been sufficiently tested yet.
Regards
Andreas Jochens
diff -urN ../tmp-orig/glibc-2.3.2.ds1/debian/control ./debian/control
--- ../tmp-orig/glibc-2.3.2.ds1/debian/control 2004-07-05 12:51:10.632483267
+0200
+++ ./debian/control2004-07-05 12:50:43.849657694 +0200
@@ -1,7 +1,7 @@
Source: g
linked to it segfaults). I will try to investigate this.
Maybe you have an idea where these problems could come from?
Regards
Andreas Jochens
few
packages inside that chroot. Everything seems to work fine.
Regards
Andreas Jochens
ies which have such functions (e.g. 'void f(struct s x)').
Please point me to the relevant documents which describe in which way
the C++ ABI has changed in gcc-3.4.
Regards
Andreas Jochens
-fomit-frame-pointer -D__USE_STRING_INLINES
to 'debian/sysdeps/amd64.mk' which uses '-g0' instead of the default '-g1'.
This is also necessary to compile glibc with gcc-3.4.
The attached patch also contains all other amd64 changes (see #246547).
Thank you for your hel
has been used since December 2004
by the ppc64 archive on alioth.
The patch uses basically the same strategy as the amd64 port,
but with an additional extra pass for the creation of an
alternative 32 bit library.
Regards
Andreas Jochens
Summary of changes:
* debian/control.in/ppc64
- New
ched patch changes debian/rules to install the ppc kernel headers
on ppc64 in '/usr/include/ppc-linux'. The implementation is similar
as for the amd64 case where the 32-bit i386 kernel headers are installed
in '/usr/include/i386-linux'.
Regards
Andreas Jochens
diff -urN ..
Hello,
thanks for your quick reply to my report!
On 05-Aug-08 16:54, Christoph Hellwig wrote:
> On Mon, Aug 08, 2005 at 04:12:28PM +0200, Andreas Jochens wrote:
> > Package: linux-kernel-headers
> > Version: 2.6.13+0rc3-1.1
> > Severity: wishlist
> > Tags: patch
>
;,' instead of '|' to make
sure that the second item gets processed by 'apt-get build-dep' and that
the gcc-3.4 version requirement (>= 3.4.4-6) cannot be fulfilled because
unstable still has 3.4.4-5.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
ppc and sparc setup.
Note that on ppc64 both '__powerpc__' and '__powerpc64__' get defined by
gcc, so one has to use '#ifndef __powerpc64__' to check for a '-m32'
32-bit compilation.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux-kernel-headers-2.6.13+
Hello,
the attached patch also adds support for /usr/include/asm wrappers
for the i386 and amd64 architectures to linux-kernel-headers.
This will allow the build of 32/64 bit alternative libc libraries
on the amd64/i386 architectures.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux
I just learned that this should better be solved by using the existing
wrapping mechanism for /usr/include/asm for i386/amd64.
Please look at #321969 for a patch which does this.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe"
An additional setting of 'arch_biarch="amd64"' has to be added to the
previous patch for the i386 case. Otherwise the autoconf.h file will
not be installed correctly.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux-kernel-headers-2.6.13+0rc3/debian/generate-asm.sh
./de
; because it will not really be usable without
it.
May be this kind of self dependency can be avoided somehow, but I could
not find a different solution for this yet.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
64-bit binaries on i386 will have to
Build-Depend on 'amd64-libs-dev' or 'libc6-dev-amd64' anyway, because
it needs to link to a 64-bit libc. In this case there is no problem.
In the case of the 'glibc' package itself, I think a Build-Depends
on 'amd64-libs-dev [i
build 32/64-bit biarch packages instead of trying to circumvent
all these issues.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
On 05-Aug-18 14:47, GOTO Masanori wrote:
> At Wed, 17 Aug 2005 22:05:42 +0200,
> Andreas Jochens wrote:
> > In the end it will be much easier to require a 64-bit machine to be
> > used to build 32/64-bit biarch packages instead of trying to circumvent
> > all these issu
On 05-Aug-18 19:38, GOTO Masanori wrote:
> At Thu, 18 Aug 2005 10:24:14 +0200,
> Andreas Jochens wrote:
> > A small number of packages still need some minor patches
> > to make this work, notably 'glibc' (see #301438 - the patch in that
> > bug is outdated, but I
Hello,
On 05-Aug-18 19:38, GOTO Masanori wrote:
> If your patch does not have any impact, please send your latest
> version to me.
I updated the ppc64 patch again so that it now applies cleanly to
the current glibc version 2.3.5-4 and also to the current svn.
Regards
Andreas Jochens
S
gcc-4.0. The same name lib32c6(-dev)
could later also be used for 32-bit libc6 packages on amd64.
However, the name 'libc6(-dev)-powerpc' is easier to implement
because it follows the current naming scheme for alternative
libc6 packages and can therefore use the already existing
infrastructure
rrect linker path '/lib64/ld-linux-x86-64.so.2' for '/lib/libc.so.6',
'/usr/bin/iconv' and the other libc binaries and libraries on amd64.
The patch could be added to the existing debian/patches/amd64-lib.dpatch.
Regards
Andreas Jochens
diff -urN ../glibc-2
Hello,
attached is the latest update of the native ppc64 support patch.
This patch applies cleanly against the current glibc version 2.3.5-7.
Regards
Andreas Jochens
Summary of changes to enable native ppc64 support:
* debian/control.in/main: Add Build-Depends on gcc-3.4 for ppc64
be automatically in the standard
library search paths while /emul/ia32-linux has to be specified
explicitly.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
s are from gcc-4.0 which also has a
simple switch that chooses between the ugly /emul/ia32-linux/usr/lib
and /usr/lib32.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
compiles and works fine. So I am not sure
if that definition is still necessary.
Thanks again for working on this. I really hope that we will get
the new libc6(-dev)-i386 packages for etch.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
single exception of the linker path.
I think it would be good to keep it that way and to let the symlinks
point to '/lib/lib*.so.*' as they do now.
Do you perhaps know of a simple way to achive this within your approach?
Anyway, thank you for your patch!
Regards
Andreas Jochens
--
HS should prohibit the saner approach
used by Debian, Ubuntu and Gentoo.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
ent.
However, the packaging process for the libc6-(dev)-i386 packages would
have to be changed a little to take this change into account.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
bit libc6 packages build correctly without
that line.
With these two small changes, glibc 2.3.6 with your patch builds
without problems and the resulting 32-bit i386 libraries work fine.
Regards
Andreas Jochens
diff -urN ../tmp-orig/glibc-2.3.6/debian/control ./debian/control
--- ../tmp-orig/
7; and not
'/lib64' to access the libraries.
> +endef
> +
With this change, glibc 2.3.6 with your patch builds and works fine for me
on amd64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
nabled again on powerpc/ppc64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
some time ago as a hack
to avoid a problem with gcc-3.4 which has been fixed in gcc-4.0.
The libc_extra_config_options line can be dropped because the
'--with-tls' and '--with-__thread' options are the defaults
since glibc version 2.3.5.
Thanks again for your work on the glibc package!
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Hello,
On 06-Feb-22 20:37, Aurelien Jarno wrote:
> Andreas Jochens a écrit :
> >This occurs because the patch changes the default library search path
> >of the dynamic linker from "/lib /usr/lib" to "/lib64 /usr/lib64".
>
> Well it's already strang
reserved for future
multiarch installations.
These changes could be implemented by simple patches without breaking
existing installations.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
On 06-Feb-24 12:06, Aurelien Jarno wrote:
> Andreas Jochens a écrit :
> >I suggest the following setup for 32-bit libraries on amd64:
> >
> >1. The ia32-libs package continues to install the 32-bit libraries in
> >/emul/ia32-linux/usr/lib but it stops to pr
Hello,
On 06-Feb-24 14:04, Aurelien Jarno wrote:
> Andreas Jochens a écrit :
> When I say linking, I don't speak about the dynamic linking, but the
> linking that occurs when building a package with gcc.
>
> >because /emul/ia32-linux/usr/lib is added to /etc/ld.so.conf b
s fine on ppc64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
case $$slibdir in \
> - /lib32 | /lib64) \
> + /lib32 | /lib64 | /emul/ia32-linux) \
I did not yet try to build glibc with this patch, but I think
the last line should be
> + /lib32 | /lib64 | /emul/ia32-linux/lib) \
because later in the patch you have
> +i38
on the ia32-libs binary package collection from i386.
The amd64 toolchain should not depend on packages from the i386 port.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Hello,
On 06-Feb-27 11:47, Aurelien Jarno wrote:
> On Mon, Feb 27, 2006 at 09:00:36AM +0100, Andreas Jochens wrote:
> The solution is probably to move both the symlink and the code of the
> preinst/postinst ia32-libs scripts into the glibc, but I don't really
> like that, e
ll packages to install native
amd64 libraries in /usr/lib is simple and sane. This should not be
changed.
Anything which makes it easier to violate this simple policy
will lead to a mixed usage of /usr/lib and /usr/lib64 and consequently
to problems which could be difficult to disentangle later.
Thi
On 06-May-19 11:02, Goswin von Brederlow wrote:
> Andreas Jochens <[EMAIL PROTECTED]> writes:
> > Anything which makes it easier to violate this simple policy
> > will lead to a mixed usage of /usr/lib and /usr/lib64 and consequently
> > to problems which could be dif
es this by changing debian/make-autoconfs.sh
and debian/rules to use 'powerpc' as the karch/kernel_arch for
ppc64.
Additionally, the patch changes debian/patches/linux-types-long-long.patch
to make sure that the '__extension__' keyword is used on ppc64 like
on the other archi
.
Regards
Andreas Jochens
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.4-1-k7
Locale: LANG=C, LC_CTYPE=C
diff -urN ../tmp-orig/glibc-2.3.2.ds1/debian/control ./debian/control
--- ../tmp-
. Most packages have been compiled for (pure64-bit) amd64
(I compiled about 8000 packages myself) and there is even a working
installer for amd64.
Regards
Andreas Jochens
. Most packages have been compiled for (pure64-bit) amd64
(I compiled about 8000 packages myself) and there is even a working
installer for amd64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
On 04-Apr-30 11:17, GOTO Masanori wrote:
> At Thu, 29 Apr 2004 17:53:06 +0200,
> Andreas Jochens wrote:
> > On 04-Apr-29 08:05, Jeff Bailey wrote:
> > > I'm a bit concerned about amd64 - Have the discussions now been settled
> > > for the whole true-64-bit-ar
On 04-Apr-30 11:17, GOTO Masanori wrote:
> At Thu, 29 Apr 2004 17:53:06 +0200,
> Andreas Jochens wrote:
> > On 04-Apr-29 08:05, Jeff Bailey wrote:
> > > I'm a bit concerned about amd64 - Have the discussions now been settled
> > > for the whole true-64-bit-ar
n /usr/lib/i386-linux/asm on amd64.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux-kernel-headers-2.5.999-test7-bk/debian/rules
./debian/rules
--- ../tmp-orig/linux-kernel-headers-2.5.999-test7-bk/debian/rules
2004-06-04 14:23:44.583272777 +0200
+++ ./debian/rules 2004-06-04 14:23:33
is does not go into sarge, because the amd64 port will
very likely not be released with sarge anyway.
Regards
Andreas Jochens
.
Regards
Andreas Jochens
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.4-1-k7
Locale: LANG=C, LC_CTYPE=C
diff -urN ../tmp-orig/glibc-2.3.2.ds1/debian/control ./debian/control
--- ../tmp-
. Most packages have been compiled for (pure64-bit) amd64
(I compiled about 8000 packages myself) and there is even a working
installer for amd64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
. Most packages have been compiled for (pure64-bit) amd64
(I compiled about 8000 packages myself) and there is even a working
installer for amd64.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
On 04-Apr-30 11:17, GOTO Masanori wrote:
> At Thu, 29 Apr 2004 17:53:06 +0200,
> Andreas Jochens wrote:
> > On 04-Apr-29 08:05, Jeff Bailey wrote:
> > > I'm a bit concerned about amd64 - Have the discussions now been settled
> > > for the whole true-64-bit-ar
On 04-Apr-30 11:17, GOTO Masanori wrote:
> At Thu, 29 Apr 2004 17:53:06 +0200,
> Andreas Jochens wrote:
> > On 04-Apr-29 08:05, Jeff Bailey wrote:
> > > I'm a bit concerned about amd64 - Have the discussions now been settled
> > > for the whole true-64-bit-ar
n /usr/lib/i386-linux/asm on amd64.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux-kernel-headers-2.5.999-test7-bk/debian/rules ./debian/rules
--- ../tmp-orig/linux-kernel-headers-2.5.999-test7-bk/debian/rules 2004-06-04
14:23:44.583272777 +0200
+++ ./debian/rules 2004-06-04 14:23:33
is does not go into sarge, because the amd64 port will
very likely not be released with sarge anyway.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
been sufficiently tested yet.
Regards
Andreas Jochens
diff -urN ../tmp-orig/glibc-2.3.2.ds1/debian/control ./debian/control
--- ../tmp-orig/glibc-2.3.2.ds1/debian/control 2004-07-05 12:51:10.632483267 +0200
+++ ./debian/control2004-07-05 12:50:43.849657694 +0200
@@ -1,7 +1,7 @@
Source: g
linked to it segfaults). I will try to investigate this.
Maybe you have an idea where these problems could come from?
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
few
packages inside that chroot. Everything seems to work fine.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
ies which have such functions (e.g. 'void f(struct s x)').
Please point me to the relevant documents which describe in which way
the C++ ABI has changed in gcc-3.4.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
-fomit-frame-pointer -D__USE_STRING_INLINES
to 'debian/sysdeps/amd64.mk' which uses '-g0' instead of the default '-g1'.
This is also necessary to compile glibc with gcc-3.4.
The attached patch also contains all other amd64 changes (see #246547).
Thank you for your he
On 04-Jul-25 23:57, GOTO Masanori wrote:
> Andreas Jochens wrote:
> > Additionally, the attached patch adds
> >
> > nptl_extra_cflags = -g0 -O3 -fomit-frame-pointer -D__USE_STRING_INLINES
> >
> > to 'debian/sysdeps/amd64.mk' which uses '-g0
;fno-unit-at-a-time.dpatch' to this email.
Thank you again for your help.
Regards
Andreas Jochens
#! /bin/sh -e
# All lines beginning with `# DP:' are a description of the patch.
# DP: Add -fno-unit-at-a-time test to configure.in and configure.
# DP: Add fno-unit-at-a-time variable to config.
mental archive on alioth since
a few weeks (without problems so far).
Regards
Andreas Jochens
--- amd64.mk2004-09-12 15:18:37.216153000 +0200
+++ debian/sysdeps/amd64.mk 2004-09-12 15:19:09.390261953 +0200
@@ -1,11 +1,14 @@
CC = gcc
BUILD_CC = gcc
-GLIBC_PASSES += nptl
-n
the whole archive without having linuxthreads in the
glibc binary packages and only five packages needed minor changes to
build properly with that setup.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
ort does not support the 2.4 kernel at all.
There is no kernel-image-2.4 package in the amd64 archives
and I think there never will be one. The 2.4 kernel does not support
the amd64 architecture very well and upstream does not do any
kernel-2.4 development for amd64 or backports to 2.4 for amd64
ut simply depends on libc6-i386 instead.
The attached patch - together with a corresponding patch to ia32-libs -
has been used for a few weeks by the experimental amd64/gcc-3.4 archive
without problems so far.
Regards
Andreas Jochens
This patch changes the following files of the glibc package:
*
able on i386.
I just checked again and it seems that 'gcc-3.4 -m64' works now.
I will try to change the patch so that it also covers the
i386/libc-amd64 case.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
On 04-Oct-04 19:48, Andreas Jochens wrote:
> On 04-Oct-04 12:27, Daniel Jacobowitz wrote:
> > When we do this for amd64, we ought to do the same thing for ia32 as
> > well. Then amd64-libs can lose the libraries.
>
> Thank you for your reply.
>
> Of course a 'li
linux' directory. This corresponds to the
amd64 case where the i386 linux kernel headers are installed in the
'/usr/include/i386-linux' directory.
Regards
Andreas Jochens
diff -urN ../tmp-orig/linux-kernel-headers-2.5.999-test7-bk/debian/rules ./debian/rules
--- ../tmp-orig/linux-ke
tags 275997 +patch
thanks
The attached patch from upstream CVS should fix this bug.
Regards
Andreas Jochens
#! /bin/sh -e
# All lines beginning with `# DP:' are a description of the patch.
# DP: C++ problem fix for nptl/sysdeps/pthread/pthread.h
# DP: DPatch author: Andreas Jochens &l
two patches are appended to the 'debian/patches/amd64-lib.dpatch' file
which already contains the change of the libdir directory from 'lib64' to
'lib' on amd64.
Regards
Andreas Jochens
diff -urN ../tmp-orig/glibc-2.3.2.ds1/debian/patches/amd64-lib.dpatch
./debian/p
\$verify_out"
> > +-s_^\(RTLDLIST=\)\(.*lib\)\(\|64\)\(/[^/]*\)\(-x86-64\)\(\.so\.[0-9.]*\)[
> > ]*$_\1"\2\4\6 \264\4\5\6"_
> > ++s_^\(RTLDLIST=\)\(.*lib\)\(\|64\)\(/[^/]*\)\(-x86-64\)\(\.so\.[0-9.]*\)[
> > ]*$_\1"\2\4\6 \2\4\5\6"_
>
> I
On 04-Oct-24 23:24, Kurt Roeckx wrote:
> On Sun, Oct 24, 2004 at 10:18:15PM +0200, Andreas Jochens wrote:
> >
> > This patch is harmless with respect to any LSB requirement.
> > The name of the dynamic loader, which is coded into every binary
> > can only be changed in
On 04-Oct-30 15:36, Bill Allombert wrote:
> On Mon, Oct 25, 2004 at 08:18:40AM +0200, Andreas Jochens wrote:
> > I patched the gcc-3.4 package in the amd64/gcc-3.4 archive to get that
> > result. For the patch I used please look at BTS #277852. I recompiled
> > the complete
On 04-Oct-30 16:55, Bill Allombert wrote:
> On Sat, Oct 30, 2004 at 04:12:01PM +0200, Andreas Jochens wrote:
> > Anyway, if you intend to run binaries on different distributions,
> > you should create binaries which conform to the LSB standard and you
> > should install
d and the binaries could not
find the dynamic linker anymore.
The '/usr/lib64' symlink is not a problem. It would be nice to have that
also in 'glibc' instead of 'base-files'. On the other hand '/usr/lib64'
is not really necessary. The 'amd64/gcc-3.4' does not have a
'/usr/lib64' at all at the moment and everything works without that
directory.
Regards
Andreas Jochens
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
if there are objections to the creation of the '/usr/lib64' and
'/usr/X11R6/lib64' directories in 'base-files', the attached minimal
patch to 'base-files' would be sufficient for the amd64 port.
Regards
Andreas Jochens
--- debian/rules.orig
79 matches
Mail list logo