few, but still not even 
half of what Wine wants. And last I checked (which I admit was a number 
of months ago), there were many X libraries (DGA, VM, stuff like that) 
that does exist there, but that you could not compile against. And most 
just don't exist, even critical ones.

For example, one important core Wine feature, i18n, depends on 
libicu36-dev. There seems to be no 32-bit version of this, therefore, if 
we use this patch, the amd64 version of Wine will be severely limited in 
functionality here, compared to the i386 version of Wine.

Most of the split-out packages will be useless too - for example, 
libwine-sane will have no functionality at all without a lib32sane-dev 
or something, and should therefore preferably not be part of an amd64 
build - the architecture fields should specify the affected packages as 
not available on amd64.

If possible, it would be a far preferable approach to actually build 
Wine in a 32-bit environment and then just set the architecture fields 
or something. It would give amd64 users a full, uncrippled, win32 emulation.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #26 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Cc: Andreas Jochens <aj@andaco.de>
Subject: Re: Bug#381341: patch update (please apply!)
Date: Thu, 18 Jan 2007 02:27:46 +0100
> When you build Wine on amd64, we know that Wine is built as a 32-bit 
> binary, of course. Therefore, we want Wine to build against 32-bit 
> libraries, not 64-bit libraries. And Wine has a *lot* of build 
> dependencies, not all of which seem to have 32-bit equivalents. When you 
> add a build dependency on ia32-libs, you get a few, but still not even 
> half of what Wine wants. And last I checked (which I admit was a number 
> of months ago), there were many X libraries (DGA, VM, stuff like that) 
> that does exist there, but that you could not compile against. And most 
> just don't exist, even critical ones.
> 
> For example, one important core Wine feature, i18n, depends on 
> libicu36-dev. There seems to be no 32-bit version of this, therefore, if 
> we use this patch, the amd64 version of Wine will be severely limited in 
> functionality here, compared to the i386 version of Wine.

I clearly prefer shipping a somewhat limited version of wine than shipping
none at all.  But that is not to say we shouldn't fix the problems you
describe.

As you can see, I filed an update to #394230 that will fill the gap for these
missing libraries.

> If possible, it would be a far preferable approach to actually build 
> Wine in a 32-bit environment and then just set the architecture fields 
> or something. It would give amd64 users a full, uncrippled, win32 emulation.

But then we'd need to add the missing libs to ia32-libs for runtime anyway, so
we wouldn't gain anything.

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #31 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Cc: Andreas Jochens <aj@andaco.de>
Subject: ***SPAM*** Re: Bug#381341: patch update (please apply!)
Date: Wed, 17 Jan 2007 21:39:42 -0500
Robert Millan wrote:
> As you can see, I filed an update to #394230 that will fill the gap for these
> missing libraries.

Thanks.

>> If possible, it would be a far preferable approach to actually build 
>> Wine in a 32-bit environment and then just set the architecture fields 
>> or something. It would give amd64 users a full, uncrippled, win32 emulation.
> 
> But then we'd need to add the missing libs to ia32-libs for runtime anyway, so
> we wouldn't gain anything.

Not everything, but we could gain *something*. For example, libicu36 
gets statically linked into Wine, so it wouldn't be a problem that it 
didn't otherwise exist in 32-bit form. Also it would fix cases where 
ia32-libs has a functional library, but not a functional -dev package or 
.so symlink.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #34 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org, Andreas Jochens <aj@andaco.de>
Subject: Re: ***SPAM*** Re: Bug#381341: patch update (please apply!)
Date: Fri, 19 Jan 2007 22:28:22 +0100
On Wed, Jan 17, 2007 at 09:39:42PM -0500, Ove Kaaven wrote:
> 
> Not everything, but we could gain *something*. For example, libicu36 
> gets statically linked into Wine, so it wouldn't be a problem that it 
> didn't otherwise exist in 32-bit form. Also it would fix cases where 
> ia32-libs has a functional library, but not a functional -dev package or 
> .so symlink.

Ok then, let's go with that.

  - I proposed to follow this approach in #394230, and requested addition of
  the missing runtime libraries without bothering about .so symlinks or static
  objects.  If you agree to it, could you post to #394230 giving your approval?

  - I made an _ugly_ patch to implement your proposal in the wine package.
    - debian/amd64.tar.lzma.uu is generated during the i386 build.
    - lzma is used for its high compression ratio, uuencode to keep diff happy.
    - You'll have to run an additional build before doing a full upload, to
    allow latest binaries to propagate from amd64.tar.lzma.uu to your diff.gz.

I'll contact the release team to make sure we get it in time.

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #37 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org, Andreas Jochens <aj@andaco.de>
Subject: Re: ***SPAM*** Re: Bug#381341: patch update (please apply!)
Date: Fri, 19 Jan 2007 22:32:58 +0100
[Message part 1 (text/plain, inline)]
Oh, the patch.

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

[amd64.diff (text/x-diff, attachment)]

Reply sent to Robert Millan <rmh@aybabtu.com>:
You have taken responsibility. (full text, mbox, link).

Notification sent to Robert Millan <rmh@aybabtu.com>:
Bug acknowledged by developer. (full text, mbox, link).

Message #42 received at 381341-done@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341-done@bugs.debian.org
Subject: Re: Bug#407052 closed by Ove Kaaven <ovek@arcticnet.no> (Bug#407052: fixed in wine 0.9.29-1)
Date: Thu, 25 Jan 2007 21:06:24 +0100
On Thu, Jan 25, 2007 at 07:48:49AM -0800, Debian Bug Tracking System wrote:
>    * Added amd64 build hack by Robert Millan (from #381341),
>      which encode the i386 binaries into the diff and then
>      just decode them for the amd64 build, to work around
>      difficulties with building a 32-bit package on amd64.

Which is working fine.  Closing this bug.

Btw, amd64 buildds aren't going to grab wine yet (I already sent an update
for P-a-s).  Can you do a manual binary-only upload?

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended for
spam harvesters.  Writing to it will get you added to my black list.

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #47 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>
Cc: 381341@bugs.debian.org
Subject: ***SPAM*** Re: Bug#407052 closed by Ove Kaaven <ovek@arcticnet.no> (Bug#407052: fixed in wine 0.9.29-1)
Date: Thu, 25 Jan 2007 16:59:03 -0500
tor, 25,.01.2007 kl. 21.06 +0100, skrev Robert Millan:
> On Thu, Jan 25, 2007 at 07:48:49AM -0800, Debian Bug Tracking System wrote:
> >    * Added amd64 build hack by Robert Millan (from #381341),
> >      which encode the i386 binaries into the diff and then
> >      just decode them for the amd64 build, to work around
> >      difficulties with building a 32-bit package on amd64.
> 
> Which is working fine.  Closing this bug.

I didn't really want to close the bug since it's a hack and we'll
probably want to do it differently...

> Btw, amd64 buildds aren't going to grab wine yet (I already sent an update
> for P-a-s).  Can you do a manual binary-only upload?

No, the only amd64 machine I have access to is at the office, on the
company network, behind the company firewall, and owned by the company.
And normally runs in 32-bit mode.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #50 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org
Subject: Re: ***SPAM*** Re: Bug#407052 closed by Ove Kaaven <ovek@arcticnet.no> (Bug#407052: fixed in wine 0.9.29-1)
Date: Thu, 25 Jan 2007 23:35:39 +0100
reopen 381341
retitle 381341 replace amd64 hack with something better
thanks

On Thu, Jan 25, 2007 at 04:59:03PM -0500, Ove Kaaven wrote:
> tor, 25,.01.2007 kl. 21.06 +0100, skrev Robert Millan:
> > On Thu, Jan 25, 2007 at 07:48:49AM -0800, Debian Bug Tracking System wrote:
> > >    * Added amd64 build hack by Robert Millan (from #381341),
> > >      which encode the i386 binaries into the diff and then
> > >      just decode them for the amd64 build, to work around
> > >      difficulties with building a 32-bit package on amd64.
> > 
> > Which is working fine.  Closing this bug.
> 
> I didn't really want to close the bug since it's a hack and we'll
> probably want to do it differently...

Ok, sorry about that.

> > Btw, amd64 buildds aren't going to grab wine yet (I already sent an update
> > for P-a-s).  Can you do a manual binary-only upload?
> 
> No, the only amd64 machine I have access to is at the office, on the
> company network, behind the company firewall, and owned by the company.
> And normally runs in 32-bit mode.

Well, LaMont said he's updating Packages-arch-specific in a few minutes
so hopefuly that won't be a problem.

-- 
Robert Millan

My spam trap is honeypot@aybabtu.com.  Note: this address is only intended
for spam harvesters.  Writing to it will get you added to my black list.

Bug reopened, originator not changed. Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (full text, mbox, link).

Changed Bug title. Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (full text, mbox, link).

Tags removed: patch Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #59 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Cc: 430845@bugs.debian.org, "Ben Hodgetts (Enverex)" <ben@atomnet.co.uk>
Subject: fix amd64 builds
Date: Tue, 7 Aug 2007 20:08:18 +0200
[Message part 1 (text/plain, inline)]
severity 381341 important
merge 381341 430845
tags 381341 patch
thanks

Hi,

It is now possible to properly compile a 32-bit wine on amd64.  See attached
patch.  It's tested and known to work on etch and sid.

As collateral benefit, this fixes #430845.  I'm merging it.

Notes on the patch:

  - While I was at it, I simplified some Build-Depends rules which were
  there to support old debian releases.
  - libicu hack is no longer necessary at all because of upstream fix:
    2007-05-07  Alexandre Julliard <julliard@winehq.org>

Btw, remember to get rid of the stale amd64.lzma.uu ;-)

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)

[fix_amd64_build.diff (text/x-diff, attachment)]

Severity set to `important' from `normal' Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (Tue, 07 Aug 2007 18:33:08 GMT) (full text, mbox, link).

Merged 381341 430845. Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (Tue, 07 Aug 2007 18:33:08 GMT) (full text, mbox, link).

Tags added: patch Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (Tue, 07 Aug 2007 18:33:10 GMT) (full text, mbox, link).

Disconnected #381341 from all other report(s). Request was from Ove Kaaven <ovek@debian.org> to control@bugs.debian.org. (Sun, 12 Aug 2007 12:51:02 GMT) (full text, mbox, link).

Severity set to `normal' from `important' Request was from Ove Kaaven <ovek@debian.org> to control@bugs.debian.org. (Sun, 12 Aug 2007 12:51:04 GMT) (full text, mbox, link).

Blocking bugs of 381341 added: 398778 Request was from Ove Kaaven <ovek@debian.org> to control@bugs.debian.org. (Sun, 12 Aug 2007 13:00:02 GMT) (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #74 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Cc: 430845@bugs.debian.org, "Ben Hodgetts (Enverex)" <ben@atomnet.co.uk>
Subject: Re: fix amd64 builds
Date: Thu, 20 Sep 2007 21:49:02 +0200
[Message part 1 (text/plain, inline)]
Now that lib32icu-dev is in unstable, I guess there should be no problem with
fixing this bug.

I'm attaching an updated patch.

On Tue, Aug 07, 2007 at 08:08:18PM +0200, Robert Millan wrote:
> severity 381341 important
> merge 381341 430845
> tags 381341 patch
> thanks
> 
> Hi,
> 
> It is now possible to properly compile a 32-bit wine on amd64.  See attached
> patch.  It's tested and known to work on etch and sid.
> 
> As collateral benefit, this fixes #430845.  I'm merging it.
> 
> Notes on the patch:
> 
>   - While I was at it, I simplified some Build-Depends rules which were
>   there to support old debian releases.
>   - libicu hack is no longer necessary at all because of upstream fix:
>     2007-05-07  Alexandre Julliard <julliard@winehq.org>
> 
> Btw, remember to get rid of the stale amd64.lzma.uu ;-)

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)

[fix_amd64_build.diff (text/x-diff, attachment)]

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@debian.org>:
Extra info received and forwarded to list. Copy sent to Ove Kaaven <ovek@arcticnet.no>. (full text, mbox, link).

Message #79 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@debian.org>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Thu, 20 Sep 2007 16:47:18 -0400
tor, 20.09.2007 kl. 21.49 +0200, skrev Robert Millan:
> Now that lib32icu-dev is in unstable, I guess there should be no
> problem with fixing this bug.

381341, maybe, but it's not urgent. I've several times noted that 430845
is an unrelated issue, one that just would not be fixed by building on
amd64 (according to what Ben Hodgetts said), so I'd like to address that
first. I tested 430845 earlier today, by the way, and it works for me,
so I've downgraded it, but I'd like to hear from the original submitters
too, maybe it can just be closed.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #82 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@debian.org>
Cc: 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Fri, 21 Sep 2007 09:47:11 +0200
On Thu, Sep 20, 2007 at 04:47:18PM -0400, Ove Kaaven wrote:
> tor, 20.09.2007 kl. 21.49 +0200, skrev Robert Millan:
> > Now that lib32icu-dev is in unstable, I guess there should be no
> > problem with fixing this bug.
> 
> 381341, maybe, but it's not urgent. I've several times noted that 430845
> is an unrelated issue, one that just would not be fixed by building on
> amd64 (according to what Ben Hodgetts said), so I'd like to address that
> first. I tested 430845 earlier today, by the way, and it works for me,
> so I've downgraded it, but I'd like to hear from the original submitters
> too, maybe it can just be closed.

Oh.  How strange, in my previous tests, the problem happened always when using
the uuencode hack and never when building properly.  Now it doesn't happen at
all.

Might be related to libc or something like that, as you pointed out.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #85 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@debian.org>
Cc: 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Sun, 9 Dec 2007 14:08:14 +0100
Any news on this?  Even if #430845 isn't solved, I'd like to have a package
that can be built properly, now that we can do it.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #90 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Sun, 09 Dec 2007 18:15:22 +0100
Robert Millan skrev:
> Any news on this?  Even if #430845 isn't solved, I'd like to have a package
> that can be built properly, now that we can do it.

Well, someone filed #454756.

I'm travelling, so right now I don't have an amd64 system to test on. I 
don't even have a Linux system, just building these packages in a vmware 
virtual machine right now. I plan to see if I can get around to doing 
packaging work (such as adding build dependencies, at least) in the 
uploads before catching up to current release, but as I know I can't 
test them much, we'll see.

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #93 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Sun, 9 Dec 2007 21:16:30 +0100
On Sun, Dec 09, 2007 at 06:15:22PM +0100, Ove Kaaven wrote:
> Robert Millan skrev:
> >Any news on this?  Even if #430845 isn't solved, I'd like to have a package
> >that can be built properly, now that we can do it.
> 
> Well, someone filed #454756.

These libraries add optional features.  I'd rather have a package that can be
built normaly and provides the motivation for other people to add biarch
versions of these, than stay with the current situation.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call, if you are unable to speak?
(as seen on /.)

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #98 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: fix amd64 builds
Date: Sun, 09 Dec 2007 21:48:47 +0100
Robert Millan skrev:
> On Sun, Dec 09, 2007 at 06:15:22PM +0100, Ove Kaaven wrote:
>> Robert Millan skrev:
>>> Any news on this?  Even if #430845 isn't solved, I'd like to have a package
>>> that can be built properly, now that we can do it.
>> Well, someone filed #454756.
> 
> These libraries add optional features.

I just mentioned it, since you asked.

> I'd rather have a package that can be
> built normaly and provides the motivation for other people to add biarch
> versions of these, than stay with the current situation.

Well, before I went travelling, I had been slowly working on 
restructuring the packaging a little (those debhelper changes, for 
example), aiming to make it slightly more flexible, perhaps allowing the 
package to be built in more than one way. Been planning to resume that 
process as time permits.

Blocking bugs of 381341 removed: Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Wed, 02 Jan 2008 01:00:04 GMT) (full text, mbox, link).

Blocking bugs of 381341 added: 458013 and 456914 Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Wed, 16 Jan 2008 03:48:03 GMT) (full text, mbox, link).

Forcibly Merged 381341 462542. Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Fri, 25 Jan 2008 16:30:07 GMT) (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Javier Serrano Polo <jasp00@terra.es>:
Extra info received and forwarded to list. Copy sent to Ove Kaaven <ovek@arcticnet.no>. (full text, mbox, link).

Message #109 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: 381341@bugs.debian.org
Cc: Robert Millan <rmh@aybabtu.com>
Subject: Re: wine: replace amd64 hack with something better
Date: Tue, 19 Feb 2008 22:38:49 +0100
I'd like to bring your attention to bug #464796. It allows to drop the
amd64 hack, using alternative build dependencies. Robert may be
interested in building from this. I have a simplified list of dependency
changes (0.9.53) if you need to start from somewhere.

In the meantime, I'll package an amd64 wine, unless you want to do it.
Ove, if you're uploading 0.9.55 before next week, please tell me and
I'll wait for it.

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #114 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Javier Serrano Polo <jasp00@terra.es>, 381341@bugs.debian.org
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Wed, 20 Feb 2008 06:11:13 +0100
Javier Serrano Polo skrev:
> I'd like to bring your attention to bug #464796. It allows to drop the
> amd64 hack, using alternative build dependencies. Robert may be
> interested in building from this. I have a simplified list of dependency
> changes (0.9.53) if you need to start from somewhere.

OK, but I updated the build dependencies in 0.9.54 (added unixodbc-dev, 
removed libungif4-dev).

> In the meantime, I'll package an amd64 wine, unless you want to do it.
> Ove, if you're uploading 0.9.55 before next week, please tell me and
> I'll wait for it.

Well, I was planning to upload a 0.9.55 one of these days, and I wanted 
to include amd64 patches if I had any (so far I've got that 
build-dependency on lib32z1-dev). Of course, I cannot actually build an 
amd64 package without the build daemon's help, but at least I can make 
it easier for the users to build the package on their own...

If I understand you right, you're suggesting building Wine amd64 
packages using build-dependencies from an unofficial repository, and 
uploading that to the official Debian repository, binNMU-style. I'm not 
sure that's really a solution. (Perhaps it would technically be a 
violation of policy on some level, too, since Wine is in main, not 
contrib.) Sure, Wine's current amd64 hack is neither prettier nor really 
more secure than that, but at least it is possible to build the amd64 
package without changing /etc/apt/sources.list, and thus, the amd64 
build daemon would be able to build it (if ia32-libs wasn't broken). 
With your unofficial repository, I don't see any way I could have the 
amd64 package updated every time I update the i386 package.

Like now, if you NMU-ed 0.9.54, and I later got around to uploading the 
0.9.55 with these ia32-*-dev build deps you used, then amd64 users would 
*still* be out of luck.

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #117 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Wed, 20 Feb 2008 21:26:53 +0100
[Message part 1 (text/plain, inline)]
El dc 20 de 02 del 2008 a les 06:11 +0100, en/na Ove Kaaven va escriure:
> > I have a simplified list of dependency
> > changes (0.9.53) if you need to start from somewhere.
> 
> OK, but I updated the build dependencies in 0.9.54 (added unixodbc-dev, 
> removed libungif4-dev).

"you" meant Robert in that sentence, because he's interested in *bsd
stuff and may know what dependencies don't apply. I know you (Ove) have
trouble to build on amd64. I'm attaching the simplified Build-Depends,
so you can see what's being changed.

> Well, I was planning to upload a 0.9.55 one of these days, and I wanted 
> to include amd64 patches if I had any (so far I've got that 
> build-dependency on lib32z1-dev). Of course, I cannot actually build an 
> amd64 package without the build daemon's help, but at least I can make 
> it easier for the users to build the package on their own...

Then you could try this new scheme, since ia32-libs is broken. The ia32-
naming is the same ia32-libs maintainers want to introduce. Differences
lie on how these packages are fetched.
You may drop lib32z1-dev, unless you've added zlib1g-dev. unixodbc-dev
will be ia32-unixodbc-dev.

> If I understand you right, you're suggesting building Wine amd64 
> packages using build-dependencies from an unofficial repository,

Initially.

> and
> uploading that to the official Debian repository, binNMU-style.

No, that requires building from official repositories only. I have to
prove this is useful first. I know this is working, but I need others to
test and confirm. When enough packages have been ported then comes the
request to be official.

Note that the repository is needed to build (ia32-*-dev) and to run
(ia32-*) the packages.

> With your unofficial repository, I don't see any way I could have the 
> amd64 package updated every time I update the i386 package.

It'd be built with lenny libraries automatically. If I added the sid
distribution, it'd use sid libraries.

> Like now, if you NMU-ed 0.9.54, and I later got around to uploading the 
> 0.9.55 with these ia32-*-dev build deps you used, then amd64 users would 
> *still* be out of luck.

They wouldn't if they used the repository and were willing to wait (it's
bandwidth limited).

[build-depends.txt (text/plain, attachment)]

Information stored:
Bug#381341; Package wine. (full text, mbox, link).

Message #120 received at 381341-quiet@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Javier Serrano Polo <jasp00@terra.es>, 381341-quiet@bugs.debian.org
Cc: Ove Kaaven <ovek@arcticnet.no>, 381341-submitter@bugs.debian.org
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Wed, 20 Feb 2008 21:58:31 +0100
On Wed, Feb 20, 2008 at 09:26:53PM +0100, Javier Serrano Polo wrote:
> El dc 20 de 02 del 2008 a les 06:11 +0100, en/na Ove Kaaven va escriure:
> > > I have a simplified list of dependency
> > > changes (0.9.53) if you need to start from somewhere.
> > 
> > OK, but I updated the build dependencies in 0.9.54 (added unixodbc-dev, 
> > removed libungif4-dev).
> 
> "you" meant Robert in that sentence, because he's interested in *bsd
> stuff and may know what dependencies don't apply. I know you (Ove) have
> trouble to build on amd64. I'm attaching the simplified Build-Depends,
> so you can see what's being changed.

Sorry I must have missed something.  What are you talking about?

> Build-Depends: debhelper (>= 5), flex | flex-old, bison,
>  libx11-dev [!amd64], ia32-libx11-dev [amd64],
>  libxext-dev [!amd64], ia32-libxext-dev [amd64],
>  libxi-dev [!amd64], ia32-libxi-dev [amd64],
>  libxrandr-dev [!amd64], ia32-libxrandr-dev [amd64],
>  libxrender-dev [!amd64], ia32-libxrender-dev [amd64],
>  libxt-dev [!amd64], ia32-libxt-dev [amd64],
>  libxkbfile-dev [!amd64], ia32-libxkbfile-dev [amd64],
>  libxxf86dga-dev [!amd64], ia32-libxxf86dga-dev [amd64],
>  libxxf86vm-dev [!amd64], ia32-libxxf86vm-dev [amd64],
>  libxinerama-dev [!amd64], ia32-libxinerama-dev [amd64],
>  libxcomposite-dev [!amd64], ia32-libxcomposite-dev [amd64],
>  xlibmesa-gl-dev [!amd64], ia32-libgl1-mesa-dev [amd64],
>  libglu1-mesa-dev [!amd64], ia32-libglu1-mesa-dev [amd64],

Who's going to do all the job of adding biarch builds in these libraries?

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #126 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Wed, 20 Feb 2008 22:13:05 +0100
> What are you talking about?

Sorry for the surprise. Please read from
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=381341#109

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #129 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: 381341-submitter@bugs.debian.org, Ove Kaaven <ovek@arcticnet.no>
Subject: Re: wine: replace amd64 hack with something better
Date: Thu, 21 Feb 2008 02:40:14 +0100
El dc 20 de 02 del 2008 a les 21:58 +0100, en/na Robert Millan va
escriure:
> Sorry I must have missed something.  What are you talking about?

Sorry for the surprise. Please read from message #109 in the bug.

I've addressed some issues with nspluginwrapper, another package I found
important. Now I should be able to package 0.9.54. Ove, let me know if
you'll try this scheme for 0.9.55 and I'll prepare the complete
Build-Depends, with old alternatives.

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #132 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Javier Serrano Polo <jasp00@terra.es>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Thu, 21 Feb 2008 05:19:47 +0100
Javier Serrano Polo skrev:
> Then you could try this new scheme, since ia32-libs is broken.

I don't see how I could. I need the build daemon to build amd64 packages 
for me, since I cannot do it myself, and this doesn't help with that.

>> uploading that to the official Debian repository, binNMU-style.
> 
> No, that requires building from official repositories only. I have to
> prove this is useful first. I know this is working, but I need others to
> test and confirm. When enough packages have been ported then comes the
> request to be official.

I don't get it. If not uploading to the official Debian repository, then 
to where? It sounded like you were saying that *your* unofficial webhost 
couldn't handle the load of tons of Wine-hungry Debian users with amd64, 
so I don't understand what else your plan was.

>> With your unofficial repository, I don't see any way I could have the 
>> amd64 package updated every time I update the i386 package.
> 
> It'd be built with lenny libraries automatically. If I added the sid
> distribution, it'd use sid libraries.

Huh? That has nothing to do with what I wondered about.

>> Like now, if you NMU-ed 0.9.54, and I later got around to uploading the 
>> 0.9.55 with these ia32-*-dev build deps you used, then amd64 users would 
>> *still* be out of luck.
> 
> They wouldn't if they used the repository and were willing to wait (it's
> bandwidth limited).

But all the users would still have to build the package themselves, is 
what you're saying? Or would *you* run a buildd?

As for bandwidth... every DD gets free webhosting on people.debian.org, 
why not just put your repository there?

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #135 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Thu, 21 Feb 2008 20:06:59 +0100
El dj 21 de 02 del 2008 a les 05:19 +0100, en/na Ove Kaaven va escriure:
> But all the users would still have to build the package themselves, is 
> what you're saying?

They could, but it isn't necessary.

> Or would *you* run a buildd?

I am, sort of. The amount of automation depends on the original
packaging. For instance, I have to rebuild gtk and pango
using /usr/lib32. In wine's case I have to keep track of dependencies
and remove the hack.

> As for bandwidth... every DD gets free webhosting on people.debian.org, 
> why not just put your repository there?

That'd be great indeed. I only need a DD. Most libraries can be
repackaged on i386. Recompiled ones can be done on i386 too.
The best way to get the amd64 wine is to build from source. But you can
try what I call a full conversion, like I did with zsnes. You'll be able
to get amd64 packages on an i386 box. I can test this option. Are you
interested?

Regarding policy, AFAIK there's nothing preventing the inclusion in
contrib. Furthermore, all material comes from main. Adding the scripts,
some could even justify its use in the main section. Besides, contrib
and non-free were created because of legal issues. I wouldn't consider
it a serious policy violation.
Anyway, it isn't my intention to ask for policy clarifications. This
scheme should be included in a proper manner when ready. First comes the
testing part.

Wine 0.9.54 is ready to be uploaded. I can start distributing it or wait
if you think we can do it in a better way.

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #138 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Mon, 25 Feb 2008 19:31:58 +0100
[Message part 1 (text/plain, inline)]
I hope there's no objection to a similar patch like the attached one.

[dev-depends.diff (text/x-patch, attachment)]

Information stored:
Bug#381341; Package wine. (full text, mbox, link).

Message #141 received at 381341-quiet@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Javier Serrano Polo <jasp00@terra.es>, 381341-quiet@bugs.debian.org
Cc: Ove Kaaven <ovek@arcticnet.no>, 381341-submitter@bugs.debian.org
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Mon, 25 Feb 2008 21:26:42 +0100
On Mon, Feb 25, 2008 at 07:31:58PM +0100, Javier Serrano Polo wrote:
> I hope there's no objection to a similar patch like the attached one.

> --- wine-0.9.55.orig/debian/control	2008-02-25 19:20:57.000000000 +0100
> +++ wine-0.9.55/debian/control	2008-02-21 17:35:38.000000000 +0100
> @@ -88,7 +88,7 @@
>  Package: libwine-dev
>  Architecture: i386 hurd-i386 kfreebsd-i386 netbsd-i386 amd64 powerpc hurd-powerpc kfreebsd-powerpc netbsd-powerpc sparc hurd-sparc kfreebsd-sparc netbsd-sparc
>  Section: libdevel
> -Depends: libwine (= ${binary:Version}), libc6-dev
> +Depends: libwine (= ${binary:Version}), libc6-dev [!amd64], libc6-dev-i386 [amd64]

I suggest you use | instead, it's clearer.  Also, please take kfreebsd-amd64
into account.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #147 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: 381341-submitter@bugs.debian.org
Cc: Ove Kaaven <ovek@arcticnet.no>
Subject: Re: wine: replace amd64 hack with something better
Date: Mon, 25 Feb 2008 22:53:36 +0100
El dl 25 de 02 del 2008 a les 21:26 +0100, en/na Robert Millan va
escriure:
> I suggest you use | instead, it's clearer.

I would but I'm following Ove's style.

> Also, please take kfreebsd-amd64
> into account.

Ok. I'll provide the full Build-Depends when 0.9.56 comes out. Will you
be able to use the amd64 repository? Or build a kfreebsd-amd64 one?

Blocking bugs of 381341 added: 466186 Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Tue, 26 Feb 2008 09:33:05 GMT) (full text, mbox, link).

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #152 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Javier Serrano Polo <jasp00@terra.es>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Tue, 26 Feb 2008 10:48:48 +0100
Javier Serrano Polo skrev:
> El dj 21 de 02 del 2008 a les 05:19 +0100, en/na Ove Kaaven va escriure:
>> But all the users would still have to build the package themselves, is 
>> what you're saying?
> 
> They could, but it isn't necessary.

Where would they get the built package then?

>> Or would *you* run a buildd?
> 
> I am, sort of. The amount of automation depends on the original
> packaging. For instance, I have to rebuild gtk and pango
> using /usr/lib32. In wine's case I have to keep track of dependencies
> and remove the hack.

Well, deleting the .tar.lzma.uu is enough to turn off the hack, at least.

>> As for bandwidth... every DD gets free webhosting on people.debian.org, 
>> why not just put your repository there?
> 
> That'd be great indeed. I only need a DD.

You're not a DD? And yet, you're messing with ia32-libs?

> Most libraries can be
> repackaged on i386. Recompiled ones can be done on i386 too.
> The best way to get the amd64 wine is to build from source. But you can
> try what I call a full conversion, like I did with zsnes. You'll be able
> to get amd64 packages on an i386 box. I can test this option. Are you
> interested?

From the sound of it that would output almost exactly what the amd64 
hack would do, just with a different mechanism, right?

In principle, that *could* be a good way to do it, especially if that 
solves the current situation quicker than the other alternatives... but 
in the long run, it's probably not the right thing. One day, Wine will 
be able to run 64-bit stuff, and when that happens, it should be 
compiled in a way that allows that.

> Regarding policy, AFAIK there's nothing preventing the inclusion in
> contrib.

Well, other than what the users want. Wine isn't currently in contrib, 
and from experience, some would prefer it stay in main...

> Furthermore, all material comes from main. Adding the scripts,
> some could even justify its use in the main section. Besides, contrib
> and non-free were created because of legal issues. I wouldn't consider
> it a serious policy violation.

What they were originally created for is less important than how the 
policy has changed since then... for example, the way all official 
repositories are GPG-signed these days, and how a FTBFS in main is not 
really a legal issue, but still RC and may subject the package to 
removal if left unfixed.

As for those build dependencies... does ia32-*-dev packages include the 
header files, or only the libraries? If they do contain the header 
files, wouldn't they conflict with the regular header files? And if they 
don't, why would I want the [!am64] on the regular -dev packages?

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #155 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Javier Serrano Polo <jasp00@terra.es>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Tue, 26 Feb 2008 11:04:25 +0100
Javier Serrano Polo skrev:
> El dl 25 de 02 del 2008 a les 21:26 +0100, en/na Robert Millan va
> escriure:
>> I suggest you use | instead, it's clearer.
> 
> I would but I'm following Ove's style.

I don't think I have set any precedent for that. If you're thinking of 
the build-dependencies, there aren't any [!amd64] entries in there. The 
only [!kfreebsd-amd64] is because kfreebsd isn't linux, so it doesn't 
support ALSA.

And putting all the amd64 packages at the end is mostly because the 
current ia32-libs is so crappy. Trying to change the build-depends to 
put | everywhere using ia32-libs would create a hornet's nest. Also, 
ia32-libs doesn't provide header files, so all the other -dev files are 
needed anyway.

This binary dependency seems somewhat different.

On the other hand, I don't see why libwine-dev should *not* depend on 
libc6-dev on amd64. Then, we could have it *actually* "follow my style":

Depends: libwine (...), libc6-dev, libc6-dev-i386 [amd64]

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #158 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Tue, 26 Feb 2008 23:23:07 +0100
El dt 26 de 02 del 2008 a les 10:48 +0100, en/na Ove Kaaven va escriure:
> Where would they get the built package then?

Instructions are at
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=464796#35

I'm waiting the answer from alioth to get a better environment.

> From the sound of it that would output almost exactly what the amd64 
> hack would do, just with a different mechanism, right?

I guess so, haven't compared.

> As for those build dependencies... does ia32-*-dev packages include the 
> header files, or only the libraries?

Just libraries.

> And if they 
> don't, why would I want the [!am64] on the regular -dev packages?

The regular -dev may be unnecessary, like libxmu-dev. This isn't very
significant at present.

With the full conversion you don't need to add ia32-* build dependencies
though. Anyway, I think you'd rather wait if I manage to get a better
server.

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Message #161 received at 381341-submitter@bugs.debian.org (full text, mbox, reply):

From: Javier Serrano Polo <jasp00@terra.es>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341-submitter@bugs.debian.org
Subject: Re: wine: replace amd64 hack with something better
Date: Tue, 26 Feb 2008 23:26:50 +0100
[Message part 1 (text/plain, inline)]
Here they are the control changes I'll be using for the next versions.

[control.diff.gz (application/x-gzip, attachment)]

Information stored:
Bug#381341; Package wine. (full text, mbox, link).

Message #164 received at 381341-quiet@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Javier Serrano Polo <jasp00@terra.es>, 381341-quiet@bugs.debian.org
Cc: Ove Kaaven <ovek@arcticnet.no>, 381341-submitter@bugs.debian.org
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Wed, 27 Feb 2008 12:23:13 +0100
On Tue, Feb 26, 2008 at 11:26:50PM +0100, Javier Serrano Polo wrote:
> Here they are the control changes I'll be using for the next versions.

Please get these ia32-libx11-dev et al added to the Debian archive before making
Debian packages rely on it.  There's enough work maintaining a distro to start
supporting random hacks that aren't getting anywhere.

Or do it right and send a patch for lib32x11-dev to the right place instead.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Information stored:
Bug#381341; Package wine. (full text, mbox, link).

Message #170 received at 381341-quiet@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Javier Serrano Polo <jasp00@terra.es>, 381341-quiet@bugs.debian.org
Cc: Ove Kaaven <ovek@arcticnet.no>, 381341-submitter@bugs.debian.org
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Wed, 27 Feb 2008 17:12:46 +0100
On Wed, Feb 27, 2008 at 12:23:13PM +0100, Robert Millan wrote:
> 
> Or do it right and send a patch for lib32x11-dev to the right place instead.

Btw, I can give you more details on how to do this if you're interested.  I've
done quite a few lib32foo libs before (it is not hard, just time-consuming).

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Message sent on to Robert Millan <rmh@aybabtu.com>:
Bug#381341. (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #176 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: 381341@bugs.debian.org
Subject: severity
Date: Tue, 25 Mar 2008 10:17:25 +0100
severity 381341 serious
thanks

This became de-facto RC untill #458013 is fixed.

(Note: the severity change here is purely informative.  wine isn't migrating
to testing anyway because of missing binaries in the archive.)

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Severity set to `serious' from `normal' Request was from Robert Millan <rmh@aybabtu.com> to control@bugs.debian.org. (Tue, 25 Mar 2008 09:37:36 GMT) (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #183 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: severity
Date: Tue, 25 Mar 2008 11:04:50 +0100
severity 381341 normal
stop

Robert Millan skrev:
> severity 381341 serious
> thanks
> 
> This became de-facto RC untill #458013 is fixed.

What's this nonsense? #458013 causes Wine to FTBFS regardless of the 
hack - in fact, until ia32-libs is fixed, removing the amd64 hack 
*worsens* the situation (*both* libxml2 and libsane causes build 
failures now, plus, of course, lesser issues like png and cups not being 
linked in). Take a look at #466186 if you want.

Severity set to `normal' from `serious' Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Tue, 25 Mar 2008 10:09:11 GMT) (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #188 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org
Subject: Re: Bug#381341: severity
Date: Tue, 25 Mar 2008 11:40:31 +0100
On Tue, Mar 25, 2008 at 11:04:50AM +0100, Ove Kaaven wrote:
> severity 381341 normal
> stop
> 
> Robert Millan skrev:
> >severity 381341 serious
> >thanks
> >
> >This became de-facto RC untill #458013 is fixed.
> 
> What's this nonsense? #458013 causes Wine to FTBFS regardless of the 
> hack

But you set #458013 as a blocker for this bug.  If #381341 is not the right bug
for that, we could have another, then?

Currently the problem is that there isn't any clear indication that wine isn't
in releasable state (and why), so people hunting down the problem will have to
waste their time guessing.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #193 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: severity
Date: Tue, 25 Mar 2008 14:12:52 +0100
Robert Millan skrev:
> But you set #458013 as a blocker for this bug.

Sure, but that doesn't actually make *this* bug RC. If some RC bug was 
blocked on this, *then* it might effectively be RC, but that's not the 
case. So there's a few RC bugs in ia32-libs that need fixing before I'll 
drop the hack, but if ia32-libs is fixed, I still don't really *have* to 
drop the hack to get into lenny, so this bug isn't RC. The blocker isn't 
here.

> If #381341 is not the right bug
> for that, we could have another, then?

Well, if you want. File one if you think it'll be of use. I suppose 
you'd call it something like "can't build on amd64", and block it on the 
ia32-libs stuff?

Note that there is actually something I *could* do to make wine 
available on amd64 without waiting for ia32-libs to be fixed. I could 
reverse the change I made in 0.9.49-1:

* Also moved the generation of the amd64.tar.lzma.uu further up in the
  build process, before the dh_makeshlibs/dh_installdeb/dh_shlibdeps,
  so that maintainer scripts and dependencies should be generated a bit
  more like they would if the binaries were compiled directly on amd64.

If I remember right, you yourself suggested that building stuff 
"natively" on amd64 would somehow encourage maintainers to add the 
missing 32-bit support to Debian. I added this change to approach that 
ideal, and look what happened: the wine packages have *never* built on 
amd64 in the 3-4 months since I did that.

Reversing this, and thus taking wine further away from a natively-built 
amd64 package again, would make wine available on amd64 again (if the 
dep-wait is killed too, of course), but it would probably be a setback 
for your own theory...

Still, it would be a solution to this "not updated on amd64" bug you 
might file (but far from a solution to the "drop amd64 hack" bug). What 
would you think?

Information forwarded to debian-bugs-dist@lists.debian.org, Ove Kaaven <ovek@arcticnet.no>:
Bug#381341; Package wine. (full text, mbox, link).

Message #196 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Robert Millan <rmh@aybabtu.com>
To: Ove Kaaven <ovek@arcticnet.no>
Cc: 381341@bugs.debian.org
Subject: FTBFS on amd64 (Re: Bug#381341: severity)
Date: Tue, 25 Mar 2008 16:19:09 +0100
Package: wine
Severity: serious
Version: 0.9.57-1
Tags: patch

Ok, here's your new bug (yes, with patch -- see below)

On Tue, Mar 25, 2008 at 02:12:52PM +0100, Ove Kaaven wrote:
> Robert Millan skrev:
> >But you set #458013 as a blocker for this bug.
> 
> Sure, but that doesn't actually make *this* bug RC. If some RC bug was 
> blocked on this, *then* it might effectively be RC, but that's not the 
> case. So there's a few RC bugs in ia32-libs that need fixing before I'll 
> drop the hack, but if ia32-libs is fixed, I still don't really *have* to 
> drop the hack to get into lenny, so this bug isn't RC. The blocker isn't 
> here.
> 
> >If #381341 is not the right bug
> >for that, we could have another, then?
> 
> Well, if you want. File one if you think it'll be of use. I suppose 
> you'd call it something like "can't build on amd64", and block it on the 
> ia32-libs stuff?
> 
> Note that there is actually something I *could* do to make wine 
> available on amd64 without waiting for ia32-libs to be fixed. I could 
> reverse the change I made in 0.9.49-1:
> 
> * Also moved the generation of the amd64.tar.lzma.uu further up in the
>   build process, before the dh_makeshlibs/dh_installdeb/dh_shlibdeps,
>   so that maintainer scripts and dependencies should be generated a bit
>   more like they would if the binaries were compiled directly on amd64.
> 
> If I remember right, you yourself suggested that building stuff 
> "natively" on amd64 would somehow encourage maintainers to add the 
> missing 32-bit support to Debian. I added this change to approach that 
> ideal, and look what happened: the wine packages have *never* built on 
> amd64 in the 3-4 months since I did that.
> 
> Reversing this, and thus taking wine further away from a natively-built 
> amd64 package again, would make wine available on amd64 again (if the 
> dep-wait is killed too, of course), but it would probably be a setback 
> for your own theory...
> 
> Still, it would be a solution to this "not updated on amd64" bug you 
> might file (but far from a solution to the "drop amd64 hack" bug). What 
> would you think?

Now that I think, you can easily disable that annoying check with:

--- wine-0.9.57.old/debian/rules        2008-03-25 16:17:32.000000000 +0100
+++ wine-0.9.57/debian/rules    2008-03-25 16:16:52.000000000 +0100
@@ -312,7 +312,7 @@
        dh_makeshlibs -plibwine -n -V "libwine (= $(VERSION))"

        bash debian/gendeps.sh $(patsubst build%,%,$(BUILDS))
-       dh_shlibdeps -s -Llibwine -ldebian/libwine/usr/lib
+       dh_shlibdeps -s -Llibwine -ldebian/libwine/usr/lib -- --ignore-missing-info
        bash debian/cleandeps.sh

        # relaxed libwine dependencies for everyone else

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #201 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: Robert Millan <rmh@aybabtu.com>, 381341@bugs.debian.org
Subject: Re: Bug#381341: FTBFS on amd64 (Re: Bug#381341: severity)
Date: Tue, 25 Mar 2008 20:47:13 +0100
Robert Millan skrev:
> Now that I think, you can easily disable that annoying check with:
> 
> --- wine-0.9.57.old/debian/rules        2008-03-25 16:17:32.000000000 +0100
> +++ wine-0.9.57/debian/rules    2008-03-25 16:16:52.000000000 +0100
> @@ -312,7 +312,7 @@
>         dh_makeshlibs -plibwine -n -V "libwine (= $(VERSION))"
> 
>         bash debian/gendeps.sh $(patsubst build%,%,$(BUILDS))
> -       dh_shlibdeps -s -Llibwine -ldebian/libwine/usr/lib
> +       dh_shlibdeps -s -Llibwine -ldebian/libwine/usr/lib -- --ignore-missing-info

What, that actually works? Even with the broken ia32-libs?

Blocking bugs of 381341 added: 473586 Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Mon, 31 Mar 2008 14:30:18 GMT) (full text, mbox, link).

Information forwarded to debian-bugs-dist@lists.debian.org:
Bug#381341; Package wine. (full text, mbox, link).

Acknowledgement sent to Ove Kaaven <ovek@arcticnet.no>:
Extra info received and forwarded to list. (full text, mbox, link).

Message #208 received at 381341@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: 381341@bugs.debian.org, Debian Bug Control <control@bugs.debian.org>
Subject: Re: Bug#381341: wine: replace amd64 hack with something better
Date: Mon, 31 Mar 2008 16:45:43 +0200
tags 381341 - patch
severity 381341 important
stop

I've filed new bugs against ia32-libs, and marked this bug as blocked by 
one of them. This bug also doesn't really contain any currently 
applicable patch that still needs to be applied to the package anymore, 
so I'll untag it. However, I'll raise its severity.

Tags removed: patch Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Mon, 31 Mar 2008 14:48:35 GMT) (full text, mbox, link).

Severity set to `important' from `normal' Request was from Ove Kaaven <ovek@arcticnet.no> to control@bugs.debian.org. (Mon, 31 Mar 2008 14:48:37 GMT) (full text, mbox, link).

Reply sent to Ove Kaaven <ovek@arcticnet.no>:
You have taken responsibility. (full text, mbox, link).

Notification sent to Robert Millan <rmh@aybabtu.com>:
Bug acknowledged by developer. (full text, mbox, link).

Message #217 received at 381341-close@bugs.debian.org (full text, mbox, reply):

From: Ove Kaaven <ovek@arcticnet.no>
To: 381341-close@bugs.debian.org
Subject: Bug#381341: fixed in wine 0.9.61-1
Date: Sat, 03 May 2008 09:03:20 +0000
Source: wine
Source-Version: 0.9.61-1

We believe that the bug you reported is fixed in the latest version of
wine, which is due to be installed in the Debian FTP archive:

libwine-alsa_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-alsa_0.9.61-1_i386.deb
libwine-capi_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-capi_0.9.61-1_i386.deb
libwine-cms_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-cms_0.9.61-1_i386.deb
libwine-dbg_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-dbg_0.9.61-1_i386.deb
libwine-dev_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-dev_0.9.61-1_i386.deb
libwine-esd_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-esd_0.9.61-1_i386.deb
libwine-gl_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-gl_0.9.61-1_i386.deb
libwine-gphoto2_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-gphoto2_0.9.61-1_i386.deb
libwine-jack_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-jack_0.9.61-1_i386.deb
libwine-ldap_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-ldap_0.9.61-1_i386.deb
libwine-nas_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-nas_0.9.61-1_i386.deb
libwine-oss_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-oss_0.9.61-1_i386.deb
libwine-print_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-print_0.9.61-1_i386.deb
libwine-sane_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-sane_0.9.61-1_i386.deb
libwine-twain_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine-twain_0.9.61-1_i386.deb
libwine_0.9.61-1_i386.deb
  to pool/main/w/wine/libwine_0.9.61-1_i386.deb
wine-bin_0.9.61-1_i386.deb
  to pool/main/w/wine/wine-bin_0.9.61-1_i386.deb
wine-utils_0.9.61-1_i386.deb
  to pool/main/w/wine/wine-utils_0.9.61-1_i386.deb
wine_0.9.61-1.diff.gz
  to pool/main/w/wine/wine_0.9.61-1.diff.gz
wine_0.9.61-1.dsc
  to pool/main/w/wine/wine_0.9.61-1.dsc
wine_0.9.61-1_i386.deb
  to pool/main/w/wine/wine_0.9.61-1_i386.deb
wine_0.9.61.orig.tar.gz
  to pool/main/w/wine/wine_0.9.61.orig.tar.gz



A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 381341@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Ove Kaaven <ovek@arcticnet.no> (supplier of updated wine package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.8
Date: Sat, 03 May 2008 07:46:16 +0200
Source: wine
Binary: wine wine-bin libwine-dbg libwine-dev libwine libwine-alsa libwine-capi libwine-cms libwine-esd libwine-gl libwine-gphoto2 libwine-jack libwine-ldap libwine-nas libwine-oss libwine-print libwine-sane libwine-twain wine-utils
Architecture: source i386
Version: 0.9.61-1
Distribution: unstable
Urgency: low
Maintainer: Ove Kaaven <ovek@arcticnet.no>
Changed-By: Ove Kaaven <ovek@arcticnet.no>
Description: 
 libwine    - Windows API implementation - library
 libwine-alsa - Windows API implementation - ALSA sound module
 libwine-capi - Windows API implementation - ISDN module
 libwine-cms - Windows API implementation - color management module
 libwine-dbg - Windows API implementation - debugging symbols
 libwine-dev - Windows API implementation - development files
 libwine-esd - Windows API implementation - EsounD sound module
 libwine-gl - Windows API implementation - OpenGL module
 libwine-gphoto2 - Windows API implementation - camera module
 libwine-jack - Windows API implementation - JACK sound module
 libwine-ldap - Windows API implementation - LDAP module
 libwine-nas - Windows API implementation - NAS sound module
 libwine-oss - Windows API implementation - OSS sound module
 libwine-print - Windows API implementation - printing module
 libwine-sane - Windows API implementation - scanner module
 libwine-twain - Windows API implementation - empty transitional package
 wine       - Windows API implementation - standard suite
 wine-bin   - Windows API implementation - binary loader
 wine-utils - Windows API implementation - utilities
Closes: 381341
Changes: 
 wine (0.9.61-1) unstable; urgency=low
 .
   * New upstream release 0.9.61, released May 2, 2008.
     - Automatic updating of the WINEPREFIX directory.
     - Winhelp now uses Richedit as display engine.
     - Many RichEdit fixes.
     - More improvements to IME support.
     - More quartz fixes.
     - Implementation for many more Gdiplus functions.
     - Lots of bug fixes.
   * Added Vcs-* fields to debian/control, pointing at the new Git
     repository for the Debian Wine packages.
   * Fix diff-amd64.sh to work if building out of the Git repository.
   * Allow clean rule to succeed if invoked more than once with
     DEB_SAVEBUILD set.
   * Added scripts to import new upstream releases and build packages
     using the pkg-wine Git repository.
   * Added debian/watch file.
   * With ia32-libs version 2.4, it's finally possible to build Wine
     properly on amd64, so I'll now finally disable the amd64 hack.
     (Closes: #381341.)
Checksums-Sha1: 
 035a0f53ce90ee6254c4ff02473b9407538c56ef 3084 wine_0.9.61-1.dsc
 699d8733fb5bbc934253f38c931aff0972c6e304 17952858 wine_0.9.61.orig.tar.gz
 d59de1b2de3f886540d93fb390d6cb603489b158 54137 wine_0.9.61-1.diff.gz
 145114ee8967318d799591b41555e83e95e8e4b0 938338 wine_0.9.61-1_i386.deb
 fc36f301301056c468e539918a4e128fc51a10bc 1772528 wine-bin_0.9.61-1_i386.deb
 e933ed8348485113d6f5ca413cb86f0db5945fad 25944462 libwine-dbg_0.9.61-1_i386.deb
 c09c7b2eb08a6f3e3e4f3d9c8b03ec7e089ea6db 4789808 libwine-dev_0.9.61-1_i386.deb
 2b55fd3d88fe5e80592ebe41e7f8cc0ac02f3354 9433810 libwine_0.9.61-1_i386.deb
 0a153c8b6bb46ef27577d687ca254fc9705dd2ab 994964 libwine-alsa_0.9.61-1_i386.deb
 998adb2f0fb169dcc7c3fb0570a853ff0ba89367 940474 libwine-capi_0.9.61-1_i386.deb
 257762736440adfda3dfb2c27652e072b8b8c479 954754 libwine-cms_0.9.61-1_i386.deb
 0b0ed798404e7863812e750ac9ff945cd29d201d 950480 libwine-esd_0.9.61-1_i386.deb
 3de1789ae4ef67ca70269ced021c9af98bed37ae 1274456 libwine-gl_0.9.61-1_i386.deb
 4ce59b113fedc11681c4d01f95bb1169d3d316b0 956010 libwine-gphoto2_0.9.61-1_i386.deb
 69daa9c7492a3ece0ccb1d12206086f1ffc870ee 952240 libwine-jack_0.9.61-1_i386.deb
 46e9301394c6b94a9504f8ec68a2320f80850f95 1001002 libwine-ldap_0.9.61-1_i386.deb
 35510fea6736d7a5746d7028605ec07171597954 946730 libwine-nas_0.9.61-1_i386.deb
 dc2f96240b57a5b1bf5bb0359b1121cf1d7b6df7 997560 libwine-oss_0.9.61-1_i386.deb
 2b2659cfed47ab707af72e5ae8b0b78df8abbe42 1050680 libwine-print_0.9.61-1_i386.deb
 e6721deaba3ee9d82d0ed18a813e1319eec0124c 953748 libwine-sane_0.9.61-1_i386.deb
 7155a344584122cfe91c935ad4729558b3df0672 935864 libwine-twain_0.9.61-1_i386.deb
 d8505d3fcca037ab4cc75ae22b7d5d408c613d39 1449932 wine-utils_0.9.61-1_i386.deb
Checksums-Sha256: 
 bb59a5d485908e7f8ecb89452a8d47442858ce3c5c5ea567b16dd989859255d9 3084 wine_0.9.61-1.dsc
 45c585876fe32a960183670231637e3dff9fae67f65ddf0eedbd1d87d0cc3db8 17952858 wine_0.9.61.orig.tar.gz
 e8d7b2363ab514f268cf1f92a815742e71b59e661b9d61ef37009025b206bb46 54137 wine_0.9.61-1.diff.gz
 042cb304891358f974ce46e5d750c73e5076be7b0a3bb2e2ff33fec7ac9b0d61 938338 wine_0.9.61-1_i386.deb
 16ac3cfe46c64b7c6ef2316ccdb7cc4ad00ea064568dc5bc631bde58176412fa 1772528 wine-bin_0.9.61-1_i386.deb
 ba7da2f10add42ffb9b2d78d2fd7596e7149d2d8a1881608f66ca0e0375599f6 25944462 libwine-dbg_0.9.61-1_i386.deb
 089f31004a2484998eeae18f232d082b7f88aab424f9dc778371149b76d8e03f 4789808 libwine-dev_0.9.61-1_i386.deb
 33e3ba4265356f39c58685413a4d7385c32ba28a3186804c7ff9441e4da16cb2 9433810 libwine_0.9.61-1_i386.deb
 86e439d2f4ab6fa19d81fd7958d4f718e370b6f76fe26ed2317c40e187fca691 994964 libwine-alsa_0.9.61-1_i386.deb
 deb93988ec903bdbd8787186fcf0d2a9a0f7a6ec86491e44dedc6b35826ce684 940474 libwine-capi_0.9.61-1_i386.deb
 7a335a202e196f93e2ba2d643953c46ee15ff3c85cbbc7db4a5f79987eccfca3 954754 libwine-cms_0.9.61-1_i386.deb
 79c72b5aff4af27587a38959077e1c832d2cda04947e1b9dac21b73cd4a9a28c 950480 libwine-esd_0.9.61-1_i386.deb
 87f876167cf977fca05b3b7ae29f20d1082a50462d283abcac7750d12c2d416f 1274456 libwine-gl_0.9.61-1_i386.deb
 60067a596bc733238a4be2e5e3e5f9b478918aec1c468a60d475bb3b3b9c4756 956010 libwine-gphoto2_0.9.61-1_i386.deb
 338f7952c5067d0af9e533ea4d30b8ad42c3474bb57cf2e15391f4e3101a4e7b 952240 libwine-jack_0.9.61-1_i386.deb
 298940bf00558e25eea9097ac58f76bbf9082384da3204952caecada54471f4b 1001002 libwine-ldap_0.9.61-1_i386.deb
 8d94b13f51797957477a1bb5f04b1d336a62b8d8138ba2312df7ffebcfbe360b 946730 libwine-nas_0.9.61-1_i386.deb
 ba9dab0a304ad48b154c9735bf3e58a1e4098d3cb2eca49ad157a9d1ab0501d8 997560 libwine-oss_0.9.61-1_i386.deb
 11e4e69a12bbbcb1f2750eb66eea2b065288facc4bc23dff3e4053eb196786fe 1050680 libwine-print_0.9.61-1_i386.deb
 99d6e3c9d41a1344a0521e6e40774582f359953e50a730024ae6056d171907a8 953748 libwine-sane_0.9.61-1_i386.deb
 5303848290c8d4876a1ed4299c00f6abcf23f49be868d7c2c63584ac97b56387 935864 libwine-twain_0.9.61-1_i386.deb
 1d26db7c3b71bac1b797079f3706f599b737a0106281dd30797e27ac3b541ebd 1449932 wine-utils_0.9.61-1_i386.deb
Files: 
 8d17b571fd82bb85011d41339955e902 3084 otherosfs optional wine_0.9.61-1.dsc
 a1942d334bd6ee12598ad1d78403e9ab 17952858 otherosfs optional wine_0.9.61.orig.tar.gz
 ee8a20e5d424bbc88ceedeb90762d6e0 54137 otherosfs optional wine_0.9.61-1.diff.gz
 f6f78233bf2729fc8b81fd577393aba5 938338 otherosfs optional wine_0.9.61-1_i386.deb
 ed833228acd73bdbdb84f0f174337721 1772528 otherosfs optional wine-bin_0.9.61-1_i386.deb
 c1584c4ca354effff7ae6dfedc79b123 25944462 libdevel extra libwine-dbg_0.9.61-1_i386.deb
 dfb5f795e6a0cff9790ea54e6a55ec59 4789808 libdevel optional libwine-dev_0.9.61-1_i386.deb
 07fe3566084520b781d0408d7b56bdac 9433810 libs optional libwine_0.9.61-1_i386.deb
 51f92e485abe619f92da666e8872f8e9 994964 libs optional libwine-alsa_0.9.61-1_i386.deb
 b629075e98fece5dc0417106a4c408d1 940474 libs extra libwine-capi_0.9.61-1_i386.deb
 f6147582834f82799b24a1dd31076e78 954754 libs optional libwine-cms_0.9.61-1_i386.deb
 e48e73715eac18065471bdbc0341332e 950480 libs optional libwine-esd_0.9.61-1_i386.deb
 ea33b25af6f057d2c141638aaee0530b 1274456 libs optional libwine-gl_0.9.61-1_i386.deb
 1a19bfc64bb86bf047252cae9d047918 956010 libs optional libwine-gphoto2_0.9.61-1_i386.deb
 6e7956926ecd3ac0aed1ebafdf4659b3 952240 libs optional libwine-jack_0.9.61-1_i386.deb
 57969e8fc7d0f3e4b1bce26427663f7e 1001002 libs optional libwine-ldap_0.9.61-1_i386.deb
 baec35ddf2357f068607e3d68b9a22ed 946730 libs optional libwine-nas_0.9.61-1_i386.deb
 c32fa872be4263ec799828afa2e8f480 997560 libs optional libwine-oss_0.9.61-1_i386.deb
 12b8909e9ca06ec05afad0c5382f5b57 1050680 libs optional libwine-print_0.9.61-1_i386.deb
 95b76504bb6a7ae31c169ed30026f417 953748 libs optional libwine-sane_0.9.61-1_i386.deb
 7d72034c2ee943497ff6849a06b72854 935864 libs optional libwine-twain_0.9.61-1_i386.deb
 2b239fe1232ffafcf4a6207dd1f31cca 1449932 otherosfs optional wine-utils_0.9.61-1_i386.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIHBxiA+GMa4PlEQ8RArb9AKCW2znpI8dqA/3UoU525fIa8+sddQCcDt3f
dhuABKeMPx1u7VvvdTJNVlQ=
=w5Hc
-----END PGP SIGNATURE-----

Reply sent to Ove Kaaven <ovek@arcticnet.no>:
You have taken responsibility. (full text, mbox, link).

Notification sent to fabo@debian.org:
Bug acknowledged by developer. (full text, mbox, link).

Bug archived. Request was from Debbugs Internal Request <owner@bugs.debian.org> to internal_control@bugs.debian.org. (Sun, 01 Jun 2008 07:31:43 GMT) (full text, mbox, link).

Send a report that this bug log contains spam.

Debian bug tracking system administrator <owner@bugs.debian.org>. Last modified: Thu Jan 11 10:12:59 2018; Machine Name: buxtehude

Debian Bug tracking system

Debbugs is free software and licensed under the terms of the GNU Public License version 2. The current version can be obtained from https://bugs.debian.org/debbugs-source/.

Copyright © 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson, 2005-2017 Don Armstrong, and many other contributors.